Implements unified model routing

Migrates the AMP module to a new unified routing system, replacing the fallback handler with a router-based approach.

This change introduces a `ModelRoutingWrapper` that handles model extraction, routing decisions, and proxying based on provider availability and model mappings.
It provides a more flexible and maintainable routing mechanism by centralizing routing logic.

The changes include:
- Introducing new `routing` package with core routing logic.
- Creating characterization tests to capture existing behavior.
- Implementing model extraction and rewriting.
- Updating AMP module routes to utilize the new routing wrapper.
- Deprecating `FallbackHandler` in favor of the new `ModelRoutingWrapper`.

.dockerignore

@@ -13,8 +13,6 @@ Dockerfile
 docs/*
 README.md
 README_CN.md
-MANAGEMENT_API.md
-MANAGEMENT_API_CN.md
 LICENSE
 
 # Runtime data folders (should be mounted as volumes)
@@ -25,6 +23,14 @@ config.yaml
 
 # Development/editor
 bin/*
-.claude/*
 .vscode/*
+.claude/*
+.codex/*
+.gemini/*
 .serena/*
+.agent/*
+.agents/*
+.opencode/*
+.bmad/*
+_bmad/*
+_bmad-output/*

.github/ISSUE_TEMPLATE/bug_report.md

@@ -7,6 +7,13 @@ assignees: ''
 
 ---
 
+**Is it a request payload issue?**
+[  ] Yes, this is a request payload issue. I am using a client/cURL to send a request payload, but I received an unexpected error.
+[  ] No, it's another issue.
+
+**If it's a request payload issue, you MUST know**
+Our team doesn't have any GODs or ORACLEs or MIND READERs. Please make sure to attach the request log or curl payload.
+
 **Describe the bug**
 A clear and concise description of what the bug is.
 

.github/workflows/docker-image.yml

@@ -10,13 +10,11 @@ env:
   DOCKERHUB_REPO: eceasy/cli-proxy-api
 
 jobs:
-  docker:
+  docker_amd64:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout
         uses: actions/checkout@v4
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
       - name: Login to DockerHub
@@ -29,18 +27,113 @@ jobs:
           echo VERSION=`git describe --tags --always --dirty` >> $GITHUB_ENV
           echo COMMIT=`git rev-parse --short HEAD` >> $GITHUB_ENV
           echo BUILD_DATE=`date -u +%Y-%m-%dT%H:%M:%SZ` >> $GITHUB_ENV
-      - name: Build and push
+      - name: Build and push (amd64)
         uses: docker/build-push-action@v6
         with:
           context: .
-          platforms: |
-            linux/amd64
-            linux/arm64
+          platforms: linux/amd64
           push: true
           build-args: |
             VERSION=${{ env.VERSION }}
             COMMIT=${{ env.COMMIT }}
             BUILD_DATE=${{ env.BUILD_DATE }}
           tags: |
-            ${{ env.DOCKERHUB_REPO }}:latest
-            ${{ env.DOCKERHUB_REPO }}:${{ env.VERSION }}
+            ${{ env.DOCKERHUB_REPO }}:latest-amd64
+            ${{ env.DOCKERHUB_REPO }}:${{ env.VERSION }}-amd64
+
+  docker_arm64:
+    runs-on: ubuntu-24.04-arm
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+      - name: Login to DockerHub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+      - name: Generate Build Metadata
+        run: |
+          echo VERSION=`git describe --tags --always --dirty` >> $GITHUB_ENV
+          echo COMMIT=`git rev-parse --short HEAD` >> $GITHUB_ENV
+          echo BUILD_DATE=`date -u +%Y-%m-%dT%H:%M:%SZ` >> $GITHUB_ENV
+      - name: Build and push (arm64)
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          platforms: linux/arm64
+          push: true
+          build-args: |
+            VERSION=${{ env.VERSION }}
+            COMMIT=${{ env.COMMIT }}
+            BUILD_DATE=${{ env.BUILD_DATE }}
+          tags: |
+            ${{ env.DOCKERHUB_REPO }}:latest-arm64
+            ${{ env.DOCKERHUB_REPO }}:${{ env.VERSION }}-arm64
+
+  docker_manifest:
+    runs-on: ubuntu-latest
+    needs:
+      - docker_amd64
+      - docker_arm64
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+      - name: Login to DockerHub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+      - name: Generate Build Metadata
+        run: |
+          echo VERSION=`git describe --tags --always --dirty` >> $GITHUB_ENV
+          echo COMMIT=`git rev-parse --short HEAD` >> $GITHUB_ENV
+          echo BUILD_DATE=`date -u +%Y-%m-%dT%H:%M:%SZ` >> $GITHUB_ENV
+      - name: Create and push multi-arch manifests
+        run: |
+          docker buildx imagetools create \
+            --tag "${DOCKERHUB_REPO}:latest" \
+            "${DOCKERHUB_REPO}:latest-amd64" \
+            "${DOCKERHUB_REPO}:latest-arm64"
+          docker buildx imagetools create \
+            --tag "${DOCKERHUB_REPO}:${VERSION}" \
+            "${DOCKERHUB_REPO}:${VERSION}-amd64" \
+            "${DOCKERHUB_REPO}:${VERSION}-arm64"
+      - name: Cleanup temporary tags
+        continue-on-error: true
+        env:
+          DOCKERHUB_USERNAME: ${{ secrets.DOCKERHUB_USERNAME }}
+          DOCKERHUB_TOKEN: ${{ secrets.DOCKERHUB_TOKEN }}
+        run: |
+          set -euo pipefail
+          namespace="${DOCKERHUB_REPO%%/*}"
+          repo_name="${DOCKERHUB_REPO#*/}"
+
+          token="$(
+            curl -fsSL \
+              -H 'Content-Type: application/json' \
+              -d "{\"username\":\"${DOCKERHUB_USERNAME}\",\"password\":\"${DOCKERHUB_TOKEN}\"}" \
+              'https://hub.docker.com/v2/users/login/' \
+              | python3 -c 'import json,sys; print(json.load(sys.stdin)["token"])'
+          )"
+
+          delete_tag() {
+            local tag="$1"
+            local url="https://hub.docker.com/v2/repositories/${namespace}/${repo_name}/tags/${tag}/"
+            local http_code
+            http_code="$(curl -sS -o /dev/null -w "%{http_code}" -X DELETE -H "Authorization: JWT ${token}" "${url}" || true)"
+            if [ "${http_code}" = "204" ] || [ "${http_code}" = "404" ]; then
+              echo "Docker Hub tag removed (or missing): ${DOCKERHUB_REPO}:${tag} (HTTP ${http_code})"
+              return 0
+            fi
+            echo "Docker Hub tag delete failed: ${DOCKERHUB_REPO}:${tag} (HTTP ${http_code})"
+            return 0
+          }
+
+          delete_tag "latest-amd64"
+          delete_tag "latest-arm64"
+          delete_tag "${VERSION}-amd64"
+          delete_tag "${VERSION}-arm64"

.github/workflows/pr-test-build.yml

@@ -0,0 +1,23 @@
+name: pr-test-build
+
+on:
+  pull_request:
+
+permissions:
+  contents: read
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Set up Go
+        uses: actions/setup-go@v5
+        with:
+          go-version-file: go.mod
+          cache: true
+      - name: Build
+        run: |
+          go build -o test-output ./cmd/server
+          rm -f test-output

.gitignore

@@ -11,11 +11,15 @@ bin/*
 logs/*
 conv/*
 temp/*
+refs/*
+
+# Storage backends
 pgstore/*
 gitstore/*
 objectstore/*
+
+# Static assets
 static/*
-refs/*
 
 # Authentication data
 auths/*
@@ -29,8 +33,17 @@ GEMINI.md
 
 # Tooling metadata
 .vscode/*
+.codex/*
 .claude/*
+.gemini/*
 .serena/*
+.agent/*
+.agents/*
+.agents/*
+.opencode/*
+.bmad/*
+_bmad/*
+_bmad-output/*
 
 # macOS
 .DS_Store

README.md

@@ -10,14 +10,29 @@ So you can use local or multi-account CLI access with OpenAI(include Responses)/
 
 ## Sponsor
 
-[![z.ai](https://assets.router-for.me/english.png)](https://z.ai/subscribe?ic=8JVLJQFSKB)
+[![z.ai](https://assets.router-for.me/english-4.7.png)](https://z.ai/subscribe?ic=8JVLJQFSKB)
 
 This project is sponsored by Z.ai, supporting us with their GLM CODING PLAN.
 
-GLM CODING PLAN is a subscription service designed for AI coding, starting at just $3/month. It provides access to their flagship GLM-4.6 model across 10+ popular AI coding tools (Claude Code, Cline, Roo Code, etc.), offering developers top-tier, fast, and stable coding experiences.
+GLM CODING PLAN is a subscription service designed for AI coding, starting at just $3/month. It provides access to their flagship GLM-4.7 model across 10+ popular AI coding tools (Claude Code, Cline, Roo Code, etc.), offering developers top-tier, fast, and stable coding experiences.
 
 Get 10% OFF GLM CODING PLAN：https://z.ai/subscribe?ic=8JVLJQFSKB
 
+---
+
+<table>
+<tbody>
+<tr>
+<td width="180"><a href="https://www.packyapi.com/register?aff=cliproxyapi"><img src="./assets/packycode.png" alt="PackyCode" width="150"></a></td>
+<td>Thanks to PackyCode for sponsoring this project! PackyCode is a reliable and efficient API relay service provider, offering relay services for Claude Code, Codex, Gemini, and more. PackyCode provides special discounts for our software users: register using <a href="https://www.packyapi.com/register?aff=cliproxyapi">this link</a> and enter the "cliproxyapi" promo code during recharge to get 10% off.</td>
+</tr>
+<tr>
+<td width="180"><a href="https://cubence.com/signup?code=CLIPROXYAPI&source=cpa"><img src="./assets/cubence.png" alt="Cubence" width="150"></a></td>
+<td>Thanks to Cubence for sponsoring this project! Cubence is a reliable and efficient API relay service provider, offering relay services for Claude Code, Codex, Gemini, and more. Cubence provides special discounts for our software users: register using <a href="https://cubence.com/signup?code=CLIPROXYAPI&source=cpa">this link</a> and enter the "CLIPROXYAPI" promo code during recharge to get 10% off.</td>
+</tr>
+</tbody>
+</table>
+
 ## Overview
 
 - OpenAI/Gemini/Claude compatible API endpoints for CLI models
@@ -59,7 +74,7 @@ CLIProxyAPI includes integrated support for [Amp CLI](https://ampcode.com) and A
 - **Model mapping** to route unavailable models to alternatives (e.g., `claude-opus-4.5` → `claude-sonnet-4`)
 - Security-first design with localhost-only management endpoints
 
-**→ [Complete Amp CLI Integration Guide](docs/amp-cli-integration.md)**
+**→ [Complete Amp CLI Integration Guide](https://help.router-for.me/agent-client/amp-cli.html)**
 
 ## SDK Docs
 
@@ -99,9 +114,48 @@ CLI wrapper for instant switching between multiple Claude accounts and alternati
 
 Native macOS GUI for managing CLIProxyAPI: configure providers, model mappings, and endpoints via OAuth - no API keys needed.
 
+### [Quotio](https://github.com/nguyenphutrong/quotio)
+
+Native macOS menu bar app that unifies Claude, Gemini, OpenAI, Qwen, and Antigravity subscriptions with real-time quota tracking and smart auto-failover for AI coding tools like Claude Code, OpenCode, and Droid - no API keys needed.
+
+### [CodMate](https://github.com/loocor/CodMate)
+
+Native macOS SwiftUI app for managing CLI AI sessions (Codex, Claude Code, Gemini CLI) with unified provider management, Git review, project organization, global search, and terminal integration. Integrates CLIProxyAPI to provide OAuth authentication for Codex, Claude, Gemini, Antigravity, and Qwen Code, with built-in and third-party provider rerouting through a single proxy endpoint - no API keys needed for OAuth providers.
+
+### [ProxyPilot](https://github.com/Finesssee/ProxyPilot)
+
+Windows-native CLIProxyAPI fork with TUI, system tray, and multi-provider OAuth for AI coding tools - no API keys needed.
+
+### [Claude Proxy VSCode](https://github.com/uzhao/claude-proxy-vscode)
+
+VSCode extension for quick switching between Claude Code models, featuring integrated CLIProxyAPI as its backend with automatic background lifecycle management.
+
+### [ZeroLimit](https://github.com/0xtbug/zero-limit)
+
+Windows desktop app built with Tauri + React for monitoring AI coding assistant quotas via CLIProxyAPI. Track usage across Gemini, Claude, OpenAI Codex, and Antigravity accounts with real-time dashboard, system tray integration, and one-click proxy control - no API keys needed.
+
+### [CPA-XXX Panel](https://github.com/ferretgeek/CPA-X)
+
+A lightweight web admin panel for CLIProxyAPI with health checks, resource monitoring, real-time logs, auto-update, request statistics and pricing display. Supports one-click installation and systemd service.
+
+### [CLIProxyAPI Tray](https://github.com/kitephp/CLIProxyAPI_Tray)
+
+A Windows tray application implemented using PowerShell scripts, without relying on any third-party libraries. The main features include: automatic creation of shortcuts, silent running, password management, channel switching (Main / Plus), and automatic downloading and updating.
+
 > [!NOTE]  
 > If you developed a project based on CLIProxyAPI, please open a PR to add it to this list.
 
+## More choices
+
+Those projects are ports of CLIProxyAPI or inspired by it:
+
+### [9Router](https://github.com/decolua/9router)
+
+A Next.js implementation inspired by CLIProxyAPI, easy to install and use, built from scratch with format translation (OpenAI/Claude/Gemini/Ollama), combo system with auto-fallback, multi-account management with exponential backoff, a Next.js web dashboard, and support for CLI tools (Cursor, Claude Code, Cline, RooCode) - no API keys needed.
+
+> [!NOTE]  
+> If you have developed a port of CLIProxyAPI or a project inspired by it, please open a PR to add it to this list.
+
 ## License
 
 This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

README_CN.md

@@ -10,14 +10,30 @@
 
 ## 赞助商
 
-[![bigmodel.cn](https://assets.router-for.me/chinese.png)](https://www.bigmodel.cn/claude-code?ic=RRVJPB5SII)
+[![bigmodel.cn](https://assets.router-for.me/chinese-4.7.png)](https://www.bigmodel.cn/claude-code?ic=RRVJPB5SII)
 
 本项目由 Z智谱 提供赞助, 他们通过 GLM CODING PLAN 对本项目提供技术支持。
 
-GLM CODING PLAN 是专为AI编码打造的订阅套餐，每月最低仅需20元，即可在十余款主流AI编码工具如 Claude Code、Cline、Roo Code 中畅享智谱旗舰模型GLM-4.6，为开发者提供顶尖的编码体验。
+GLM CODING PLAN 是专为AI编码打造的订阅套餐，每月最低仅需20元，即可在十余款主流AI编码工具如 Claude Code、Cline、Roo Code 中畅享智谱旗舰模型GLM-4.7，为开发者提供顶尖的编码体验。
 
 智谱AI为本软件提供了特别优惠，使用以下链接购买可以享受九折优惠：https://www.bigmodel.cn/claude-code?ic=RRVJPB5SII
 
+---
+
+<table>
+<tbody>
+<tr>
+<td width="180"><a href="https://www.packyapi.com/register?aff=cliproxyapi"><img src="./assets/packycode.png" alt="PackyCode" width="150"></a></td>
+<td>感谢 PackyCode 对本项目的赞助！PackyCode 是一家可靠高效的 API 中转服务商，提供 Claude Code、Codex、Gemini 等多种服务的中转。PackyCode 为本软件用户提供了特别优惠：使用<a href="https://www.packyapi.com/register?aff=cliproxyapi">此链接</a>注册，并在充值时输入 "cliproxyapi" 优惠码即可享受九折优惠。</td>
+</tr>
+<tr>
+<td width="180"><a href="https://cubence.com/signup?code=CLIPROXYAPI&source=cpa"><img src="./assets/cubence.png" alt="Cubence" width="150"></a></td>
+<td>感谢 Cubence 对本项目的赞助！Cubence 是一家可靠高效的 API 中转服务商，提供 Claude Code、Codex、Gemini 等多种服务的中转。Cubence 为本软件用户提供了特别优惠：使用<a href="https://cubence.com/signup?code=CLIPROXYAPI&source=cpa">此链接</a>注册，并在充值时输入 "CLIPROXYAPI" 优惠码即可享受九折优惠。</td>
+</tr>
+</tbody>
+</table>
+
+
 ## 功能特性
 
 - 为 CLI 模型提供 OpenAI/Gemini/Claude/Codex 兼容的 API 端点
@@ -57,7 +73,7 @@ CLIProxyAPI 已内置对 [Amp CLI](https://ampcode.com) 和 Amp IDE 扩展的支
 - 智能模型回退与自动路由
 - 以安全为先的设计，管理端点仅限 localhost
 
-**→ [Amp CLI 完整集成指南](docs/amp-cli-integration_CN.md)**
+**→ [Amp CLI 完整集成指南](https://help.router-for.me/cn/agent-client/amp-cli.html)**
 
 ## SDK 文档
 
@@ -97,9 +113,48 @@ CLI 封装器，用于通过 CLIProxyAPI OAuth 即时切换多个 Claude 账户
 
 基于 macOS 平台的原生 CLIProxyAPI GUI：配置供应商、模型映射以及OAuth端点，无需 API 密钥。
 
+### [Quotio](https://github.com/nguyenphutrong/quotio)
+
+原生 macOS 菜单栏应用，统一管理 Claude、Gemini、OpenAI、Qwen 和 Antigravity 订阅，提供实时配额追踪和智能自动故障转移，支持 Claude Code、OpenCode 和 Droid 等 AI 编程工具，无需 API 密钥。
+
+### [CodMate](https://github.com/loocor/CodMate)
+
+原生 macOS SwiftUI 应用，用于管理 CLI AI 会话（Claude Code、Codex、Gemini CLI），提供统一的提供商管理、Git 审查、项目组织、全局搜索和终端集成。集成 CLIProxyAPI 为 Codex、Claude、Gemini、Antigravity 和 Qwen Code 提供统一的 OAuth 认证，支持内置和第三方提供商通过单一代理端点重路由 - OAuth 提供商无需 API 密钥。
+
+### [ProxyPilot](https://github.com/Finesssee/ProxyPilot)
+
+原生 Windows CLIProxyAPI 分支，集成 TUI、系统托盘及多服务商 OAuth 认证，专为 AI 编程工具打造，无需 API 密钥。
+
+### [Claude Proxy VSCode](https://github.com/uzhao/claude-proxy-vscode)
+
+一款 VSCode 扩展，提供了在 VSCode 中快速切换 Claude Code 模型的功能，内置 CLIProxyAPI 作为其后端，支持后台自动启动和关闭。
+
+### [ZeroLimit](https://github.com/0xtbug/zero-limit)
+
+Windows 桌面应用，基于 Tauri + React 构建，用于通过 CLIProxyAPI 监控 AI 编程助手配额。支持跨 Gemini、Claude、OpenAI Codex 和 Antigravity 账户的使用量追踪，提供实时仪表盘、系统托盘集成和一键代理控制，无需 API 密钥。
+
+### [CPA-XXX Panel](https://github.com/ferretgeek/CPA-X)
+
+面向 CLIProxyAPI 的 Web 管理面板，提供健康检查、资源监控、日志查看、自动更新、请求统计与定价展示，支持一键安装与 systemd 服务。
+
 > [!NOTE]  
 > 如果你开发了基于 CLIProxyAPI 的项目，请提交一个 PR（拉取请求）将其添加到此列表中。
 
+## 更多选择
+
+以下项目是 CLIProxyAPI 的移植版或受其启发：
+
+### [9Router](https://github.com/decolua/9router)
+
+基于 Next.js 的实现，灵感来自 CLIProxyAPI，易于安装使用；自研格式转换（OpenAI/Claude/Gemini/Ollama）、组合系统与自动回退、多账户管理（指数退避）、Next.js Web 控制台，并支持 Cursor、Claude Code、Cline、RooCode 等 CLI 工具，无需 API 密钥。
+
+### [CLIProxyAPI Tray](https://github.com/kitephp/CLIProxyAPI_Tray)
+
+Windows 托盘应用，基于 PowerShell 脚本实现，不依赖任何第三方库。主要功能包括：自动创建快捷方式、静默运行、密码管理、通道切换（Main / Plus）以及自动下载与更新。
+
+> [!NOTE]  
+> 如果你开发了 CLIProxyAPI 的移植或衍生项目，请提交 PR 将其添加到此列表中。
+
 ## 许可证
 
 此项目根据 MIT 许可证授权 - 有关详细信息，请参阅 [LICENSE](LICENSE) 文件。

cmd/server/main.go

@@ -61,6 +61,7 @@ func main() {
 	var iflowLogin bool
 	var iflowCookie bool
 	var noBrowser bool
+	var oauthCallbackPort int
 	var antigravityLogin bool
 	var projectID string
 	var vertexImport string
@@ -75,6 +76,7 @@ func main() {
 	flag.BoolVar(&iflowLogin, "iflow-login", false, "Login to iFlow using OAuth")
 	flag.BoolVar(&iflowCookie, "iflow-cookie", false, "Login to iFlow using Cookie")
 	flag.BoolVar(&noBrowser, "no-browser", false, "Don't open browser automatically for OAuth")
+	flag.IntVar(&oauthCallbackPort, "oauth-callback-port", 0, "Override OAuth callback port (defaults to provider-specific port)")
 	flag.BoolVar(&antigravityLogin, "antigravity-login", false, "Login to Antigravity using OAuth")
 	flag.StringVar(&projectID, "project_id", "", "Project ID (Gemini only, not required)")
 	flag.StringVar(&configPath, "config", DefaultConfigPath, "Configure File Path")
@@ -405,7 +407,7 @@ func main() {
 	usage.SetStatisticsEnabled(cfg.UsageStatisticsEnabled)
 	coreauth.SetQuotaCooldownDisabled(cfg.DisableCooling)
 
-	if err = logging.ConfigureLogOutput(cfg.LoggingToFile); err != nil {
+	if err = logging.ConfigureLogOutput(cfg); err != nil {
 		log.Errorf("failed to configure log output: %v", err)
 		return
 	}
@@ -426,6 +428,7 @@ func main() {
 	// Create login options to be used in authentication flows.
 	options := &cmd.LoginOptions{
 		NoBrowser:    noBrowser,
+		CallbackPort: oauthCallbackPort,
 	}
 
 	// Register the shared token store once so all components use the same persistence backend.

config.example.yaml

@@ -25,6 +25,9 @@ remote-management:
   # Disable the bundled management control panel asset download and HTTP route when true.
   disable-control-panel: false
 
+  # GitHub repository for the management control panel. Accepts a repository URL or releases API URL.
+  panel-github-repository: "https://github.com/router-for-me/Cli-Proxy-API-Management-Center"
+
 # Authentication directory (supports ~ for home directory)
 auth-dir: "~/.cli-proxy-api"
 
@@ -32,19 +35,34 @@ auth-dir: "~/.cli-proxy-api"
 api-keys:
   - "your-api-key-1"
   - "your-api-key-2"
+  - "your-api-key-3"
 
 # Enable debug logging
 debug: false
 
+# When true, disable high-overhead HTTP middleware features to reduce per-request memory usage under high concurrency.
+commercial-mode: false
+
 # When true, write application logs to rotating files instead of stdout
 logging-to-file: false
 
+# Maximum total size (MB) of log files under the logs directory. When exceeded, the oldest log
+# files are deleted until within the limit. Set to 0 to disable.
+logs-max-total-size-mb: 0
+
+# Maximum number of error log files retained when request logging is disabled.
+# When exceeded, the oldest error log files are deleted. Default is 10. Set to 0 to disable cleanup.
+error-logs-max-files: 10
+
 # When false, disable in-memory usage statistics aggregation
 usage-statistics-enabled: false
 
 # Proxy URL. Supports socks5/http/https protocols. Example: socks5://user:pass@192.168.1.1:1080/
 proxy-url: ""
 
+# When true, unprefixed model requests only use credentials without a prefix (except when prefix == model name).
+force-model-prefix: false
+
 # Number of times to retry a request. Retries will occur if the HTTP response code is 403, 408, 500, 502, 503, or 504.
 request-retry: 3
 
@@ -56,16 +74,36 @@ quota-exceeded:
   switch-project: true # Whether to automatically switch to another project when a quota is exceeded
   switch-preview-model: true # Whether to automatically switch to a preview model when a quota is exceeded
 
+# Routing strategy for selecting credentials when multiple match.
+routing:
+  strategy: "round-robin" # round-robin (default), fill-first
+
 # When true, enable authentication for the WebSocket API (/v1/ws).
 ws-auth: false
 
+# When > 0, emit blank lines every N seconds for non-streaming responses to prevent idle timeouts.
+nonstream-keepalive-interval: 0
+
+# Streaming behavior (SSE keep-alives + safe bootstrap retries).
+# streaming:
+#   keepalive-seconds: 15   # Default: 0 (disabled). <= 0 disables keep-alives.
+#   bootstrap-retries: 1    # Default: 0 (disabled). Retries before first byte is sent.
+
+# When true, enable official Codex instructions injection for Codex API requests.
+# When false (default), CodexInstructionsForModel returns immediately without modification.
+codex-instructions-enabled: false
+
 # Gemini API keys
 # gemini-api-key:
 #   - api-key: "AIzaSy...01"
+#     prefix: "test" # optional: require calls like "test/gemini-3-pro-preview" to target this credential
 #     base-url: "https://generativelanguage.googleapis.com"
 #     headers:
 #       X-Custom-Header: "custom-value"
 #     proxy-url: "socks5://proxy.example.com:1080"
+#     models:
+#       - name: "gemini-2.5-flash" # upstream model name
+#         alias: "gemini-flash"    # client alias mapped to the upstream model
 #     excluded-models:
 #       - "gemini-2.5-pro"     # exclude specific models from this provider (exact match)
 #       - "gemini-2.5-*"       # wildcard matching prefix (e.g. gemini-2.5-flash, gemini-2.5-pro)
@@ -76,10 +114,14 @@ ws-auth: false
 # Codex API keys
 # codex-api-key:
 #   - api-key: "sk-atSM..."
+#     prefix: "test" # optional: require calls like "test/gpt-5-codex" to target this credential
 #     base-url: "https://www.example.com" # use the custom codex API endpoint
 #     headers:
 #       X-Custom-Header: "custom-value"
 #     proxy-url: "socks5://proxy.example.com:1080" # optional: per-key proxy override
+#     models:
+#       - name: "gpt-5-codex"   # upstream model name
+#         alias: "codex-latest" # client alias mapped to the upstream model
 #     excluded-models:
 #       - "gpt-5.1"         # exclude specific models (exact match)
 #       - "gpt-5-*"         # wildcard matching prefix (e.g. gpt-5-medium, gpt-5-codex)
@@ -90,6 +132,7 @@ ws-auth: false
 # claude-api-key:
 #   - api-key: "sk-atSM..." # use the official claude API key, no need to set the base url
 #   - api-key: "sk-atSM..."
+#     prefix: "test" # optional: require calls like "test/claude-sonnet-latest" to target this credential
 #     base-url: "https://www.example.com" # use the custom claude API endpoint
 #     headers:
 #       X-Custom-Header: "custom-value"
@@ -102,10 +145,20 @@ ws-auth: false
 #       - "claude-3-*"               # wildcard matching prefix (e.g. claude-3-7-sonnet-20250219)
 #       - "*-thinking"               # wildcard matching suffix (e.g. claude-opus-4-5-thinking)
 #       - "*haiku*"                  # wildcard matching substring (e.g. claude-3-5-haiku-20241022)
+#     cloak:                         # optional: request cloaking for non-Claude-Code clients
+#       mode: "auto"                 # "auto" (default): cloak only when client is not Claude Code
+#                                    # "always": always apply cloaking
+#                                    # "never": never apply cloaking
+#       strict-mode: false           # false (default): prepend Claude Code prompt to user system messages
+#                                    # true: strip all user system messages, keep only Claude Code prompt
+#       sensitive-words:             # optional: words to obfuscate with zero-width characters
+#         - "API"
+#         - "proxy"
 
 # OpenAI compatibility providers
 # openai-compatibility:
 #   - name: "openrouter" # The name of the provider; it will be used in the user agent and other places.
+#     prefix: "test" # optional: require calls like "test/kimi-k2" to target this provider's credentials
 #     base-url: "https://openrouter.ai/api/v1" # The base URL of the provider.
 #     headers:
 #       X-Custom-Header: "custom-value"
@@ -120,14 +173,15 @@ ws-auth: false
 # Vertex API keys (Vertex-compatible endpoints, use API key + base URL)
 # vertex-api-key:
 #   - api-key: "vk-123..."                        # x-goog-api-key header
+#     prefix: "test"                              # optional: require calls like "test/vertex-pro" to target this credential
 #     base-url: "https://example.com/api"         # e.g. https://zenmux.ai/api
 #     proxy-url: "socks5://proxy.example.com:1080" # optional per-key proxy override
 #     headers:
 #       X-Custom-Header: "custom-value"
 #     models:                                     # optional: map aliases to upstream model names
-#       - name: "gemini-2.0-flash"                # upstream model name
+#       - name: "gemini-2.5-flash"                # upstream model name
 #         alias: "vertex-flash"                   # client-visible alias
-#       - name: "gemini-1.5-pro"
+#       - name: "gemini-2.5-pro"
 #         alias: "vertex-pro"
 
 # Amp Integration
@@ -136,8 +190,20 @@ ws-auth: false
 #   upstream-url: "https://ampcode.com"
 #   # Optional: Override API key for Amp upstream (otherwise uses env or file)
 #   upstream-api-key: ""
-#   # Restrict Amp management routes (/api/auth, /api/user, etc.) to localhost only (recommended)
-#   restrict-management-to-localhost: true
+#   # Per-client upstream API key mapping
+#   # Maps client API keys (from top-level api-keys) to different Amp upstream API keys.
+#   # Useful when different clients need to use different Amp accounts/quotas.
+#   # If a client key isn't mapped, falls back to upstream-api-key (default behavior).
+#   upstream-api-keys:
+#     - upstream-api-key: "amp_key_for_team_a"    # Upstream key to use for these clients
+#       api-keys:                                 # Client keys that use this upstream key
+#         - "your-api-key-1"
+#         - "your-api-key-2"
+#     - upstream-api-key: "amp_key_for_team_b"
+#       api-keys:
+#         - "your-api-key-3"
+#   # Restrict Amp management routes (/api/auth, /api/user, etc.) to localhost only (default: false)
+#   restrict-management-to-localhost: false
 #   # Force model mappings to run before checking local API keys (default: false)
 #   force-model-mappings: false
 #   # Amp Model Mappings
@@ -145,12 +211,56 @@ ws-auth: false
 #   # Useful when Amp CLI requests models you don't have access to (e.g., Claude Opus 4.5)
 #   # but you have a similar model available (e.g., Claude Sonnet 4).
 #   model-mappings:
-#     - from: "claude-opus-4.5"       # Model requested by Amp CLI
-#       to: "claude-sonnet-4"         # Route to this available model instead
-#     - from: "gpt-5"
-#       to: "gemini-2.5-pro"
-#     - from: "claude-3-opus-20240229"
-#       to: "claude-3-5-sonnet-20241022"
+#     - from: "claude-opus-4-5-20251101"          # Model requested by Amp CLI
+#       to: "gemini-claude-opus-4-5-thinking"     # Route to this available model instead
+#     - from: "claude-sonnet-4-5-20250929"
+#       to: "gemini-claude-sonnet-4-5-thinking"
+#     - from: "claude-haiku-4-5-20251001"
+#       to: "gemini-2.5-flash"
+
+# Global OAuth model name aliases (per channel)
+# These aliases rename model IDs for both model listing and request routing.
+# Supported channels: gemini-cli, vertex, aistudio, antigravity, claude, codex, qwen, iflow.
+# NOTE: Aliases do not apply to gemini-api-key, codex-api-key, claude-api-key, openai-compatibility, vertex-api-key, or ampcode.
+# You can repeat the same name with different aliases to expose multiple client model names.
+oauth-model-alias:
+  antigravity:
+    - name: "rev19-uic3-1p"
+      alias: "gemini-2.5-computer-use-preview-10-2025"
+    - name: "gemini-3-pro-image"
+      alias: "gemini-3-pro-image-preview"
+    - name: "gemini-3-pro-high"
+      alias: "gemini-3-pro-preview"
+    - name: "gemini-3-flash"
+      alias: "gemini-3-flash-preview"
+    - name: "claude-sonnet-4-5"
+      alias: "gemini-claude-sonnet-4-5"
+    - name: "claude-sonnet-4-5-thinking"
+      alias: "gemini-claude-sonnet-4-5-thinking"
+    - name: "claude-opus-4-5-thinking"
+      alias: "gemini-claude-opus-4-5-thinking"
+#   gemini-cli:
+#     - name: "gemini-2.5-pro"          # original model name under this channel
+#       alias: "g2.5p"                  # client-visible alias
+#       fork: true                      # when true, keep original and also add the alias as an extra model (default: false)
+#   vertex:
+#     - name: "gemini-2.5-pro"
+#       alias: "g2.5p"
+#   aistudio:
+#     - name: "gemini-2.5-pro"
+#       alias: "g2.5p"
+#   claude:
+#     - name: "claude-sonnet-4-5-20250929"
+#       alias: "cs4.5"
+#   codex:
+#     - name: "gpt-5"
+#       alias: "g5"
+#   qwen:
+#     - name: "qwen3-coder-plus"
+#       alias: "qwen-plus"
+#   iflow:
+#     - name: "glm-4.7"
+#       alias: "glm-god"
 
 # OAuth provider excluded models
 # oauth-excluded-models:
@@ -179,12 +289,31 @@ ws-auth: false
 #   default: # Default rules only set parameters when they are missing in the payload.
 #     - models:
 #         - name: "gemini-2.5-pro" # Supports wildcards (e.g., "gemini-*")
-#           protocol: "gemini" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex
+#           protocol: "gemini" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex, antigravity
 #       params: # JSON path (gjson/sjson syntax) -> value
 #         "generationConfig.thinkingConfig.thinkingBudget": 32768
+#   default-raw: # Default raw rules set parameters using raw JSON when missing (must be valid JSON).
+#     - models:
+#         - name: "gemini-2.5-pro" # Supports wildcards (e.g., "gemini-*")
+#           protocol: "gemini" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex, antigravity
+#       params: # JSON path (gjson/sjson syntax) -> raw JSON value (strings are used as-is, must be valid JSON)
+#         "generationConfig.responseJsonSchema": "{\"type\":\"object\",\"properties\":{\"answer\":{\"type\":\"string\"}}}"
 #   override: # Override rules always set parameters, overwriting any existing values.
 #     - models:
 #         - name: "gpt-*" # Supports wildcards (e.g., "gpt-*")
-#           protocol: "codex" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex
+#           protocol: "codex" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex, antigravity
 #       params: # JSON path (gjson/sjson syntax) -> value
 #         "reasoning.effort": "high"
+#   override-raw: # Override raw rules always set parameters using raw JSON (must be valid JSON).
+#     - models:
+#         - name: "gpt-*" # Supports wildcards (e.g., "gpt-*")
+#           protocol: "codex" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex, antigravity
+#       params: # JSON path (gjson/sjson syntax) -> raw JSON value (strings are used as-is, must be valid JSON)
+#         "response_format": "{\"type\":\"json_schema\",\"json_schema\":{\"name\":\"answer\",\"schema\":{\"type\":\"object\"}}}"
+#   filter: # Filter rules remove specified parameters from the payload.
+#     - models:
+#         - name: "gemini-2.5-pro" # Supports wildcards (e.g., "gemini-*")
+#           protocol: "gemini" # restricts the rule to a specific protocol, options: openai, gemini, claude, codex, antigravity
+#       params: # JSON paths (gjson/sjson syntax) to remove from the payload
+#         - "generationConfig.thinkingConfig.thinkingBudget"
+#         - "generationConfig.responseJsonSchema"

docker-build.sh

@@ -5,9 +5,115 @@
 # This script automates the process of building and running the Docker container
 # with version information dynamically injected at build time.
 
-# Exit immediately if a command exits with a non-zero status.
+# Hidden feature: Preserve usage statistics across rebuilds
+# Usage: ./docker-build.sh --with-usage
+# First run prompts for management API key, saved to temp/stats/.api_secret
+
 set -euo pipefail
 
+STATS_DIR="temp/stats"
+STATS_FILE="${STATS_DIR}/.usage_backup.json"
+SECRET_FILE="${STATS_DIR}/.api_secret"
+WITH_USAGE=false
+
+get_port() {
+  if [[ -f "config.yaml" ]]; then
+    grep -E "^port:" config.yaml | sed -E 's/^port: *["'"'"']?([0-9]+)["'"'"']?.*$/\1/'
+  else
+    echo "8317"
+  fi
+}
+
+export_stats_api_secret() {
+  if [[ -f "${SECRET_FILE}" ]]; then
+    API_SECRET=$(cat "${SECRET_FILE}")
+  else
+    if [[ ! -d "${STATS_DIR}" ]]; then
+      mkdir -p "${STATS_DIR}"
+    fi
+    echo "First time using --with-usage. Management API key required."
+    read -r -p "Enter management key: " -s API_SECRET
+    echo
+    echo "${API_SECRET}" > "${SECRET_FILE}"
+    chmod 600 "${SECRET_FILE}"
+  fi
+}
+
+check_container_running() {
+  local port
+  port=$(get_port)
+
+  if ! curl -s -o /dev/null -w "%{http_code}" "http://localhost:${port}/" | grep -q "200"; then
+    echo "Error: cli-proxy-api service is not responding at localhost:${port}"
+    echo "Please start the container first or use without --with-usage flag."
+    exit 1
+  fi
+}
+
+export_stats() {
+  local port
+  port=$(get_port)
+
+  if [[ ! -d "${STATS_DIR}" ]]; then
+    mkdir -p "${STATS_DIR}"
+  fi
+  check_container_running
+  echo "Exporting usage statistics..."
+  EXPORT_RESPONSE=$(curl -s -w "\n%{http_code}" -H "X-Management-Key: ${API_SECRET}" \
+    "http://localhost:${port}/v0/management/usage/export")
+  HTTP_CODE=$(echo "${EXPORT_RESPONSE}" | tail -n1)
+  RESPONSE_BODY=$(echo "${EXPORT_RESPONSE}" | sed '$d')
+
+  if [[ "${HTTP_CODE}" != "200" ]]; then
+    echo "Export failed (HTTP ${HTTP_CODE}): ${RESPONSE_BODY}"
+    exit 1
+  fi
+
+  echo "${RESPONSE_BODY}" > "${STATS_FILE}"
+  echo "Statistics exported to ${STATS_FILE}"
+}
+
+import_stats() {
+  local port
+  port=$(get_port)
+
+  echo "Importing usage statistics..."
+  IMPORT_RESPONSE=$(curl -s -w "\n%{http_code}" -X POST \
+    -H "X-Management-Key: ${API_SECRET}" \
+    -H "Content-Type: application/json" \
+    -d @"${STATS_FILE}" \
+    "http://localhost:${port}/v0/management/usage/import")
+  IMPORT_CODE=$(echo "${IMPORT_RESPONSE}" | tail -n1)
+  IMPORT_BODY=$(echo "${IMPORT_RESPONSE}" | sed '$d')
+
+  if [[ "${IMPORT_CODE}" == "200" ]]; then
+    echo "Statistics imported successfully"
+  else
+    echo "Import failed (HTTP ${IMPORT_CODE}): ${IMPORT_BODY}"
+  fi
+
+  rm -f "${STATS_FILE}"
+}
+
+wait_for_service() {
+  local port
+  port=$(get_port)
+
+  echo "Waiting for service to be ready..."
+  for i in {1..30}; do
+    if curl -s -o /dev/null -w "%{http_code}" "http://localhost:${port}/" | grep -q "200"; then
+      break
+    fi
+    sleep 1
+  done
+  sleep 2
+}
+
+if [[ "${1:-}" == "--with-usage" ]]; then
+  WITH_USAGE=true
+  export_stats_api_secret
+fi
+
 # --- Step 1: Choose Environment ---
 echo "Please select an option:"
 echo "1) Run using Pre-built Image (Recommended)"
@@ -18,7 +124,14 @@ read -r -p "Enter choice [1-2]: " choice
 case "$choice" in
   1)
     echo "--- Running with Pre-built Image ---"
+    if [[ "${WITH_USAGE}" == "true" ]]; then
+      export_stats
+    fi
     docker compose up -d --remove-orphans --no-build
+    if [[ "${WITH_USAGE}" == "true" ]]; then
+      wait_for_service
+      import_stats
+    fi
     echo "Services are starting from remote image."
     echo "Run 'docker compose logs -f' to see the logs."
     ;;
@@ -45,9 +158,18 @@ case "$choice" in
       --build-arg COMMIT="${COMMIT}" \
       --build-arg BUILD_DATE="${BUILD_DATE}"
 
+    if [[ "${WITH_USAGE}" == "true" ]]; then
+      export_stats
+    fi
+
     echo "Starting the services..."
     docker compose up -d --remove-orphans --pull never
 
+    if [[ "${WITH_USAGE}" == "true" ]]; then
+      wait_for_service
+      import_stats
+    fi
+
     echo "Build complete. Services are starting."
     echo "Run 'docker compose logs -f' to see the logs."
     ;;

docker-compose.yml

@@ -22,7 +22,7 @@ services:
       - "51121:51121"
       - "11451:11451"
     volumes:
-      - ./config.yaml:/CLIProxyAPI/config.yaml
-      - ./auths:/root/.cli-proxy-api
-      - ./logs:/CLIProxyAPI/logs
+      - ${CLI_PROXY_CONFIG_PATH:-./config.yaml}:/CLIProxyAPI/config.yaml
+      - ${CLI_PROXY_AUTH_PATH:-./auths}:/root/.cli-proxy-api
+      - ${CLI_PROXY_LOG_PATH:-./logs}:/CLIProxyAPI/logs
     restart: unless-stopped

docs/amp-cli-integration.md

@@ -1,443 +0,0 @@
-# Amp CLI Integration Guide
-
-This guide explains how to use CLIProxyAPI with Amp CLI and Amp IDE extensions, enabling you to use your existing Google/ChatGPT/Claude subscriptions (via OAuth) with Amp's CLI.
-
-## Table of Contents
-
-- [Overview](#overview)
-  - [Which Providers Should You Authenticate?](#which-providers-should-you-authenticate)
-- [Architecture](#architecture)
-- [Configuration](#configuration)
-  - [Model Mapping Configuration](#model-mapping-configuration)
-- [Setup](#setup)
-- [Usage](#usage)
-- [Troubleshooting](#troubleshooting)
-
-## Overview
-
-The Amp CLI integration adds specialized routing to support Amp's API patterns while maintaining full compatibility with all existing CLIProxyAPI features. This allows you to use both traditional CLIProxyAPI features and Amp CLI with the same proxy server.
-
-### Key Features
-
-- **Provider route aliases**: Maps Amp's `/api/provider/{provider}/v1...` patterns to CLIProxyAPI handlers
-- **Management proxy**: Forwards OAuth and account management requests to Amp's control plane
-- **Smart fallback**: Automatically routes unconfigured models to ampcode.com
-- **Model mapping**: Route unavailable models to alternatives you have access to (e.g., `claude-opus-4.5` → `claude-sonnet-4`)
-- **Secret management**: Configurable precedence (config > env > file) with 5-minute caching
-- **Security-first**: Management routes restricted to localhost by default
-- **Automatic gzip handling**: Decompresses responses from Amp upstream
-
-### What You Can Do
-
-- Use Amp CLI with your Google account (Gemini 3 Pro Preview, Gemini 2.5 Pro, Gemini 2.5 Flash)
-- Use Amp CLI with your ChatGPT Plus/Pro subscription (GPT-5, GPT-5 Codex models)
-- Use Amp CLI with your Claude Pro/Max subscription (Claude Sonnet 4.5, Opus 4.1)
-- Use Amp IDE extensions (VS Code, Cursor, Windsurf, etc.) with the same proxy
-- Run multiple CLI tools (Factory + Amp) through one proxy server
-- Route unconfigured models automatically through ampcode.com
-
-### Which Providers Should You Authenticate?
-
-**Important**: The providers you need to authenticate depend on which models and features your installed version of Amp currently uses. Amp employs different providers for various agent modes and specialized subagents:
-
-- **Smart mode**: Uses Google/Gemini models (Gemini 3 Pro)
-- **Rush mode**: Uses Anthropic/Claude models (Claude Haiku 4.5)
-- **Oracle subagent**: Uses OpenAI/GPT models (GPT-5 medium reasoning)
-- **Librarian subagent**: Uses Anthropic/Claude models (Claude Sonnet 4.5)
-- **Search subagent**: Uses Anthropic/Claude models (Claude Haiku 4.5)
-- **Review feature**: Uses Google/Gemini models (Gemini 2.5 Flash-Lite)
-
-For the most current information about which models Amp uses, see the **[Amp Models Documentation](https://ampcode.com/models)**.
-
-#### Fallback Behavior
-
-CLIProxyAPI uses a smart fallback system:
-
-1. **Provider authenticated locally** (`--login`, `--codex-login`, `--claude-login`):
-   - Requests use **your OAuth subscription** (ChatGPT Plus/Pro, Claude Pro/Max, Google account)
-   - You benefit from your subscription's included usage quotas
-   - No Amp credits consumed
-
-2. **Provider NOT authenticated locally**:
-   - Requests automatically forward to **ampcode.com**
-   - Uses Amp's backend provider connections
-   - **Requires Amp credits** if the provider is paid (OpenAI, Anthropic paid tiers)
-   - May result in errors if Amp credit balance is insufficient
-
-**Recommendation**: Authenticate all providers you have subscriptions for to maximize value and minimize Amp credit usage. If you don't have subscriptions to all providers Amp uses, ensure you have sufficient Amp credits available for fallback requests.
-
-## Architecture
-
-### Request Flow
-
-```
-Amp CLI/IDE
-  ↓
-  ├─ Provider API requests (/api/provider/{provider}/v1/...)
-  │   ↓
-  │   ├─ Model configured locally?
-  │   │   YES → Use local OAuth tokens (OpenAI/Claude/Gemini handlers)
-  │   │   NO  ↓
-  │   │       ├─ Model mapping configured?
-  │   │       │   YES → Rewrite model → Use local handler (free)
-  │   │       │   NO  → Forward to ampcode.com (uses Amp credits)
-  │   ↓
-  │   Response
-  │
-  └─ Management requests (/api/auth, /api/user, /api/threads, ...)
-      ↓
-      ├─ Localhost check (security)
-      ↓
-      └─ Reverse proxy to ampcode.com
-          ↓
-          Response (auto-decompressed if gzipped)
-```
-
-### Components
-
-The Amp integration is implemented as a modular routing module (`internal/api/modules/amp/`) with these components:
-
-1. **Route Aliases** (`routes.go`): Maps Amp-style paths to standard handlers
-2. **Reverse Proxy** (`proxy.go`): Forwards management requests to ampcode.com
-3. **Fallback Handler** (`fallback_handlers.go`): Routes unconfigured models to ampcode.com
-4. **Secret Management** (`secret.go`): Multi-source API key resolution with caching
-5. **Main Module** (`amp.go`): Orchestrates registration and configuration
-
-## Configuration
-
-### Basic Configuration
-
-Add these fields to your `config.yaml`:
-
-```yaml
-# Amp upstream control plane (required for management routes)
-amp-upstream-url: "https://ampcode.com"
-
-# Optional: Override API key (otherwise uses env or file)
-# amp-upstream-api-key: "your-amp-api-key"
-
-# Security: restrict management routes to localhost (recommended)
-amp-restrict-management-to-localhost: true
-```
-
-### Model Mapping Configuration
-
-When Amp CLI requests a model that you don't have access to, you can configure mappings to route those requests to alternative models that you DO have available. This avoids consuming Amp credits for models you could handle locally.
-
-```yaml
-# Route unavailable models to alternatives
-amp-model-mappings:
-  # Example: Route Claude Opus 4.5 requests to Claude Sonnet 4
-  - from: "claude-opus-4.5"
-    to: "claude-sonnet-4"
-  
-  # Example: Route GPT-5 requests to Gemini 2.5 Pro
-  - from: "gpt-5"
-    to: "gemini-2.5-pro"
-  
-  # Example: Map older model names to newer versions
-  - from: "claude-3-opus-20240229"
-    to: "claude-3-5-sonnet-20241022"
-```
-
-**How it works:**
-
-1. Amp CLI requests a model (e.g., `claude-opus-4.5`)
-2. CLIProxyAPI checks if a local provider is available for that model
-3. If not available, it checks the model mappings
-4. If a mapping exists, the request is rewritten to use the target model
-5. The request is then handled locally (free, using your OAuth subscription)
-
-**Benefits:**
-- **Save Amp credits**: Use your local subscriptions instead of forwarding to ampcode.com
-- **Hot-reload**: Mappings can be updated without restarting the proxy
-- **Structured logging**: Clear logs show when mappings are applied
-
-**Routing Decision Logs:**
-
-The proxy logs each routing decision with structured fields:
-
-```
-[AMP] Using local provider for model: gemini-2.5-pro          # Local provider (free)
-[AMP] Model mapped: claude-opus-4.5 -> claude-sonnet-4        # Mapping applied (free)
-[AMP] Forwarding to ampcode.com (uses Amp credits) - model_id: gpt-5  # Fallback (costs credits)
-```
-
-### Secret Resolution Precedence
-
-The Amp module resolves API keys using this precedence order:
-
-| Source | Key | Priority | Cache |
-|--------|-----|----------|-------|
-| Config file | `amp-upstream-api-key` | High | No |
-| Environment | `AMP_API_KEY` | Medium | No |
-| Amp secrets file | `~/.local/share/amp/secrets.json` | Low | 5 min |
-
-**Recommendation**: Use the Amp secrets file (lowest precedence) for normal usage. This file is automatically managed by `amp login`.
-
-### Security Settings
-
-**`amp-restrict-management-to-localhost`** (default: `true`)
-
-When enabled, management routes (`/api/auth`, `/api/user`, `/api/threads`, etc.) only accept connections from localhost (127.0.0.1, ::1). This prevents:
-- Drive-by browser attacks
-- Remote access to management endpoints
-- CORS-based attacks
-- Header spoofing attacks (e.g., `X-Forwarded-For: 127.0.0.1`)
-
-#### How It Works
-
-This restriction uses the **actual TCP connection address** (`RemoteAddr`), not HTTP headers like `X-Forwarded-For`. This prevents header spoofing attacks but has important implications:
-
-- ✅ **Works for direct connections**: Running CLIProxyAPI directly on your machine or server
-- ⚠️ **May not work behind reverse proxies**: If deploying behind nginx, Cloudflare, or other proxies, the connection will appear to come from the proxy's IP, not localhost
-
-#### Reverse Proxy Deployments
-
-If you need to run CLIProxyAPI behind a reverse proxy (nginx, Caddy, Cloudflare Tunnel, etc.):
-
-1. **Disable the localhost restriction**:
-   ```yaml
-   amp-restrict-management-to-localhost: false
-   ```
-
-2. **Use alternative security measures**:
-   - Firewall rules restricting access to management routes
-   - Proxy-level authentication (HTTP Basic Auth, OAuth)
-   - Network-level isolation (VPN, Tailscale, Cloudflare Access)
-   - Bind CLIProxyAPI to `127.0.0.1` only and access via SSH tunnel
-
-3. **Example nginx configuration** (blocks external access to management routes):
-   ```nginx
-   location /api/auth { deny all; }
-   location /api/user { deny all; }
-   location /api/threads { deny all; }
-   location /api/internal { deny all; }
-   ```
-
-**Important**: Only disable `amp-restrict-management-to-localhost` if you understand the security implications and have other protections in place.
-
-## Setup
-
-### 1. Configure CLIProxyAPI
-
-Create or edit `config.yaml`:
-
-```yaml
-port: 8317
-auth-dir: "~/.cli-proxy-api"
-
-# Amp integration
-amp-upstream-url: "https://ampcode.com"
-amp-restrict-management-to-localhost: true
-
-# Other standard settings...
-debug: false
-logging-to-file: true
-```
-
-### 2. Authenticate with Providers
-
-Run OAuth login for the providers you want to use:
-
-**Google Account (Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 3 Pro Preview):**
-```bash
-./cli-proxy-api --login
-```
-
-**ChatGPT Plus/Pro (GPT-5, GPT-5 Codex):**
-```bash
-./cli-proxy-api --codex-login
-```
-
-**Claude Pro/Max (Claude Sonnet 4.5, Opus 4.1):**
-```bash
-./cli-proxy-api --claude-login
-```
-
-Tokens are saved to:
-- Gemini: `~/.cli-proxy-api/gemini-<email>.json`
-- OpenAI Codex: `~/.cli-proxy-api/codex-<email>.json`
-- Claude: `~/.cli-proxy-api/claude-<email>.json`
-
-### 3. Start the Proxy
-
-```bash
-./cli-proxy-api --config config.yaml
-```
-
-Or run in background with tmux (recommended for remote servers):
-
-```bash
-tmux new-session -d -s proxy "./cli-proxy-api --config config.yaml"
-```
-
-### 4. Configure Amp CLI
-
-#### Option A: Settings File
-
-Edit `~/.config/amp/settings.json`:
-
-```json
-{
-  "amp.url": "http://localhost:8317"
-}
-```
-
-#### Option B: Environment Variable
-
-```bash
-export AMP_URL=http://localhost:8317
-```
-
-### 5. Login and Use Amp
-
-Login through the proxy (proxied to ampcode.com):
-
-```bash
-amp login
-```
-
-Use Amp as normal:
-
-```bash
-amp "Write a hello world program in Python"
-```
-
-### 6. (Optional) Configure Amp IDE Extension
-
-The proxy also works with Amp IDE extensions for VS Code, Cursor, Windsurf, etc.
-
-1. Open Amp extension settings in your IDE
-2. Set **Amp URL** to `http://localhost:8317`
-3. Login with your Amp account
-4. Start using Amp in your IDE
-
-Both CLI and IDE can use the proxy simultaneously.
-
-## Usage
-
-### Supported Routes
-
-#### Provider Aliases (Always Available)
-
-These routes work even without `amp-upstream-url` configured:
-
-- `/api/provider/openai/v1/chat/completions`
-- `/api/provider/openai/v1/responses`
-- `/api/provider/anthropic/v1/messages`
-- `/api/provider/google/v1beta/models/:action`
-
-Amp CLI calls these routes with your OAuth-authenticated models configured in CLIProxyAPI.
-
-#### Management Routes (Require `amp-upstream-url`)
-
-These routes are proxied to ampcode.com:
-
-- `/api/auth` - Authentication
-- `/api/user` - User profile
-- `/api/meta` - Metadata
-- `/api/threads` - Conversation threads
-- `/api/telemetry` - Usage telemetry
-- `/api/internal` - Internal APIs
-
-**Security**: Restricted to localhost by default.
-
-### Model Fallback Behavior
-
-When Amp requests a model:
-
-1. **Check local configuration**: Does CLIProxyAPI have OAuth tokens for this model's provider?
-2. **If YES**: Route to local handler (use your OAuth subscription)
-3. **If NO**: Check if a model mapping exists
-4. **If mapping exists**: Rewrite request to mapped model → Route to local handler (free)
-5. **If no mapping**: Forward to ampcode.com (uses Amp credits)
-
-This enables seamless mixed usage:
-- Models you've configured (Gemini, ChatGPT, Claude) → Your OAuth subscriptions
-- Models with mappings configured → Routed to alternative local models (free)
-- Models you haven't configured and have no mapping → Amp's default providers (uses credits)
-
-### Example API Calls
-
-**Chat completion with local OAuth:**
-```bash
-curl http://localhost:8317/api/provider/openai/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "gpt-5",
-    "messages": [{"role": "user", "content": "Hello"}]
-  }'
-```
-
-**Management endpoint (localhost only):**
-```bash
-curl http://localhost:8317/api/user
-```
-
-## Troubleshooting
-
-### Common Issues
-
-| Symptom | Likely Cause | Fix |
-|---------|--------------|-----|
-| 404 on `/api/provider/...` | Incorrect route path | Ensure exact path: `/api/provider/{provider}/v1...` |
-| 403 on `/api/user` | Non-localhost request | Run from same machine or disable `amp-restrict-management-to-localhost` (not recommended) |
-| 401/403 from provider | Missing/expired OAuth | Re-run `--codex-login` or `--claude-login` |
-| Amp gzip errors | Response decompression issue | Update to latest build; auto-decompression should handle this |
-| Models not using proxy | Wrong Amp URL | Verify `amp.url` setting or `AMP_URL` environment variable |
-| CORS errors | Protected management endpoint | Use CLI/terminal, not browser |
-
-### Diagnostics
-
-**Check proxy logs:**
-```bash
-# If logging-to-file: true
-tail -f logs/requests.log
-
-# If running in tmux
-tmux attach-session -t proxy
-```
-
-**Enable debug mode** (temporarily):
-```yaml
-debug: true
-```
-
-**Test basic connectivity:**
-```bash
-# Check if proxy is running
-curl http://localhost:8317/v1/models
-
-# Check Amp-specific route
-curl http://localhost:8317/api/provider/openai/v1/models
-```
-
-**Verify Amp configuration:**
-```bash
-# Check if Amp is using proxy
-amp config get amp.url
-
-# Or check environment
-echo $AMP_URL
-```
-
-### Security Checklist
-
-- ✅ Keep `amp-restrict-management-to-localhost: true` (default)
-- ✅ Don't expose proxy publicly (bind to localhost or use firewall/VPN)
-- ✅ Use the Amp secrets file (`~/.local/share/amp/secrets.json`) managed by `amp login`
-- ✅ Rotate OAuth tokens periodically by re-running login commands
-- ✅ Store config and auth-dir on encrypted disk if handling sensitive data
-- ✅ Keep proxy binary up to date for security fixes
-
-## Additional Resources
-
-- [CLIProxyAPI Main Documentation](https://help.router-for.me/)
-- [Amp CLI Official Manual](https://ampcode.com/manual)
-- [Management API Reference](https://help.router-for.me/management/api)
-- [SDK Documentation](sdk-usage.md)
-
-## Disclaimer
-
-This integration is for personal/educational use. Using reverse proxies or alternate API bases may violate provider Terms of Service. You are solely responsible for how you use this software. Accounts may be rate-limited, locked, or banned. No warranties. Use at your own risk.

docs/amp-cli-integration_CN.md

@@ -1,392 +0,0 @@
-# Amp CLI 集成指南
-
-本指南说明如何在 Amp CLI 和 Amp IDE 扩展中使用 CLIProxyAPI，通过 OAuth 让你能够把已有的 Google/ChatGPT/Claude 订阅与 Amp 的 CLI 一起使用。
-
-## 目录
-
-- [概述](#概述)
-  - [应该认证哪些服务提供商？](#应该认证哪些服务提供商)
-- [架构](#架构)
-- [配置](#配置)
-- [设置](#设置)
-- [用法](#用法)
-- [故障排查](#故障排查)
-
-## 概述
-
-Amp CLI 集成为 Amp 的 API 模式添加了专用路由，同时保持与现有 CLIProxyAPI 功能的完全兼容。这样你可以在同一个代理服务器上同时使用传统 CLIProxyAPI 功能和 Amp CLI。
-
-### 主要特性
-
-- **提供者路由别名**：将 Amp 的 `/api/provider/{provider}/v1...` 路径映射到 CLIProxyAPI 处理器
-- **管理代理**：将 OAuth 和账号管理请求转发到 Amp 控制平面
-- **智能回退**：自动将未配置的模型路由到 ampcode.com
-- **密钥管理**：可配置优先级（配置 > 环境变量 > 文件），缓存 5 分钟
-- **安全优先**：管理路由默认限制为 localhost
-- **自动 gzip 处理**：自动解压来自 Amp 上游的响应
-
-### 你可以做什么
-
-- 使用 Amp CLI 搭配你的 Google 账号（Gemini 3 Pro Preview、Gemini 2.5 Pro、Gemini 2.5 Flash）
-- 使用 Amp CLI 搭配你的 ChatGPT Plus/Pro 订阅（GPT-5、GPT-5 Codex 模型）
-- 使用 Amp CLI 搭配你的 Claude Pro/Max 订阅（Claude Sonnet 4.5、Opus 4.1）
-- 将 Amp IDE 扩展（VS Code、Cursor、Windsurf 等）与同一个代理一起使用
-- 通过一个代理同时运行多个 CLI 工具（Factory + Amp）
-- 将未配置的模型自动路由到 ampcode.com
-
-### 应该认证哪些服务提供商？
-
-**重要**：需要认证的提供商取决于你安装的 Amp 版本当前使用的模型和功能。Amp 的不同智能模式和子代理会使用不同的提供商：
-
-- **Smart 模式**：使用 Google/Gemini 模型（Gemini 3 Pro）
-- **Rush 模式**：使用 Anthropic/Claude 模型（Claude Haiku 4.5）
-- **Oracle 子代理**：使用 OpenAI/GPT 模型（GPT-5 medium reasoning）
-- **Librarian 子代理**：使用 Anthropic/Claude 模型（Claude Sonnet 4.5）
-- **Search 子代理**：使用 Anthropic/Claude 模型（Claude Haiku 4.5）
-- **Review 功能**：使用 Google/Gemini 模型（Gemini 2.5 Flash-Lite）
-
-有关 Amp 当前使用哪些模型的最新信息，请参阅 **[Amp 模型文档](https://ampcode.com/models)**。
-
-#### 回退行为
-
-CLIProxyAPI 采用智能回退机制：
-
-1. **本地已认证提供商**（`--login`、`--codex-login`、`--claude-login`）：
-   - 请求使用**你的 OAuth 订阅**（ChatGPT Plus/Pro、Claude Pro/Max、Google 账号）
-   - 享受订阅自带的额度
-   - 不消耗 Amp 额度
-
-2. **本地未认证提供商**：
-   - 请求自动转发到 **ampcode.com**
-   - 使用 Amp 的后端提供商连接
-   - 如果提供商是付费的（OpenAI、Anthropic 付费档），**需要消耗 Amp 额度**
-   - 若 Amp 额度不足，可能产生错误
-
-**建议**：对你有订阅的所有提供商都进行认证，以最大化价值并尽量减少 Amp 额度消耗。如果没有覆盖 Amp 使用的全部提供商，请确保为回退请求准备足够的 Amp 额度。
-
-## 架构
-
-### 请求流
-
-```
-Amp CLI/IDE
-  ↓
-  ├─ Provider API requests (/api/provider/{provider}/v1/...)
-  │   ↓
-  │   ├─ Model configured locally?
-  │   │   YES → Use local OAuth tokens (OpenAI/Claude/Gemini handlers)
-  │   │   NO  → Forward to ampcode.com (reverse proxy)
-  │   ↓
-  │   Response
-  │
-  └─ Management requests (/api/auth, /api/user, /api/threads, ...)
-      ↓
-      ├─ Localhost check (security)
-      ↓
-      └─ Reverse proxy to ampcode.com
-          ↓
-          Response (auto-decompressed if gzipped)
-```
-
-### 组件
-
-Amp 集成以模块化路由模块（`internal/api/modules/amp/`）实现，包含以下组件：
-
-1. **路由别名**（`routes.go`）：将 Amp 风格的路径映射到标准处理器
-2. **反向代理**（`proxy.go`）：将管理请求转发到 ampcode.com
-3. **回退处理器**（`fallback_handlers.go`）：将未配置的模型路由到 ampcode.com
-4. **密钥管理**（`secret.go`）：多来源 API 密钥解析并带缓存
-5. **主模块**（`amp.go`）：负责注册和配置
-
-## 配置
-
-### 基础配置
-
-在 `config.yaml` 中新增以下字段：
-
-```yaml
-# Amp 上游控制平面（管理路由必需）
-amp-upstream-url: "https://ampcode.com"
-
-# 可选：覆盖 API key（否则使用环境变量或文件）
-# amp-upstream-api-key: "your-amp-api-key"
-
-# 安全性：将管理路由限制为 localhost（推荐）
-amp-restrict-management-to-localhost: true
-```
-
-### 密钥解析优先级
-
-Amp 模块以如下优先级解析 API key：
-
-| 来源 | 键名 | 优先级 | 缓存 |
-|------|------|--------|------|
-| 配置文件 | `amp-upstream-api-key` | 高 | 无 |
-| 环境变量 | `AMP_API_KEY` | 中 | 无 |
-| Amp 密钥文件 | `~/.local/share/amp/secrets.json` | 低 | 5 分钟 |
-
-**建议**：日常使用时采用 Amp 密钥文件（最低优先级）。该文件由 `amp login` 自动管理。
-
-### 安全设置
-
-**`amp-restrict-management-to-localhost`**（默认：`true`）
-
-启用后，管理路由（`/api/auth`、`/api/user`、`/api/threads` 等）只接受来自 localhost（127.0.0.1、::1）的连接，可防止：
-- 浏览器探测式攻击
-- 对管理端点的远程访问
-- 基于 CORS 的攻击
-- 伪造头攻击（例如 `X-Forwarded-For: 127.0.0.1`）
-
-#### 工作原理
-
-此限制使用**实际的 TCP 连接地址**（`RemoteAddr`），而非 `X-Forwarded-For` 等 HTTP 头，能防止头部伪造，但有重要影响：
-
-- ✅ **直接连接可用**：在本机或服务器直接运行 CLIProxyAPI 时适用
-- ⚠️ **可能不适用于反向代理场景**：部署在 nginx、Cloudflare 等代理后，请求源会显示为代理 IP 而非 localhost
-
-#### 反向代理部署
-
-若需要在反向代理（nginx、Caddy、Cloudflare Tunnel 等）后运行 CLIProxyAPI：
-
-1. **关闭 localhost 限制**：
-   ```yaml
-   amp-restrict-management-to-localhost: false
-   ```
-
-2. **使用替代安全措施**：
-   - 防火墙规则限制管理路由访问
-   - 代理层认证（HTTP Basic Auth、OAuth）
-   - 网络隔离（VPN、Tailscale、Cloudflare Access）
-   - 将 CLIProxyAPI 仅绑定 `127.0.0.1`，并通过 SSH 隧道访问
-
-3. **nginx 示例配置**（阻止外部访问管理路由）：
-   ```nginx
-   location /api/auth { deny all; }
-   location /api/user { deny all; }
-   location /api/threads { deny all; }
-   location /api/internal { deny all; }
-   ```
-
-**重要**：只有在理解安全影响并已采取其他防护措施时，才关闭 `amp-restrict-management-to-localhost`。
-
-## 设置
-
-### 1. 配置 CLIProxyAPI
-
-创建或编辑 `config.yaml`：
-
-```yaml
-port: 8317
-auth-dir: "~/.cli-proxy-api"
-
-# Amp 集成
-amp-upstream-url: "https://ampcode.com"
-amp-restrict-management-to-localhost: true
-
-# 其他常规设置...
-debug: false
-logging-to-file: true
-```
-
-### 2. 认证提供商
-
-为要使用的提供商执行 OAuth 登录：
-
-**Google 账号（Gemini 2.5 Pro、Gemini 2.5 Flash、Gemini 3 Pro Preview）：**
-```bash
-./cli-proxy-api --login
-```
-
-**ChatGPT Plus/Pro（GPT-5、GPT-5 Codex）：**
-```bash
-./cli-proxy-api --codex-login
-```
-
-**Claude Pro/Max（Claude Sonnet 4.5、Opus 4.1）：**
-```bash
-./cli-proxy-api --claude-login
-```
-
-令牌会保存到：
-- Gemini: `~/.cli-proxy-api/gemini-<email>.json`
-- OpenAI Codex: `~/.cli-proxy-api/codex-<email>.json`
-- Claude: `~/.cli-proxy-api/claude-<email>.json`
-
-### 3. 启动代理
-
-```bash
-./cli-proxy-api --config config.yaml
-```
-
-或使用 tmux 在后台运行（推荐用于远程服务器）：
-
-```bash
-tmux new-session -d -s proxy "./cli-proxy-api --config config.yaml"
-```
-
-### 4. 配置 Amp CLI
-
-#### 方案 A：配置文件
-
-编辑 `~/.config/amp/settings.json`：
-
-```json
-{
-  "amp.url": "http://localhost:8317"
-}
-```
-
-#### 方案 B：环境变量
-
-```bash
-export AMP_URL=http://localhost:8317
-```
-
-### 5. 登录并使用 Amp
-
-通过代理登录（请求会被代理到 ampcode.com）：
-
-```bash
-amp login
-```
-
-像平常一样使用 Amp：
-
-```bash
-amp "Write a hello world program in Python"
-```
-
-### 6. （可选）配置 Amp IDE 扩展
-
-该代理同样适用于 VS Code、Cursor、Windsurf 等 Amp IDE 扩展。
-
-1. 在 IDE 中打开 Amp 扩展设置
-2. 将 **Amp URL** 设置为 `http://localhost:8317`
-3. 用你的 Amp 账号登录
-4. 在 IDE 中开始使用 Amp
-
-CLI 和 IDE 可同时使用该代理。
-
-## 用法
-
-### 支持的路由
-
-#### 提供商别名（始终可用）
-
-这些路由即使未配置 `amp-upstream-url` 也可使用：
-
-- `/api/provider/openai/v1/chat/completions`
-- `/api/provider/openai/v1/responses`
-- `/api/provider/anthropic/v1/messages`
-- `/api/provider/google/v1beta/models/:action`
-
-Amp CLI 会使用你在 CLIProxyAPI 中通过 OAuth 认证的模型来调用这些路由。
-
-#### 管理路由（需要 `amp-upstream-url`）
-
-这些路由会被代理到 ampcode.com：
-
-- `/api/auth` - 认证
-- `/api/user` - 用户资料
-- `/api/meta` - 元数据
-- `/api/threads` - 会话线程
-- `/api/telemetry` - 使用遥测
-- `/api/internal` - 内部 API
-
-**安全性**：默认限制为 localhost。
-
-### 模型回退行为
-
-当 Amp 请求模型时：
-
-1. **检查本地配置**：CLIProxyAPI 是否有该模型提供商的 OAuth 令牌？
-2. **如果有**：路由到本地处理器（使用你的 OAuth 订阅）
-3. **如果没有**：转发到 ampcode.com（使用 Amp 的默认路由）
-
-这实现了无缝混用：
-- 你已配置的模型（Gemini、ChatGPT、Claude）→ 你的 OAuth 订阅
-- 未配置的模型 → Amp 的默认提供商
-
-### 示例 API 调用
-
-**使用本地 OAuth 的聊天补全：**
-```bash
-curl http://localhost:8317/api/provider/openai/v1/chat/completions \
-  -H "Content-Type: application/json" \
-  -d '{
-    "model": "gpt-5",
-    "messages": [{"role": "user", "content": "Hello"}]
-  }'
-```
-
-**管理端点（仅限 localhost）：**
-```bash
-curl http://localhost:8317/api/user
-```
-
-## 故障排查
-
-### 常见问题
-
-| 症状 | 可能原因 | 解决方案 |
-|------|----------|----------|
-| `/api/provider/...` 返回 404 | 路径错误 | 确保路径准确：`/api/provider/{provider}/v1...` |
-| `/api/user` 返回 403 | 非 localhost 请求 | 在同一机器上访问，或关闭 `amp-restrict-management-to-localhost`（不推荐） |
-| 提供商返回 401/403 | OAuth 缺失或过期 | 重新运行 `--codex-login` 或 `--claude-login` |
-| Amp gzip 错误 | 响应解压问题 | 更新到最新构建；自动解压应能处理 |
-| 模型未走代理 | Amp URL 设置错误 | 检查 `amp.url` 设置或 `AMP_URL` 环境变量 |
-| CORS 错误 | 受保护的管理端点 | 使用 CLI/终端而非浏览器 |
-
-### 诊断
-
-**查看代理日志：**
-```bash
-# 若 logging-to-file: true
-tail -f logs/requests.log
-
-# 若运行在 tmux 中
-tmux attach-session -t proxy
-```
-
-**临时开启调试模式：**
-```yaml
-debug: true
-```
-
-**测试基础连通性：**
-```bash
-# 检查代理是否运行
-curl http://localhost:8317/v1/models
-
-# 检查 Amp 特定路由
-curl http://localhost:8317/api/provider/openai/v1/models
-```
-
-**验证 Amp 配置：**
-```bash
-# 检查 Amp 是否使用代理
-amp config get amp.url
-
-# 或检查环境变量
-echo $AMP_URL
-```
-
-### 安全清单
-
-- ✅ 保持 `amp-restrict-management-to-localhost: true`（默认）
-- ✅ 不要将代理暴露到公共网络（绑定到 localhost 或使用防火墙/VPN）
-- ✅ 使用 `amp login` 管理的 Amp 密钥文件（`~/.local/share/amp/secrets.json`）
-- ✅ 定期重新登录轮换 OAuth 令牌
-- ✅ 若处理敏感数据，使用加密磁盘存储配置和 auth-dir
-- ✅ 保持代理二进制为最新版本以获取安全修复
-
-## 其他资源
-
-- [CLIProxyAPI 主文档](https://help.router-for.me/)
-- [Amp CLI 官方手册](https://ampcode.com/manual)
-- [管理 API 参考](https://help.router-for.me/management/api)
-- [SDK 文档](sdk-usage.md)
-
-## 免责声明
-
-此集成仅用于个人或教育用途。使用反向代理或替代 API 基址可能违反提供商的服务条款。你需要对自己的使用方式负责。账号可能会被限速、锁定或封禁。软件不附带任何保证，使用风险自负。

examples/custom-provider/main.go

@@ -14,6 +14,7 @@ import (
 	"bytes"
 	"context"
 	"errors"
+	"fmt"
 	"io"
 	"net/http"
 	"net/url"
@@ -23,13 +24,13 @@ import (
 	"time"
 
 	"github.com/gin-gonic/gin"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/api"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api"
 	sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy"
 	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
 	clipexec "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/logging"
 	sdktr "github.com/router-for-me/CLIProxyAPI/v6/sdk/translator"
 )
 
@@ -122,7 +123,9 @@ func (MyExecutor) Execute(ctx context.Context, a *coreauth.Auth, req clipexec.Re
 	httpReq.Header.Set("Content-Type", "application/json")
 
 	// Inject credentials via PrepareRequest hook.
-	_ = (MyExecutor{}).PrepareRequest(httpReq, a)
+	if errPrep := (MyExecutor{}).PrepareRequest(httpReq, a); errPrep != nil {
+		return clipexec.Response{}, errPrep
+	}
 
 	resp, errDo := client.Do(httpReq)
 	if errDo != nil {
@@ -130,13 +133,28 @@ func (MyExecutor) Execute(ctx context.Context, a *coreauth.Auth, req clipexec.Re
 	}
 	defer func() {
 		if errClose := resp.Body.Close(); errClose != nil {
-			// Best-effort close; log if needed in real projects.
+			fmt.Fprintf(os.Stderr, "close response body error: %v\n", errClose)
 		}
 	}()
 	body, _ := io.ReadAll(resp.Body)
 	return clipexec.Response{Payload: body}, nil
 }
 
+func (MyExecutor) HttpRequest(ctx context.Context, a *coreauth.Auth, req *http.Request) (*http.Response, error) {
+	if req == nil {
+		return nil, fmt.Errorf("myprov executor: request is nil")
+	}
+	if ctx == nil {
+		ctx = req.Context()
+	}
+	httpReq := req.WithContext(ctx)
+	if errPrep := (MyExecutor{}).PrepareRequest(httpReq, a); errPrep != nil {
+		return nil, errPrep
+	}
+	client := buildHTTPClient(a)
+	return client.Do(httpReq)
+}
+
 func (MyExecutor) CountTokens(context.Context, *coreauth.Auth, clipexec.Request, clipexec.Options) (clipexec.Response, error) {
 	return clipexec.Response{}, errors.New("count tokens not implemented")
 }
@@ -187,7 +205,7 @@ func main() {
 			// Optional: add a simple middleware + custom request logger
 			api.WithMiddleware(func(c *gin.Context) { c.Header("X-Example", "custom-provider"); c.Next() }),
 			api.WithRequestLoggerFactory(func(cfg *config.Config, cfgPath string) logging.RequestLogger {
-				return logging.NewFileRequestLogger(true, "logs", filepath.Dir(cfgPath))
+				return logging.NewFileRequestLoggerWithOptions(true, "logs", filepath.Dir(cfgPath), cfg.ErrorLogsMaxFiles)
 			}),
 		).
 		WithHooks(hooks).
@@ -199,8 +217,8 @@ func main() {
 	ctx, cancel := context.WithCancel(context.Background())
 	defer cancel()
 
-	if err := svc.Run(ctx); err != nil && !errors.Is(err, context.Canceled) {
-		panic(err)
+	if errRun := svc.Run(ctx); errRun != nil && !errors.Is(errRun, context.Canceled) {
+		panic(errRun)
 	}
 	_ = os.Stderr // keep os import used (demo only)
 	_ = time.Second

examples/http-request/main.go

@@ -0,0 +1,140 @@
+// Package main demonstrates how to use coreauth.Manager.HttpRequest/NewHttpRequest
+// to execute arbitrary HTTP requests with provider credentials injected.
+//
+// This example registers a minimal custom executor that injects an Authorization
+// header from auth.Attributes["api_key"], then performs two requests against
+// httpbin.org to show the injected headers.
+package main
+
+import (
+	"bytes"
+	"context"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"strings"
+	"time"
+
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+	clipexec "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+	log "github.com/sirupsen/logrus"
+)
+
+const providerKey = "echo"
+
+// EchoExecutor is a minimal provider implementation for demonstration purposes.
+type EchoExecutor struct{}
+
+func (EchoExecutor) Identifier() string { return providerKey }
+
+func (EchoExecutor) PrepareRequest(req *http.Request, auth *coreauth.Auth) error {
+	if req == nil || auth == nil {
+		return nil
+	}
+	if auth.Attributes != nil {
+		if apiKey := strings.TrimSpace(auth.Attributes["api_key"]); apiKey != "" {
+			req.Header.Set("Authorization", "Bearer "+apiKey)
+		}
+	}
+	return nil
+}
+
+func (EchoExecutor) HttpRequest(ctx context.Context, auth *coreauth.Auth, req *http.Request) (*http.Response, error) {
+	if req == nil {
+		return nil, fmt.Errorf("echo executor: request is nil")
+	}
+	if ctx == nil {
+		ctx = req.Context()
+	}
+	httpReq := req.WithContext(ctx)
+	if errPrep := (EchoExecutor{}).PrepareRequest(httpReq, auth); errPrep != nil {
+		return nil, errPrep
+	}
+	return http.DefaultClient.Do(httpReq)
+}
+
+func (EchoExecutor) Execute(context.Context, *coreauth.Auth, clipexec.Request, clipexec.Options) (clipexec.Response, error) {
+	return clipexec.Response{}, errors.New("echo executor: Execute not implemented")
+}
+
+func (EchoExecutor) ExecuteStream(context.Context, *coreauth.Auth, clipexec.Request, clipexec.Options) (<-chan clipexec.StreamChunk, error) {
+	return nil, errors.New("echo executor: ExecuteStream not implemented")
+}
+
+func (EchoExecutor) Refresh(context.Context, *coreauth.Auth) (*coreauth.Auth, error) {
+	return nil, errors.New("echo executor: Refresh not implemented")
+}
+
+func (EchoExecutor) CountTokens(context.Context, *coreauth.Auth, clipexec.Request, clipexec.Options) (clipexec.Response, error) {
+	return clipexec.Response{}, errors.New("echo executor: CountTokens not implemented")
+}
+
+func main() {
+	log.SetLevel(log.InfoLevel)
+
+	ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+	defer cancel()
+
+	core := coreauth.NewManager(nil, nil, nil)
+	core.RegisterExecutor(EchoExecutor{})
+
+	auth := &coreauth.Auth{
+		ID:       "demo-echo",
+		Provider: providerKey,
+		Attributes: map[string]string{
+			"api_key": "demo-api-key",
+		},
+	}
+
+	// Example 1: Build a prepared request and execute it using your own http.Client.
+	reqPrepared, errReqPrepared := core.NewHttpRequest(
+		ctx,
+		auth,
+		http.MethodGet,
+		"https://httpbin.org/anything",
+		nil,
+		http.Header{"X-Example": []string{"prepared"}},
+	)
+	if errReqPrepared != nil {
+		panic(errReqPrepared)
+	}
+	respPrepared, errDoPrepared := http.DefaultClient.Do(reqPrepared)
+	if errDoPrepared != nil {
+		panic(errDoPrepared)
+	}
+	defer func() {
+		if errClose := respPrepared.Body.Close(); errClose != nil {
+			log.Errorf("close response body error: %v", errClose)
+		}
+	}()
+	bodyPrepared, errReadPrepared := io.ReadAll(respPrepared.Body)
+	if errReadPrepared != nil {
+		panic(errReadPrepared)
+	}
+	fmt.Printf("Prepared request status: %d\n%s\n\n", respPrepared.StatusCode, bodyPrepared)
+
+	// Example 2: Execute a raw request via core.HttpRequest (auto inject + do).
+	rawBody := []byte(`{"hello":"world"}`)
+	rawReq, errRawReq := http.NewRequestWithContext(ctx, http.MethodPost, "https://httpbin.org/anything", bytes.NewReader(rawBody))
+	if errRawReq != nil {
+		panic(errRawReq)
+	}
+	rawReq.Header.Set("Content-Type", "application/json")
+	rawReq.Header.Set("X-Example", "executed")
+
+	respExec, errDoExec := core.HttpRequest(ctx, auth, rawReq)
+	if errDoExec != nil {
+		panic(errDoExec)
+	}
+	defer func() {
+		if errClose := respExec.Body.Close(); errClose != nil {
+			log.Errorf("close response body error: %v", errClose)
+		}
+	}()
+	bodyExec, errReadExec := io.ReadAll(respExec.Body)
+	if errReadExec != nil {
+		panic(errReadExec)
+	}
+	fmt.Printf("Manager HttpRequest status: %d\n%s\n", respExec.StatusCode, bodyExec)
+}

go.mod

@@ -13,13 +13,14 @@ require (
 	github.com/joho/godotenv v1.5.1
 	github.com/klauspost/compress v1.17.4
 	github.com/minio/minio-go/v7 v7.0.66
+	github.com/refraction-networking/utls v1.8.2
 	github.com/sirupsen/logrus v1.9.3
 	github.com/skratchdot/open-golang v0.0.0-20200116055534-eef842397966
 	github.com/tidwall/gjson v1.18.0
 	github.com/tidwall/sjson v1.2.5
 	github.com/tiktoken-go/tokenizer v0.7.0
-	golang.org/x/crypto v0.43.0
-	golang.org/x/net v0.46.0
+	golang.org/x/crypto v0.45.0
+	golang.org/x/net v0.47.0
 	golang.org/x/oauth2 v0.30.0
 	gopkg.in/natefinch/lumberjack.v2 v2.2.1
 	gopkg.in/yaml.v3 v3.0.1
@@ -68,9 +69,9 @@ require (
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	github.com/ugorji/go/codec v1.2.12 // indirect
 	golang.org/x/arch v0.8.0 // indirect
-	golang.org/x/sync v0.17.0 // indirect
-	golang.org/x/sys v0.37.0 // indirect
-	golang.org/x/text v0.30.0 // indirect
+	golang.org/x/sync v0.18.0 // indirect
+	golang.org/x/sys v0.38.0 // indirect
+	golang.org/x/text v0.31.0 // indirect
 	google.golang.org/protobuf v1.34.1 // indirect
 	gopkg.in/ini.v1 v1.67.0 // indirect
 )

go.sum

@@ -118,6 +118,8 @@ github.com/pjbgf/sha1cd v0.5.0 h1:a+UkboSi1znleCDUNT3M5YxjOnN1fz2FhN48FlwCxs0=
 github.com/pjbgf/sha1cd v0.5.0/go.mod h1:lhpGlyHLpQZoxMv8HcgXvZEhcGs0PG/vsZnEJ7H0iCM=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/refraction-networking/utls v1.8.2 h1:j4Q1gJj0xngdeH+Ox/qND11aEfhpgoEvV+S9iJ2IdQo=
+github.com/refraction-networking/utls v1.8.2/go.mod h1:jkSOEkLqn+S/jtpEHPOsVv/4V4EVnelwbMQl4vCWXAM=
 github.com/rogpeppe/go-internal v1.14.1 h1:UQB4HGPB6osV0SQTLymcB4TgvyWu6ZyliaW0tI/otEQ=
 github.com/rogpeppe/go-internal v1.14.1/go.mod h1:MaRKkUm5W0goXpeCfT7UZI6fk/L7L7so1lCWt35ZSgc=
 github.com/rs/xid v1.5.0 h1:mKX4bl4iPYJtEIxp6CYiUuLQ/8DYMoz0PUdtGgMFRVc=
@@ -160,22 +162,22 @@ github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZ
 golang.org/x/arch v0.0.0-20210923205945-b76863e36670/go.mod h1:5om86z9Hs0C8fWVUuoMHwpExlXzs5Tkyp9hOrfG7pp8=
 golang.org/x/arch v0.8.0 h1:3wRIsP3pM4yUptoR96otTUOXI367OS0+c9eeRi9doIc=
 golang.org/x/arch v0.8.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys=
-golang.org/x/crypto v0.43.0 h1:dduJYIi3A3KOfdGOHX8AVZ/jGiyPa3IbBozJ5kNuE04=
-golang.org/x/crypto v0.43.0/go.mod h1:BFbav4mRNlXJL4wNeejLpWxB7wMbc79PdRGhWKncxR0=
-golang.org/x/net v0.46.0 h1:giFlY12I07fugqwPuWJi68oOnpfqFnJIJzaIIm2JVV4=
-golang.org/x/net v0.46.0/go.mod h1:Q9BGdFy1y4nkUwiLvT5qtyhAnEHgnQ/zd8PfU6nc210=
+golang.org/x/crypto v0.45.0 h1:jMBrvKuj23MTlT0bQEOBcAE0mjg8mK9RXFhRH6nyF3Q=
+golang.org/x/crypto v0.45.0/go.mod h1:XTGrrkGJve7CYK7J8PEww4aY7gM3qMCElcJQ8n8JdX4=
+golang.org/x/net v0.47.0 h1:Mx+4dIFzqraBXUugkia1OOvlD6LemFo1ALMHjrXDOhY=
+golang.org/x/net v0.47.0/go.mod h1:/jNxtkgq5yWUGYkaZGqo27cfGZ1c5Nen03aYrrKpVRU=
 golang.org/x/oauth2 v0.30.0 h1:dnDm7JmhM45NNpd8FDDeLhK6FwqbOf4MLCM9zb1BOHI=
 golang.org/x/oauth2 v0.30.0/go.mod h1:B++QgG3ZKulg6sRPGD/mqlHQs5rB3Ml9erfeDY7xKlU=
-golang.org/x/sync v0.17.0 h1:l60nONMj9l5drqw6jlhIELNv9I0A4OFgRsG9k2oT9Ug=
-golang.org/x/sync v0.17.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
+golang.org/x/sync v0.18.0 h1:kr88TuHDroi+UVf+0hZnirlk8o8T+4MrK6mr60WkH/I=
+golang.org/x/sync v0.18.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
 golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.37.0 h1:fdNQudmxPjkdUTPnLn5mdQv7Zwvbvpaxqs831goi9kQ=
-golang.org/x/sys v0.37.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
-golang.org/x/term v0.36.0 h1:zMPR+aF8gfksFprF/Nc/rd1wRS1EI6nDBGyWAvDzx2Q=
-golang.org/x/term v0.36.0/go.mod h1:Qu394IJq6V6dCBRgwqshf3mPF85AqzYEzofzRdZkWss=
-golang.org/x/text v0.30.0 h1:yznKA/E9zq54KzlzBEAWn1NXSQ8DIp/NYMy88xJjl4k=
-golang.org/x/text v0.30.0/go.mod h1:yDdHFIX9t+tORqspjENWgzaCVXgk0yYnYuSZ8UzzBVM=
+golang.org/x/sys v0.38.0 h1:3yZWxaJjBmCWXqhN1qh02AkOnCQ1poK6oF+a7xWL6Gc=
+golang.org/x/sys v0.38.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/term v0.37.0 h1:8EGAD0qCmHYZg6J17DvsMy9/wJ7/D/4pV/wfnld5lTU=
+golang.org/x/term v0.37.0/go.mod h1:5pB4lxRNYYVZuTLmy8oR2BH8dflOR+IbTYFD8fi3254=
+golang.org/x/text v0.31.0 h1:aC8ghyu4JhP8VojJ2lEHBnochRno1sgL6nEi9WGFGMM=
+golang.org/x/text v0.31.0/go.mod h1:tKRAlv61yKIjGGHX/4tP1LTbc13YSec1pxVEWXzfoeM=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 google.golang.org/protobuf v1.34.1 h1:9ddQBjfCyZPOHPUiPxpYESBLc+T8P3E+Vo4IbKZgFWg=

internal/api/handlers/management/api_tools.go

@@ -0,0 +1,704 @@
+package management
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net"
+	"net/http"
+	"net/url"
+	"strings"
+	"time"
+
+	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/runtime/geminicli"
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/net/proxy"
+	"golang.org/x/oauth2"
+	"golang.org/x/oauth2/google"
+)
+
+const defaultAPICallTimeout = 60 * time.Second
+
+const (
+	geminiOAuthClientID     = "681255809395-oo8ft2oprdrnp9e3aqf6av3hmdib135j.apps.googleusercontent.com"
+	geminiOAuthClientSecret = "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"
+)
+
+var geminiOAuthScopes = []string{
+	"https://www.googleapis.com/auth/cloud-platform",
+	"https://www.googleapis.com/auth/userinfo.email",
+	"https://www.googleapis.com/auth/userinfo.profile",
+}
+
+const (
+	antigravityOAuthClientID     = "1071006060591-tmhssin2h21lcre235vtolojh4g403ep.apps.googleusercontent.com"
+	antigravityOAuthClientSecret = "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"
+)
+
+var antigravityOAuthTokenURL = "https://oauth2.googleapis.com/token"
+
+type apiCallRequest struct {
+	AuthIndexSnake  *string           `json:"auth_index"`
+	AuthIndexCamel  *string           `json:"authIndex"`
+	AuthIndexPascal *string           `json:"AuthIndex"`
+	Method          string            `json:"method"`
+	URL             string            `json:"url"`
+	Header          map[string]string `json:"header"`
+	Data            string            `json:"data"`
+}
+
+type apiCallResponse struct {
+	StatusCode int                 `json:"status_code"`
+	Header     map[string][]string `json:"header"`
+	Body       string              `json:"body"`
+}
+
+// APICall makes a generic HTTP request on behalf of the management API caller.
+// It is protected by the management middleware.
+//
+// Endpoint:
+//
+//	POST /v0/management/api-call
+//
+// Authentication:
+//
+//	Same as other management APIs (requires a management key and remote-management rules).
+//	You can provide the key via:
+//	- Authorization: Bearer <key>
+//	- X-Management-Key: <key>
+//
+// Request JSON:
+//   - auth_index / authIndex / AuthIndex (optional):
+//     The credential "auth_index" from GET /v0/management/auth-files (or other endpoints returning it).
+//     If omitted or not found, credential-specific proxy/token substitution is skipped.
+//   - method (required): HTTP method, e.g. GET, POST, PUT, PATCH, DELETE.
+//   - url (required): Absolute URL including scheme and host, e.g. "https://api.example.com/v1/ping".
+//   - header (optional): Request headers map.
+//     Supports magic variable "$TOKEN$" which is replaced using the selected credential:
+//     1) metadata.access_token
+//     2) attributes.api_key
+//     3) metadata.token / metadata.id_token / metadata.cookie
+//     Example: {"Authorization":"Bearer $TOKEN$"}.
+//     Note: if you need to override the HTTP Host header, set header["Host"].
+//   - data (optional): Raw request body as string (useful for POST/PUT/PATCH).
+//
+// Proxy selection (highest priority first):
+//  1. Selected credential proxy_url
+//  2. Global config proxy-url
+//  3. Direct connect (environment proxies are not used)
+//
+// Response JSON (returned with HTTP 200 when the APICall itself succeeds):
+//   - status_code: Upstream HTTP status code.
+//   - header: Upstream response headers.
+//   - body: Upstream response body as string.
+//
+// Example:
+//
+//	curl -sS -X POST "http://127.0.0.1:8317/v0/management/api-call" \
+//	  -H "Authorization: Bearer <MANAGEMENT_KEY>" \
+//	  -H "Content-Type: application/json" \
+//	  -d '{"auth_index":"<AUTH_INDEX>","method":"GET","url":"https://api.example.com/v1/ping","header":{"Authorization":"Bearer $TOKEN$"}}'
+//
+//	curl -sS -X POST "http://127.0.0.1:8317/v0/management/api-call" \
+//	  -H "Authorization: Bearer 831227" \
+//	  -H "Content-Type: application/json" \
+//	  -d '{"auth_index":"<AUTH_INDEX>","method":"POST","url":"https://api.example.com/v1/fetchAvailableModels","header":{"Authorization":"Bearer $TOKEN$","Content-Type":"application/json","User-Agent":"cliproxyapi"},"data":"{}"}'
+func (h *Handler) APICall(c *gin.Context) {
+	var body apiCallRequest
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid body"})
+		return
+	}
+
+	method := strings.ToUpper(strings.TrimSpace(body.Method))
+	if method == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "missing method"})
+		return
+	}
+
+	urlStr := strings.TrimSpace(body.URL)
+	if urlStr == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "missing url"})
+		return
+	}
+	parsedURL, errParseURL := url.Parse(urlStr)
+	if errParseURL != nil || parsedURL.Scheme == "" || parsedURL.Host == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid url"})
+		return
+	}
+
+	authIndex := firstNonEmptyString(body.AuthIndexSnake, body.AuthIndexCamel, body.AuthIndexPascal)
+	auth := h.authByIndex(authIndex)
+
+	reqHeaders := body.Header
+	if reqHeaders == nil {
+		reqHeaders = map[string]string{}
+	}
+
+	var hostOverride string
+	var token string
+	var tokenResolved bool
+	var tokenErr error
+	for key, value := range reqHeaders {
+		if !strings.Contains(value, "$TOKEN$") {
+			continue
+		}
+		if !tokenResolved {
+			token, tokenErr = h.resolveTokenForAuth(c.Request.Context(), auth)
+			tokenResolved = true
+		}
+		if auth != nil && token == "" {
+			if tokenErr != nil {
+				c.JSON(http.StatusBadRequest, gin.H{"error": "auth token refresh failed"})
+				return
+			}
+			c.JSON(http.StatusBadRequest, gin.H{"error": "auth token not found"})
+			return
+		}
+		if token == "" {
+			continue
+		}
+		reqHeaders[key] = strings.ReplaceAll(value, "$TOKEN$", token)
+	}
+
+	var requestBody io.Reader
+	if body.Data != "" {
+		requestBody = strings.NewReader(body.Data)
+	}
+
+	req, errNewRequest := http.NewRequestWithContext(c.Request.Context(), method, urlStr, requestBody)
+	if errNewRequest != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "failed to build request"})
+		return
+	}
+
+	for key, value := range reqHeaders {
+		if strings.EqualFold(key, "host") {
+			hostOverride = strings.TrimSpace(value)
+			continue
+		}
+		req.Header.Set(key, value)
+	}
+	if hostOverride != "" {
+		req.Host = hostOverride
+	}
+
+	httpClient := &http.Client{
+		Timeout: defaultAPICallTimeout,
+	}
+	httpClient.Transport = h.apiCallTransport(auth)
+
+	resp, errDo := httpClient.Do(req)
+	if errDo != nil {
+		log.WithError(errDo).Debug("management APICall request failed")
+		c.JSON(http.StatusBadGateway, gin.H{"error": "request failed"})
+		return
+	}
+	defer func() {
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("response body close error: %v", errClose)
+		}
+	}()
+
+	respBody, errReadAll := io.ReadAll(resp.Body)
+	if errReadAll != nil {
+		c.JSON(http.StatusBadGateway, gin.H{"error": "failed to read response"})
+		return
+	}
+
+	c.JSON(http.StatusOK, apiCallResponse{
+		StatusCode: resp.StatusCode,
+		Header:     resp.Header,
+		Body:       string(respBody),
+	})
+}
+
+func firstNonEmptyString(values ...*string) string {
+	for _, v := range values {
+		if v == nil {
+			continue
+		}
+		if out := strings.TrimSpace(*v); out != "" {
+			return out
+		}
+	}
+	return ""
+}
+
+func tokenValueForAuth(auth *coreauth.Auth) string {
+	if auth == nil {
+		return ""
+	}
+	if v := tokenValueFromMetadata(auth.Metadata); v != "" {
+		return v
+	}
+	if auth.Attributes != nil {
+		if v := strings.TrimSpace(auth.Attributes["api_key"]); v != "" {
+			return v
+		}
+	}
+	if shared := geminicli.ResolveSharedCredential(auth.Runtime); shared != nil {
+		if v := tokenValueFromMetadata(shared.MetadataSnapshot()); v != "" {
+			return v
+		}
+	}
+	return ""
+}
+
+func (h *Handler) resolveTokenForAuth(ctx context.Context, auth *coreauth.Auth) (string, error) {
+	if auth == nil {
+		return "", nil
+	}
+
+	provider := strings.ToLower(strings.TrimSpace(auth.Provider))
+	if provider == "gemini-cli" {
+		token, errToken := h.refreshGeminiOAuthAccessToken(ctx, auth)
+		return token, errToken
+	}
+	if provider == "antigravity" {
+		token, errToken := h.refreshAntigravityOAuthAccessToken(ctx, auth)
+		return token, errToken
+	}
+
+	return tokenValueForAuth(auth), nil
+}
+
+func (h *Handler) refreshGeminiOAuthAccessToken(ctx context.Context, auth *coreauth.Auth) (string, error) {
+	if ctx == nil {
+		ctx = context.Background()
+	}
+	if auth == nil {
+		return "", nil
+	}
+
+	metadata, updater := geminiOAuthMetadata(auth)
+	if len(metadata) == 0 {
+		return "", fmt.Errorf("gemini oauth metadata missing")
+	}
+
+	base := make(map[string]any)
+	if tokenRaw, ok := metadata["token"].(map[string]any); ok && tokenRaw != nil {
+		base = cloneMap(tokenRaw)
+	}
+
+	var token oauth2.Token
+	if len(base) > 0 {
+		if raw, errMarshal := json.Marshal(base); errMarshal == nil {
+			_ = json.Unmarshal(raw, &token)
+		}
+	}
+
+	if token.AccessToken == "" {
+		token.AccessToken = stringValue(metadata, "access_token")
+	}
+	if token.RefreshToken == "" {
+		token.RefreshToken = stringValue(metadata, "refresh_token")
+	}
+	if token.TokenType == "" {
+		token.TokenType = stringValue(metadata, "token_type")
+	}
+	if token.Expiry.IsZero() {
+		if expiry := stringValue(metadata, "expiry"); expiry != "" {
+			if ts, errParseTime := time.Parse(time.RFC3339, expiry); errParseTime == nil {
+				token.Expiry = ts
+			}
+		}
+	}
+
+	conf := &oauth2.Config{
+		ClientID:     geminiOAuthClientID,
+		ClientSecret: geminiOAuthClientSecret,
+		Scopes:       geminiOAuthScopes,
+		Endpoint:     google.Endpoint,
+	}
+
+	ctxToken := ctx
+	httpClient := &http.Client{
+		Timeout:   defaultAPICallTimeout,
+		Transport: h.apiCallTransport(auth),
+	}
+	ctxToken = context.WithValue(ctxToken, oauth2.HTTPClient, httpClient)
+
+	src := conf.TokenSource(ctxToken, &token)
+	currentToken, errToken := src.Token()
+	if errToken != nil {
+		return "", errToken
+	}
+
+	merged := buildOAuthTokenMap(base, currentToken)
+	fields := buildOAuthTokenFields(currentToken, merged)
+	if updater != nil {
+		updater(fields)
+	}
+	return strings.TrimSpace(currentToken.AccessToken), nil
+}
+
+func (h *Handler) refreshAntigravityOAuthAccessToken(ctx context.Context, auth *coreauth.Auth) (string, error) {
+	if ctx == nil {
+		ctx = context.Background()
+	}
+	if auth == nil {
+		return "", nil
+	}
+
+	metadata := auth.Metadata
+	if len(metadata) == 0 {
+		return "", fmt.Errorf("antigravity oauth metadata missing")
+	}
+
+	current := strings.TrimSpace(tokenValueFromMetadata(metadata))
+	if current != "" && !antigravityTokenNeedsRefresh(metadata) {
+		return current, nil
+	}
+
+	refreshToken := stringValue(metadata, "refresh_token")
+	if refreshToken == "" {
+		return "", fmt.Errorf("antigravity refresh token missing")
+	}
+
+	tokenURL := strings.TrimSpace(antigravityOAuthTokenURL)
+	if tokenURL == "" {
+		tokenURL = "https://oauth2.googleapis.com/token"
+	}
+	form := url.Values{}
+	form.Set("client_id", antigravityOAuthClientID)
+	form.Set("client_secret", antigravityOAuthClientSecret)
+	form.Set("grant_type", "refresh_token")
+	form.Set("refresh_token", refreshToken)
+
+	req, errReq := http.NewRequestWithContext(ctx, http.MethodPost, tokenURL, strings.NewReader(form.Encode()))
+	if errReq != nil {
+		return "", errReq
+	}
+	req.Header.Set("Content-Type", "application/x-www-form-urlencoded")
+
+	httpClient := &http.Client{
+		Timeout:   defaultAPICallTimeout,
+		Transport: h.apiCallTransport(auth),
+	}
+	resp, errDo := httpClient.Do(req)
+	if errDo != nil {
+		return "", errDo
+	}
+	defer func() {
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("response body close error: %v", errClose)
+		}
+	}()
+
+	bodyBytes, errRead := io.ReadAll(resp.Body)
+	if errRead != nil {
+		return "", errRead
+	}
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
+		return "", fmt.Errorf("antigravity oauth token refresh failed: status %d: %s", resp.StatusCode, strings.TrimSpace(string(bodyBytes)))
+	}
+
+	var tokenResp struct {
+		AccessToken  string `json:"access_token"`
+		RefreshToken string `json:"refresh_token"`
+		ExpiresIn    int64  `json:"expires_in"`
+		TokenType    string `json:"token_type"`
+	}
+	if errUnmarshal := json.Unmarshal(bodyBytes, &tokenResp); errUnmarshal != nil {
+		return "", errUnmarshal
+	}
+
+	if strings.TrimSpace(tokenResp.AccessToken) == "" {
+		return "", fmt.Errorf("antigravity oauth token refresh returned empty access_token")
+	}
+
+	if auth.Metadata == nil {
+		auth.Metadata = make(map[string]any)
+	}
+	now := time.Now()
+	auth.Metadata["access_token"] = strings.TrimSpace(tokenResp.AccessToken)
+	if strings.TrimSpace(tokenResp.RefreshToken) != "" {
+		auth.Metadata["refresh_token"] = strings.TrimSpace(tokenResp.RefreshToken)
+	}
+	if tokenResp.ExpiresIn > 0 {
+		auth.Metadata["expires_in"] = tokenResp.ExpiresIn
+		auth.Metadata["timestamp"] = now.UnixMilli()
+		auth.Metadata["expired"] = now.Add(time.Duration(tokenResp.ExpiresIn) * time.Second).Format(time.RFC3339)
+	}
+	auth.Metadata["type"] = "antigravity"
+
+	if h != nil && h.authManager != nil {
+		auth.LastRefreshedAt = now
+		auth.UpdatedAt = now
+		_, _ = h.authManager.Update(ctx, auth)
+	}
+
+	return strings.TrimSpace(tokenResp.AccessToken), nil
+}
+
+func antigravityTokenNeedsRefresh(metadata map[string]any) bool {
+	// Refresh a bit early to avoid requests racing token expiry.
+	const skew = 30 * time.Second
+
+	if metadata == nil {
+		return true
+	}
+	if expStr, ok := metadata["expired"].(string); ok {
+		if ts, errParse := time.Parse(time.RFC3339, strings.TrimSpace(expStr)); errParse == nil {
+			return !ts.After(time.Now().Add(skew))
+		}
+	}
+	expiresIn := int64Value(metadata["expires_in"])
+	timestampMs := int64Value(metadata["timestamp"])
+	if expiresIn > 0 && timestampMs > 0 {
+		exp := time.UnixMilli(timestampMs).Add(time.Duration(expiresIn) * time.Second)
+		return !exp.After(time.Now().Add(skew))
+	}
+	return true
+}
+
+func int64Value(raw any) int64 {
+	switch typed := raw.(type) {
+	case int:
+		return int64(typed)
+	case int32:
+		return int64(typed)
+	case int64:
+		return typed
+	case uint:
+		return int64(typed)
+	case uint32:
+		return int64(typed)
+	case uint64:
+		if typed > uint64(^uint64(0)>>1) {
+			return 0
+		}
+		return int64(typed)
+	case float32:
+		return int64(typed)
+	case float64:
+		return int64(typed)
+	case json.Number:
+		if i, errParse := typed.Int64(); errParse == nil {
+			return i
+		}
+	case string:
+		if s := strings.TrimSpace(typed); s != "" {
+			if i, errParse := json.Number(s).Int64(); errParse == nil {
+				return i
+			}
+		}
+	}
+	return 0
+}
+
+func geminiOAuthMetadata(auth *coreauth.Auth) (map[string]any, func(map[string]any)) {
+	if auth == nil {
+		return nil, nil
+	}
+	if shared := geminicli.ResolveSharedCredential(auth.Runtime); shared != nil {
+		snapshot := shared.MetadataSnapshot()
+		return snapshot, func(fields map[string]any) { shared.MergeMetadata(fields) }
+	}
+	return auth.Metadata, func(fields map[string]any) {
+		if auth.Metadata == nil {
+			auth.Metadata = make(map[string]any)
+		}
+		for k, v := range fields {
+			auth.Metadata[k] = v
+		}
+	}
+}
+
+func stringValue(metadata map[string]any, key string) string {
+	if len(metadata) == 0 || key == "" {
+		return ""
+	}
+	if v, ok := metadata[key].(string); ok {
+		return strings.TrimSpace(v)
+	}
+	return ""
+}
+
+func cloneMap(in map[string]any) map[string]any {
+	if len(in) == 0 {
+		return nil
+	}
+	out := make(map[string]any, len(in))
+	for k, v := range in {
+		out[k] = v
+	}
+	return out
+}
+
+func buildOAuthTokenMap(base map[string]any, tok *oauth2.Token) map[string]any {
+	merged := cloneMap(base)
+	if merged == nil {
+		merged = make(map[string]any)
+	}
+	if tok == nil {
+		return merged
+	}
+	if raw, errMarshal := json.Marshal(tok); errMarshal == nil {
+		var tokenMap map[string]any
+		if errUnmarshal := json.Unmarshal(raw, &tokenMap); errUnmarshal == nil {
+			for k, v := range tokenMap {
+				merged[k] = v
+			}
+		}
+	}
+	return merged
+}
+
+func buildOAuthTokenFields(tok *oauth2.Token, merged map[string]any) map[string]any {
+	fields := make(map[string]any, 5)
+	if tok != nil && tok.AccessToken != "" {
+		fields["access_token"] = tok.AccessToken
+	}
+	if tok != nil && tok.TokenType != "" {
+		fields["token_type"] = tok.TokenType
+	}
+	if tok != nil && tok.RefreshToken != "" {
+		fields["refresh_token"] = tok.RefreshToken
+	}
+	if tok != nil && !tok.Expiry.IsZero() {
+		fields["expiry"] = tok.Expiry.Format(time.RFC3339)
+	}
+	if len(merged) > 0 {
+		fields["token"] = cloneMap(merged)
+	}
+	return fields
+}
+
+func tokenValueFromMetadata(metadata map[string]any) string {
+	if len(metadata) == 0 {
+		return ""
+	}
+	if v, ok := metadata["accessToken"].(string); ok && strings.TrimSpace(v) != "" {
+		return strings.TrimSpace(v)
+	}
+	if v, ok := metadata["access_token"].(string); ok && strings.TrimSpace(v) != "" {
+		return strings.TrimSpace(v)
+	}
+	if tokenRaw, ok := metadata["token"]; ok && tokenRaw != nil {
+		switch typed := tokenRaw.(type) {
+		case string:
+			if v := strings.TrimSpace(typed); v != "" {
+				return v
+			}
+		case map[string]any:
+			if v, ok := typed["access_token"].(string); ok && strings.TrimSpace(v) != "" {
+				return strings.TrimSpace(v)
+			}
+			if v, ok := typed["accessToken"].(string); ok && strings.TrimSpace(v) != "" {
+				return strings.TrimSpace(v)
+			}
+		case map[string]string:
+			if v := strings.TrimSpace(typed["access_token"]); v != "" {
+				return v
+			}
+			if v := strings.TrimSpace(typed["accessToken"]); v != "" {
+				return v
+			}
+		}
+	}
+	if v, ok := metadata["token"].(string); ok && strings.TrimSpace(v) != "" {
+		return strings.TrimSpace(v)
+	}
+	if v, ok := metadata["id_token"].(string); ok && strings.TrimSpace(v) != "" {
+		return strings.TrimSpace(v)
+	}
+	if v, ok := metadata["cookie"].(string); ok && strings.TrimSpace(v) != "" {
+		return strings.TrimSpace(v)
+	}
+	return ""
+}
+
+func (h *Handler) authByIndex(authIndex string) *coreauth.Auth {
+	authIndex = strings.TrimSpace(authIndex)
+	if authIndex == "" || h == nil || h.authManager == nil {
+		return nil
+	}
+	auths := h.authManager.List()
+	for _, auth := range auths {
+		if auth == nil {
+			continue
+		}
+		auth.EnsureIndex()
+		if auth.Index == authIndex {
+			return auth
+		}
+	}
+	return nil
+}
+
+func (h *Handler) apiCallTransport(auth *coreauth.Auth) http.RoundTripper {
+	var proxyCandidates []string
+	if auth != nil {
+		if proxyStr := strings.TrimSpace(auth.ProxyURL); proxyStr != "" {
+			proxyCandidates = append(proxyCandidates, proxyStr)
+		}
+	}
+	if h != nil && h.cfg != nil {
+		if proxyStr := strings.TrimSpace(h.cfg.ProxyURL); proxyStr != "" {
+			proxyCandidates = append(proxyCandidates, proxyStr)
+		}
+	}
+
+	for _, proxyStr := range proxyCandidates {
+		if transport := buildProxyTransport(proxyStr); transport != nil {
+			return transport
+		}
+	}
+
+	transport, ok := http.DefaultTransport.(*http.Transport)
+	if !ok || transport == nil {
+		return &http.Transport{Proxy: nil}
+	}
+	clone := transport.Clone()
+	clone.Proxy = nil
+	return clone
+}
+
+func buildProxyTransport(proxyStr string) *http.Transport {
+	proxyStr = strings.TrimSpace(proxyStr)
+	if proxyStr == "" {
+		return nil
+	}
+
+	proxyURL, errParse := url.Parse(proxyStr)
+	if errParse != nil {
+		log.WithError(errParse).Debug("parse proxy URL failed")
+		return nil
+	}
+	if proxyURL.Scheme == "" || proxyURL.Host == "" {
+		log.Debug("proxy URL missing scheme/host")
+		return nil
+	}
+
+	if proxyURL.Scheme == "socks5" {
+		var proxyAuth *proxy.Auth
+		if proxyURL.User != nil {
+			username := proxyURL.User.Username()
+			password, _ := proxyURL.User.Password()
+			proxyAuth = &proxy.Auth{User: username, Password: password}
+		}
+		dialer, errSOCKS5 := proxy.SOCKS5("tcp", proxyURL.Host, proxyAuth, proxy.Direct)
+		if errSOCKS5 != nil {
+			log.WithError(errSOCKS5).Debug("create SOCKS5 dialer failed")
+			return nil
+		}
+		return &http.Transport{
+			Proxy: nil,
+			DialContext: func(ctx context.Context, network, addr string) (net.Conn, error) {
+				return dialer.Dial(network, addr)
+			},
+		}
+	}
+
+	if proxyURL.Scheme == "http" || proxyURL.Scheme == "https" {
+		return &http.Transport{Proxy: http.ProxyURL(proxyURL)}
+	}
+
+	log.Debugf("unsupported proxy scheme: %s", proxyURL.Scheme)
+	return nil
+}

internal/api/handlers/management/api_tools_test.go

@@ -0,0 +1,173 @@
+package management
+
+import (
+	"context"
+	"encoding/json"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"net/url"
+	"strings"
+	"sync"
+	"testing"
+	"time"
+
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+)
+
+type memoryAuthStore struct {
+	mu    sync.Mutex
+	items map[string]*coreauth.Auth
+}
+
+func (s *memoryAuthStore) List(ctx context.Context) ([]*coreauth.Auth, error) {
+	_ = ctx
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	out := make([]*coreauth.Auth, 0, len(s.items))
+	for _, a := range s.items {
+		out = append(out, a.Clone())
+	}
+	return out, nil
+}
+
+func (s *memoryAuthStore) Save(ctx context.Context, auth *coreauth.Auth) (string, error) {
+	_ = ctx
+	if auth == nil {
+		return "", nil
+	}
+	s.mu.Lock()
+	if s.items == nil {
+		s.items = make(map[string]*coreauth.Auth)
+	}
+	s.items[auth.ID] = auth.Clone()
+	s.mu.Unlock()
+	return auth.ID, nil
+}
+
+func (s *memoryAuthStore) Delete(ctx context.Context, id string) error {
+	_ = ctx
+	s.mu.Lock()
+	delete(s.items, id)
+	s.mu.Unlock()
+	return nil
+}
+
+func TestResolveTokenForAuth_Antigravity_RefreshesExpiredToken(t *testing.T) {
+	var callCount int
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		callCount++
+		if r.Method != http.MethodPost {
+			t.Fatalf("expected POST, got %s", r.Method)
+		}
+		if ct := r.Header.Get("Content-Type"); !strings.HasPrefix(ct, "application/x-www-form-urlencoded") {
+			t.Fatalf("unexpected content-type: %s", ct)
+		}
+		bodyBytes, _ := io.ReadAll(r.Body)
+		_ = r.Body.Close()
+		values, err := url.ParseQuery(string(bodyBytes))
+		if err != nil {
+			t.Fatalf("parse form: %v", err)
+		}
+		if values.Get("grant_type") != "refresh_token" {
+			t.Fatalf("unexpected grant_type: %s", values.Get("grant_type"))
+		}
+		if values.Get("refresh_token") != "rt" {
+			t.Fatalf("unexpected refresh_token: %s", values.Get("refresh_token"))
+		}
+		if values.Get("client_id") != antigravityOAuthClientID {
+			t.Fatalf("unexpected client_id: %s", values.Get("client_id"))
+		}
+		if values.Get("client_secret") != antigravityOAuthClientSecret {
+			t.Fatalf("unexpected client_secret")
+		}
+
+		w.Header().Set("Content-Type", "application/json")
+		_ = json.NewEncoder(w).Encode(map[string]any{
+			"access_token":  "new-token",
+			"refresh_token": "rt2",
+			"expires_in":    int64(3600),
+			"token_type":    "Bearer",
+		})
+	}))
+	t.Cleanup(srv.Close)
+
+	originalURL := antigravityOAuthTokenURL
+	antigravityOAuthTokenURL = srv.URL
+	t.Cleanup(func() { antigravityOAuthTokenURL = originalURL })
+
+	store := &memoryAuthStore{}
+	manager := coreauth.NewManager(store, nil, nil)
+
+	auth := &coreauth.Auth{
+		ID:       "antigravity-test.json",
+		FileName: "antigravity-test.json",
+		Provider: "antigravity",
+		Metadata: map[string]any{
+			"type":          "antigravity",
+			"access_token":  "old-token",
+			"refresh_token": "rt",
+			"expires_in":    int64(3600),
+			"timestamp":     time.Now().Add(-2 * time.Hour).UnixMilli(),
+			"expired":       time.Now().Add(-1 * time.Hour).Format(time.RFC3339),
+		},
+	}
+	if _, err := manager.Register(context.Background(), auth); err != nil {
+		t.Fatalf("register auth: %v", err)
+	}
+
+	h := &Handler{authManager: manager}
+	token, err := h.resolveTokenForAuth(context.Background(), auth)
+	if err != nil {
+		t.Fatalf("resolveTokenForAuth: %v", err)
+	}
+	if token != "new-token" {
+		t.Fatalf("expected refreshed token, got %q", token)
+	}
+	if callCount != 1 {
+		t.Fatalf("expected 1 refresh call, got %d", callCount)
+	}
+
+	updated, ok := manager.GetByID(auth.ID)
+	if !ok || updated == nil {
+		t.Fatalf("expected auth in manager after update")
+	}
+	if got := tokenValueFromMetadata(updated.Metadata); got != "new-token" {
+		t.Fatalf("expected manager metadata updated, got %q", got)
+	}
+}
+
+func TestResolveTokenForAuth_Antigravity_SkipsRefreshWhenTokenValid(t *testing.T) {
+	var callCount int
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		callCount++
+		w.WriteHeader(http.StatusInternalServerError)
+	}))
+	t.Cleanup(srv.Close)
+
+	originalURL := antigravityOAuthTokenURL
+	antigravityOAuthTokenURL = srv.URL
+	t.Cleanup(func() { antigravityOAuthTokenURL = originalURL })
+
+	auth := &coreauth.Auth{
+		ID:       "antigravity-valid.json",
+		FileName: "antigravity-valid.json",
+		Provider: "antigravity",
+		Metadata: map[string]any{
+			"type":         "antigravity",
+			"access_token": "ok-token",
+			"expired":      time.Now().Add(30 * time.Minute).Format(time.RFC3339),
+		},
+	}
+	h := &Handler{}
+	token, err := h.resolveTokenForAuth(context.Background(), auth)
+	if err != nil {
+		t.Fatalf("resolveTokenForAuth: %v", err)
+	}
+	if token != "ok-token" {
+		t.Fatalf("expected existing token, got %q", token)
+	}
+	if callCount != 0 {
+		t.Fatalf("expected no refresh calls, got %d", callCount)
+	}
+}

internal/api/handlers/management/config_basic.go

@@ -202,6 +202,46 @@ func (h *Handler) PutLoggingToFile(c *gin.Context) {
 	h.updateBoolField(c, func(v bool) { h.cfg.LoggingToFile = v })
 }
 
+// LogsMaxTotalSizeMB
+func (h *Handler) GetLogsMaxTotalSizeMB(c *gin.Context) {
+	c.JSON(200, gin.H{"logs-max-total-size-mb": h.cfg.LogsMaxTotalSizeMB})
+}
+func (h *Handler) PutLogsMaxTotalSizeMB(c *gin.Context) {
+	var body struct {
+		Value *int `json:"value"`
+	}
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil || body.Value == nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid body"})
+		return
+	}
+	value := *body.Value
+	if value < 0 {
+		value = 0
+	}
+	h.cfg.LogsMaxTotalSizeMB = value
+	h.persist(c)
+}
+
+// ErrorLogsMaxFiles
+func (h *Handler) GetErrorLogsMaxFiles(c *gin.Context) {
+	c.JSON(200, gin.H{"error-logs-max-files": h.cfg.ErrorLogsMaxFiles})
+}
+func (h *Handler) PutErrorLogsMaxFiles(c *gin.Context) {
+	var body struct {
+		Value *int `json:"value"`
+	}
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil || body.Value == nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid body"})
+		return
+	}
+	value := *body.Value
+	if value < 0 {
+		value = 10
+	}
+	h.cfg.ErrorLogsMaxFiles = value
+	h.persist(c)
+}
+
 // Request log
 func (h *Handler) GetRequestLog(c *gin.Context) { c.JSON(200, gin.H{"request-log": h.cfg.RequestLog}) }
 func (h *Handler) PutRequestLog(c *gin.Context) {
@@ -232,6 +272,52 @@ func (h *Handler) PutMaxRetryInterval(c *gin.Context) {
 	h.updateIntField(c, func(v int) { h.cfg.MaxRetryInterval = v })
 }
 
+// ForceModelPrefix
+func (h *Handler) GetForceModelPrefix(c *gin.Context) {
+	c.JSON(200, gin.H{"force-model-prefix": h.cfg.ForceModelPrefix})
+}
+func (h *Handler) PutForceModelPrefix(c *gin.Context) {
+	h.updateBoolField(c, func(v bool) { h.cfg.ForceModelPrefix = v })
+}
+
+func normalizeRoutingStrategy(strategy string) (string, bool) {
+	normalized := strings.ToLower(strings.TrimSpace(strategy))
+	switch normalized {
+	case "", "round-robin", "roundrobin", "rr":
+		return "round-robin", true
+	case "fill-first", "fillfirst", "ff":
+		return "fill-first", true
+	default:
+		return "", false
+	}
+}
+
+// RoutingStrategy
+func (h *Handler) GetRoutingStrategy(c *gin.Context) {
+	strategy, ok := normalizeRoutingStrategy(h.cfg.Routing.Strategy)
+	if !ok {
+		c.JSON(200, gin.H{"strategy": strings.TrimSpace(h.cfg.Routing.Strategy)})
+		return
+	}
+	c.JSON(200, gin.H{"strategy": strategy})
+}
+func (h *Handler) PutRoutingStrategy(c *gin.Context) {
+	var body struct {
+		Value *string `json:"value"`
+	}
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil || body.Value == nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid body"})
+		return
+	}
+	normalized, ok := normalizeRoutingStrategy(*body.Value)
+	if !ok {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid strategy"})
+		return
+	}
+	h.cfg.Routing.Strategy = normalized
+	h.persist(c)
+}
+
 // Proxy URL
 func (h *Handler) GetProxyURL(c *gin.Context) { c.JSON(200, gin.H{"proxy-url": h.cfg.ProxyURL}) }
 func (h *Handler) PutProxyURL(c *gin.Context) {

internal/api/handlers/management/config_lists.go

@@ -145,71 +145,74 @@ func (h *Handler) PutGeminiKeys(c *gin.Context) {
 	h.persist(c)
 }
 func (h *Handler) PatchGeminiKey(c *gin.Context) {
+	type geminiKeyPatch struct {
+		APIKey         *string            `json:"api-key"`
+		Prefix         *string            `json:"prefix"`
+		BaseURL        *string            `json:"base-url"`
+		ProxyURL       *string            `json:"proxy-url"`
+		Headers        *map[string]string `json:"headers"`
+		ExcludedModels *[]string          `json:"excluded-models"`
+	}
 	var body struct {
 		Index *int            `json:"index"`
 		Match *string         `json:"match"`
-		Value *config.GeminiKey `json:"value"`
+		Value *geminiKeyPatch `json:"value"`
 	}
 	if err := c.ShouldBindJSON(&body); err != nil || body.Value == nil {
 		c.JSON(400, gin.H{"error": "invalid body"})
 		return
 	}
-	value := *body.Value
-	value.APIKey = strings.TrimSpace(value.APIKey)
-	value.BaseURL = strings.TrimSpace(value.BaseURL)
-	value.ProxyURL = strings.TrimSpace(value.ProxyURL)
-	value.ExcludedModels = config.NormalizeExcludedModels(value.ExcludedModels)
-	if value.APIKey == "" {
-		// Treat empty API key as delete.
+	targetIndex := -1
 	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.GeminiKey) {
-			h.cfg.GeminiKey = append(h.cfg.GeminiKey[:*body.Index], h.cfg.GeminiKey[*body.Index+1:]...)
-			h.cfg.SanitizeGeminiKeys()
-			h.persist(c)
-			return
+		targetIndex = *body.Index
 	}
-		if body.Match != nil {
+	if targetIndex == -1 && body.Match != nil {
 		match := strings.TrimSpace(*body.Match)
 		if match != "" {
-				out := make([]config.GeminiKey, 0, len(h.cfg.GeminiKey))
-				removed := false
 			for i := range h.cfg.GeminiKey {
-					if !removed && h.cfg.GeminiKey[i].APIKey == match {
-						removed = true
-						continue
-					}
-					out = append(out, h.cfg.GeminiKey[i])
-				}
-				if removed {
-					h.cfg.GeminiKey = out
-					h.cfg.SanitizeGeminiKeys()
-					h.persist(c)
-					return
+				if h.cfg.GeminiKey[i].APIKey == match {
+					targetIndex = i
+					break
 				}
 			}
 		}
+	}
+	if targetIndex == -1 {
 		c.JSON(404, gin.H{"error": "item not found"})
 		return
 	}
 
-	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.GeminiKey) {
-		h.cfg.GeminiKey[*body.Index] = value
+	entry := h.cfg.GeminiKey[targetIndex]
+	if body.Value.APIKey != nil {
+		trimmed := strings.TrimSpace(*body.Value.APIKey)
+		if trimmed == "" {
+			h.cfg.GeminiKey = append(h.cfg.GeminiKey[:targetIndex], h.cfg.GeminiKey[targetIndex+1:]...)
 			h.cfg.SanitizeGeminiKeys()
 			h.persist(c)
 			return
 		}
-	if body.Match != nil {
-		match := strings.TrimSpace(*body.Match)
-		for i := range h.cfg.GeminiKey {
-			if h.cfg.GeminiKey[i].APIKey == match {
-				h.cfg.GeminiKey[i] = value
+		entry.APIKey = trimmed
+	}
+	if body.Value.Prefix != nil {
+		entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
+	}
+	if body.Value.BaseURL != nil {
+		entry.BaseURL = strings.TrimSpace(*body.Value.BaseURL)
+	}
+	if body.Value.ProxyURL != nil {
+		entry.ProxyURL = strings.TrimSpace(*body.Value.ProxyURL)
+	}
+	if body.Value.Headers != nil {
+		entry.Headers = config.NormalizeHeaders(*body.Value.Headers)
+	}
+	if body.Value.ExcludedModels != nil {
+		entry.ExcludedModels = config.NormalizeExcludedModels(*body.Value.ExcludedModels)
+	}
+	h.cfg.GeminiKey[targetIndex] = entry
 	h.cfg.SanitizeGeminiKeys()
 	h.persist(c)
-				return
-			}
-		}
-	}
-	c.JSON(404, gin.H{"error": "item not found"})
 }
+
 func (h *Handler) DeleteGeminiKey(c *gin.Context) {
 	if val := strings.TrimSpace(c.Query("api-key")); val != "" {
 		out := make([]config.GeminiKey, 0, len(h.cfg.GeminiKey))
@@ -268,35 +271,70 @@ func (h *Handler) PutClaudeKeys(c *gin.Context) {
 	h.persist(c)
 }
 func (h *Handler) PatchClaudeKey(c *gin.Context) {
+	type claudeKeyPatch struct {
+		APIKey         *string               `json:"api-key"`
+		Prefix         *string               `json:"prefix"`
+		BaseURL        *string               `json:"base-url"`
+		ProxyURL       *string               `json:"proxy-url"`
+		Models         *[]config.ClaudeModel `json:"models"`
+		Headers        *map[string]string    `json:"headers"`
+		ExcludedModels *[]string             `json:"excluded-models"`
+	}
 	var body struct {
 		Index *int            `json:"index"`
 		Match *string         `json:"match"`
-		Value *config.ClaudeKey `json:"value"`
+		Value *claudeKeyPatch `json:"value"`
 	}
 	if err := c.ShouldBindJSON(&body); err != nil || body.Value == nil {
 		c.JSON(400, gin.H{"error": "invalid body"})
 		return
 	}
-	value := *body.Value
-	normalizeClaudeKey(&value)
+	targetIndex := -1
 	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.ClaudeKey) {
-		h.cfg.ClaudeKey[*body.Index] = value
-		h.cfg.SanitizeClaudeKeys()
-		h.persist(c)
-		return
+		targetIndex = *body.Index
 	}
-	if body.Match != nil {
+	if targetIndex == -1 && body.Match != nil {
+		match := strings.TrimSpace(*body.Match)
 		for i := range h.cfg.ClaudeKey {
-			if h.cfg.ClaudeKey[i].APIKey == *body.Match {
-				h.cfg.ClaudeKey[i] = value
-				h.cfg.SanitizeClaudeKeys()
-				h.persist(c)
+			if h.cfg.ClaudeKey[i].APIKey == match {
+				targetIndex = i
+				break
+			}
+		}
+	}
+	if targetIndex == -1 {
+		c.JSON(404, gin.H{"error": "item not found"})
 		return
 	}
+
+	entry := h.cfg.ClaudeKey[targetIndex]
+	if body.Value.APIKey != nil {
+		entry.APIKey = strings.TrimSpace(*body.Value.APIKey)
 	}
+	if body.Value.Prefix != nil {
+		entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
 	}
-	c.JSON(404, gin.H{"error": "item not found"})
+	if body.Value.BaseURL != nil {
+		entry.BaseURL = strings.TrimSpace(*body.Value.BaseURL)
 	}
+	if body.Value.ProxyURL != nil {
+		entry.ProxyURL = strings.TrimSpace(*body.Value.ProxyURL)
+	}
+	if body.Value.Models != nil {
+		entry.Models = append([]config.ClaudeModel(nil), (*body.Value.Models)...)
+	}
+	if body.Value.Headers != nil {
+		entry.Headers = config.NormalizeHeaders(*body.Value.Headers)
+	}
+	if body.Value.ExcludedModels != nil {
+		entry.ExcludedModels = config.NormalizeExcludedModels(*body.Value.ExcludedModels)
+	}
+	normalizeClaudeKey(&entry)
+	h.cfg.ClaudeKey[targetIndex] = entry
+	h.cfg.SanitizeClaudeKeys()
+	h.persist(c)
+}
+
 func (h *Handler) DeleteClaudeKey(c *gin.Context) {
 	if val := c.Query("api-key"); val != "" {
 		out := make([]config.ClaudeKey, 0, len(h.cfg.ClaudeKey))
@@ -356,62 +394,73 @@ func (h *Handler) PutOpenAICompat(c *gin.Context) {
 	h.persist(c)
 }
 func (h *Handler) PatchOpenAICompat(c *gin.Context) {
+	type openAICompatPatch struct {
+		Name          *string                             `json:"name"`
+		Prefix        *string                             `json:"prefix"`
+		BaseURL       *string                             `json:"base-url"`
+		APIKeyEntries *[]config.OpenAICompatibilityAPIKey `json:"api-key-entries"`
+		Models        *[]config.OpenAICompatibilityModel  `json:"models"`
+		Headers       *map[string]string                  `json:"headers"`
+	}
 	var body struct {
 		Name  *string            `json:"name"`
 		Index *int               `json:"index"`
-		Value *config.OpenAICompatibility `json:"value"`
+		Value *openAICompatPatch `json:"value"`
 	}
 	if err := c.ShouldBindJSON(&body); err != nil || body.Value == nil {
 		c.JSON(400, gin.H{"error": "invalid body"})
 		return
 	}
-	normalizeOpenAICompatibilityEntry(body.Value)
-	// If base-url becomes empty, delete the provider instead of updating
-	if strings.TrimSpace(body.Value.BaseURL) == "" {
+	targetIndex := -1
 	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.OpenAICompatibility) {
-			h.cfg.OpenAICompatibility = append(h.cfg.OpenAICompatibility[:*body.Index], h.cfg.OpenAICompatibility[*body.Index+1:]...)
-			h.cfg.SanitizeOpenAICompatibility()
-			h.persist(c)
-			return
+		targetIndex = *body.Index
 	}
-		if body.Name != nil {
-			out := make([]config.OpenAICompatibility, 0, len(h.cfg.OpenAICompatibility))
-			removed := false
+	if targetIndex == -1 && body.Name != nil {
+		match := strings.TrimSpace(*body.Name)
 		for i := range h.cfg.OpenAICompatibility {
-				if !removed && h.cfg.OpenAICompatibility[i].Name == *body.Name {
-					removed = true
-					continue
-				}
-				out = append(out, h.cfg.OpenAICompatibility[i])
-			}
-			if removed {
-				h.cfg.OpenAICompatibility = out
-				h.cfg.SanitizeOpenAICompatibility()
-				h.persist(c)
-				return
+			if h.cfg.OpenAICompatibility[i].Name == match {
+				targetIndex = i
+				break
 			}
 		}
+	}
+	if targetIndex == -1 {
 		c.JSON(404, gin.H{"error": "item not found"})
 		return
 	}
-	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.OpenAICompatibility) {
-		h.cfg.OpenAICompatibility[*body.Index] = *body.Value
+
+	entry := h.cfg.OpenAICompatibility[targetIndex]
+	if body.Value.Name != nil {
+		entry.Name = strings.TrimSpace(*body.Value.Name)
+	}
+	if body.Value.Prefix != nil {
+		entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
+	}
+	if body.Value.BaseURL != nil {
+		trimmed := strings.TrimSpace(*body.Value.BaseURL)
+		if trimmed == "" {
+			h.cfg.OpenAICompatibility = append(h.cfg.OpenAICompatibility[:targetIndex], h.cfg.OpenAICompatibility[targetIndex+1:]...)
 			h.cfg.SanitizeOpenAICompatibility()
 			h.persist(c)
 			return
 		}
-	if body.Name != nil {
-		for i := range h.cfg.OpenAICompatibility {
-			if h.cfg.OpenAICompatibility[i].Name == *body.Name {
-				h.cfg.OpenAICompatibility[i] = *body.Value
+		entry.BaseURL = trimmed
+	}
+	if body.Value.APIKeyEntries != nil {
+		entry.APIKeyEntries = append([]config.OpenAICompatibilityAPIKey(nil), (*body.Value.APIKeyEntries)...)
+	}
+	if body.Value.Models != nil {
+		entry.Models = append([]config.OpenAICompatibilityModel(nil), (*body.Value.Models)...)
+	}
+	if body.Value.Headers != nil {
+		entry.Headers = config.NormalizeHeaders(*body.Value.Headers)
+	}
+	normalizeOpenAICompatibilityEntry(&entry)
+	h.cfg.OpenAICompatibility[targetIndex] = entry
 	h.cfg.SanitizeOpenAICompatibility()
 	h.persist(c)
-				return
-			}
-		}
-	}
-	c.JSON(404, gin.H{"error": "item not found"})
 }
+
 func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
 	if name := c.Query("name"); name != "" {
 		out := make([]config.OpenAICompatibility, 0, len(h.cfg.OpenAICompatibility))
@@ -438,6 +487,137 @@ func (h *Handler) DeleteOpenAICompat(c *gin.Context) {
 	c.JSON(400, gin.H{"error": "missing name or index"})
 }
 
+// vertex-api-key: []VertexCompatKey
+func (h *Handler) GetVertexCompatKeys(c *gin.Context) {
+	c.JSON(200, gin.H{"vertex-api-key": h.cfg.VertexCompatAPIKey})
+}
+func (h *Handler) PutVertexCompatKeys(c *gin.Context) {
+	data, err := c.GetRawData()
+	if err != nil {
+		c.JSON(400, gin.H{"error": "failed to read body"})
+		return
+	}
+	var arr []config.VertexCompatKey
+	if err = json.Unmarshal(data, &arr); err != nil {
+		var obj struct {
+			Items []config.VertexCompatKey `json:"items"`
+		}
+		if err2 := json.Unmarshal(data, &obj); err2 != nil || len(obj.Items) == 0 {
+			c.JSON(400, gin.H{"error": "invalid body"})
+			return
+		}
+		arr = obj.Items
+	}
+	for i := range arr {
+		normalizeVertexCompatKey(&arr[i])
+	}
+	h.cfg.VertexCompatAPIKey = arr
+	h.cfg.SanitizeVertexCompatKeys()
+	h.persist(c)
+}
+func (h *Handler) PatchVertexCompatKey(c *gin.Context) {
+	type vertexCompatPatch struct {
+		APIKey   *string                     `json:"api-key"`
+		Prefix   *string                     `json:"prefix"`
+		BaseURL  *string                     `json:"base-url"`
+		ProxyURL *string                     `json:"proxy-url"`
+		Headers  *map[string]string          `json:"headers"`
+		Models   *[]config.VertexCompatModel `json:"models"`
+	}
+	var body struct {
+		Index *int               `json:"index"`
+		Match *string            `json:"match"`
+		Value *vertexCompatPatch `json:"value"`
+	}
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil || body.Value == nil {
+		c.JSON(400, gin.H{"error": "invalid body"})
+		return
+	}
+	targetIndex := -1
+	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.VertexCompatAPIKey) {
+		targetIndex = *body.Index
+	}
+	if targetIndex == -1 && body.Match != nil {
+		match := strings.TrimSpace(*body.Match)
+		if match != "" {
+			for i := range h.cfg.VertexCompatAPIKey {
+				if h.cfg.VertexCompatAPIKey[i].APIKey == match {
+					targetIndex = i
+					break
+				}
+			}
+		}
+	}
+	if targetIndex == -1 {
+		c.JSON(404, gin.H{"error": "item not found"})
+		return
+	}
+
+	entry := h.cfg.VertexCompatAPIKey[targetIndex]
+	if body.Value.APIKey != nil {
+		trimmed := strings.TrimSpace(*body.Value.APIKey)
+		if trimmed == "" {
+			h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:targetIndex], h.cfg.VertexCompatAPIKey[targetIndex+1:]...)
+			h.cfg.SanitizeVertexCompatKeys()
+			h.persist(c)
+			return
+		}
+		entry.APIKey = trimmed
+	}
+	if body.Value.Prefix != nil {
+		entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
+	}
+	if body.Value.BaseURL != nil {
+		trimmed := strings.TrimSpace(*body.Value.BaseURL)
+		if trimmed == "" {
+			h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:targetIndex], h.cfg.VertexCompatAPIKey[targetIndex+1:]...)
+			h.cfg.SanitizeVertexCompatKeys()
+			h.persist(c)
+			return
+		}
+		entry.BaseURL = trimmed
+	}
+	if body.Value.ProxyURL != nil {
+		entry.ProxyURL = strings.TrimSpace(*body.Value.ProxyURL)
+	}
+	if body.Value.Headers != nil {
+		entry.Headers = config.NormalizeHeaders(*body.Value.Headers)
+	}
+	if body.Value.Models != nil {
+		entry.Models = append([]config.VertexCompatModel(nil), (*body.Value.Models)...)
+	}
+	normalizeVertexCompatKey(&entry)
+	h.cfg.VertexCompatAPIKey[targetIndex] = entry
+	h.cfg.SanitizeVertexCompatKeys()
+	h.persist(c)
+}
+
+func (h *Handler) DeleteVertexCompatKey(c *gin.Context) {
+	if val := strings.TrimSpace(c.Query("api-key")); val != "" {
+		out := make([]config.VertexCompatKey, 0, len(h.cfg.VertexCompatAPIKey))
+		for _, v := range h.cfg.VertexCompatAPIKey {
+			if v.APIKey != val {
+				out = append(out, v)
+			}
+		}
+		h.cfg.VertexCompatAPIKey = out
+		h.cfg.SanitizeVertexCompatKeys()
+		h.persist(c)
+		return
+	}
+	if idxStr := c.Query("index"); idxStr != "" {
+		var idx int
+		_, errScan := fmt.Sscanf(idxStr, "%d", &idx)
+		if errScan == nil && idx >= 0 && idx < len(h.cfg.VertexCompatAPIKey) {
+			h.cfg.VertexCompatAPIKey = append(h.cfg.VertexCompatAPIKey[:idx], h.cfg.VertexCompatAPIKey[idx+1:]...)
+			h.cfg.SanitizeVertexCompatKeys()
+			h.persist(c)
+			return
+		}
+	}
+	c.JSON(400, gin.H{"error": "missing api-key or index"})
+}
+
 // oauth-excluded-models: map[string][]string
 func (h *Handler) GetOAuthExcludedModels(c *gin.Context) {
 	c.JSON(200, gin.H{"oauth-excluded-models": config.NormalizeOAuthExcludedModels(h.cfg.OAuthExcludedModels)})
@@ -523,6 +703,103 @@ func (h *Handler) DeleteOAuthExcludedModels(c *gin.Context) {
 	h.persist(c)
 }
 
+// oauth-model-alias: map[string][]OAuthModelAlias
+func (h *Handler) GetOAuthModelAlias(c *gin.Context) {
+	c.JSON(200, gin.H{"oauth-model-alias": sanitizedOAuthModelAlias(h.cfg.OAuthModelAlias)})
+}
+
+func (h *Handler) PutOAuthModelAlias(c *gin.Context) {
+	data, err := c.GetRawData()
+	if err != nil {
+		c.JSON(400, gin.H{"error": "failed to read body"})
+		return
+	}
+	var entries map[string][]config.OAuthModelAlias
+	if err = json.Unmarshal(data, &entries); err != nil {
+		var wrapper struct {
+			Items map[string][]config.OAuthModelAlias `json:"items"`
+		}
+		if err2 := json.Unmarshal(data, &wrapper); err2 != nil {
+			c.JSON(400, gin.H{"error": "invalid body"})
+			return
+		}
+		entries = wrapper.Items
+	}
+	h.cfg.OAuthModelAlias = sanitizedOAuthModelAlias(entries)
+	h.persist(c)
+}
+
+func (h *Handler) PatchOAuthModelAlias(c *gin.Context) {
+	var body struct {
+		Provider *string                  `json:"provider"`
+		Channel  *string                  `json:"channel"`
+		Aliases  []config.OAuthModelAlias `json:"aliases"`
+	}
+	if errBindJSON := c.ShouldBindJSON(&body); errBindJSON != nil {
+		c.JSON(400, gin.H{"error": "invalid body"})
+		return
+	}
+	channelRaw := ""
+	if body.Channel != nil {
+		channelRaw = *body.Channel
+	} else if body.Provider != nil {
+		channelRaw = *body.Provider
+	}
+	channel := strings.ToLower(strings.TrimSpace(channelRaw))
+	if channel == "" {
+		c.JSON(400, gin.H{"error": "invalid channel"})
+		return
+	}
+
+	normalizedMap := sanitizedOAuthModelAlias(map[string][]config.OAuthModelAlias{channel: body.Aliases})
+	normalized := normalizedMap[channel]
+	if len(normalized) == 0 {
+		if h.cfg.OAuthModelAlias == nil {
+			c.JSON(404, gin.H{"error": "channel not found"})
+			return
+		}
+		if _, ok := h.cfg.OAuthModelAlias[channel]; !ok {
+			c.JSON(404, gin.H{"error": "channel not found"})
+			return
+		}
+		delete(h.cfg.OAuthModelAlias, channel)
+		if len(h.cfg.OAuthModelAlias) == 0 {
+			h.cfg.OAuthModelAlias = nil
+		}
+		h.persist(c)
+		return
+	}
+	if h.cfg.OAuthModelAlias == nil {
+		h.cfg.OAuthModelAlias = make(map[string][]config.OAuthModelAlias)
+	}
+	h.cfg.OAuthModelAlias[channel] = normalized
+	h.persist(c)
+}
+
+func (h *Handler) DeleteOAuthModelAlias(c *gin.Context) {
+	channel := strings.ToLower(strings.TrimSpace(c.Query("channel")))
+	if channel == "" {
+		channel = strings.ToLower(strings.TrimSpace(c.Query("provider")))
+	}
+	if channel == "" {
+		c.JSON(400, gin.H{"error": "missing channel"})
+		return
+	}
+	if h.cfg.OAuthModelAlias == nil {
+		c.JSON(404, gin.H{"error": "channel not found"})
+		return
+	}
+	if _, ok := h.cfg.OAuthModelAlias[channel]; !ok {
+		c.JSON(404, gin.H{"error": "channel not found"})
+		return
+	}
+	delete(h.cfg.OAuthModelAlias, channel)
+	if len(h.cfg.OAuthModelAlias) == 0 {
+		h.cfg.OAuthModelAlias = nil
+	}
+	h.persist(c)
+}
+
 // codex-api-key: []CodexKey
 func (h *Handler) GetCodexKeys(c *gin.Context) {
 	c.JSON(200, gin.H{"codex-api-key": h.cfg.CodexKey})
@@ -548,11 +825,7 @@ func (h *Handler) PutCodexKeys(c *gin.Context) {
 	filtered := make([]config.CodexKey, 0, len(arr))
 	for i := range arr {
 		entry := arr[i]
-		entry.APIKey = strings.TrimSpace(entry.APIKey)
-		entry.BaseURL = strings.TrimSpace(entry.BaseURL)
-		entry.ProxyURL = strings.TrimSpace(entry.ProxyURL)
-		entry.Headers = config.NormalizeHeaders(entry.Headers)
-		entry.ExcludedModels = config.NormalizeExcludedModels(entry.ExcludedModels)
+		normalizeCodexKey(&entry)
 		if entry.BaseURL == "" {
 			continue
 		}
@@ -563,66 +836,77 @@ func (h *Handler) PutCodexKeys(c *gin.Context) {
 	h.persist(c)
 }
 func (h *Handler) PatchCodexKey(c *gin.Context) {
+	type codexKeyPatch struct {
+		APIKey         *string              `json:"api-key"`
+		Prefix         *string              `json:"prefix"`
+		BaseURL        *string              `json:"base-url"`
+		ProxyURL       *string              `json:"proxy-url"`
+		Models         *[]config.CodexModel `json:"models"`
+		Headers        *map[string]string   `json:"headers"`
+		ExcludedModels *[]string            `json:"excluded-models"`
+	}
 	var body struct {
 		Index *int           `json:"index"`
 		Match *string        `json:"match"`
-		Value *config.CodexKey `json:"value"`
+		Value *codexKeyPatch `json:"value"`
 	}
 	if err := c.ShouldBindJSON(&body); err != nil || body.Value == nil {
 		c.JSON(400, gin.H{"error": "invalid body"})
 		return
 	}
-	value := *body.Value
-	value.APIKey = strings.TrimSpace(value.APIKey)
-	value.BaseURL = strings.TrimSpace(value.BaseURL)
-	value.ProxyURL = strings.TrimSpace(value.ProxyURL)
-	value.Headers = config.NormalizeHeaders(value.Headers)
-	value.ExcludedModels = config.NormalizeExcludedModels(value.ExcludedModels)
-	// If base-url becomes empty, delete instead of update
-	if value.BaseURL == "" {
+	targetIndex := -1
 	if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.CodexKey) {
-			h.cfg.CodexKey = append(h.cfg.CodexKey[:*body.Index], h.cfg.CodexKey[*body.Index+1:]...)
-			h.cfg.SanitizeCodexKeys()
-			h.persist(c)
-			return
+		targetIndex = *body.Index
 	}
-		if body.Match != nil {
-			out := make([]config.CodexKey, 0, len(h.cfg.CodexKey))
-			removed := false
+	if targetIndex == -1 && body.Match != nil {
+		match := strings.TrimSpace(*body.Match)
 		for i := range h.cfg.CodexKey {
-				if !removed && h.cfg.CodexKey[i].APIKey == *body.Match {
-					removed = true
-					continue
-				}
-				out = append(out, h.cfg.CodexKey[i])
-			}
-			if removed {
-				h.cfg.CodexKey = out
-				h.cfg.SanitizeCodexKeys()
-				h.persist(c)
-				return
-			}
-		}
-	} else {
-		if body.Index != nil && *body.Index >= 0 && *body.Index < len(h.cfg.CodexKey) {
-			h.cfg.CodexKey[*body.Index] = value
-			h.cfg.SanitizeCodexKeys()
-			h.persist(c)
-			return
-		}
-		if body.Match != nil {
-			for i := range h.cfg.CodexKey {
-				if h.cfg.CodexKey[i].APIKey == *body.Match {
-					h.cfg.CodexKey[i] = value
-					h.cfg.SanitizeCodexKeys()
-					h.persist(c)
-					return
-				}
+			if h.cfg.CodexKey[i].APIKey == match {
+				targetIndex = i
+				break
 			}
 		}
 	}
+	if targetIndex == -1 {
 		c.JSON(404, gin.H{"error": "item not found"})
+		return
 	}
+
+	entry := h.cfg.CodexKey[targetIndex]
+	if body.Value.APIKey != nil {
+		entry.APIKey = strings.TrimSpace(*body.Value.APIKey)
+	}
+	if body.Value.Prefix != nil {
+		entry.Prefix = strings.TrimSpace(*body.Value.Prefix)
+	}
+	if body.Value.BaseURL != nil {
+		trimmed := strings.TrimSpace(*body.Value.BaseURL)
+		if trimmed == "" {
+			h.cfg.CodexKey = append(h.cfg.CodexKey[:targetIndex], h.cfg.CodexKey[targetIndex+1:]...)
+			h.cfg.SanitizeCodexKeys()
+			h.persist(c)
+			return
+		}
+		entry.BaseURL = trimmed
+	}
+	if body.Value.ProxyURL != nil {
+		entry.ProxyURL = strings.TrimSpace(*body.Value.ProxyURL)
+	}
+	if body.Value.Models != nil {
+		entry.Models = append([]config.CodexModel(nil), (*body.Value.Models)...)
+	}
+	if body.Value.Headers != nil {
+		entry.Headers = config.NormalizeHeaders(*body.Value.Headers)
+	}
+	if body.Value.ExcludedModels != nil {
+		entry.ExcludedModels = config.NormalizeExcludedModels(*body.Value.ExcludedModels)
+	}
+	normalizeCodexKey(&entry)
+	h.cfg.CodexKey[targetIndex] = entry
+	h.cfg.SanitizeCodexKeys()
+	h.persist(c)
+}
+
 func (h *Handler) DeleteCodexKey(c *gin.Context) {
 	if val := c.Query("api-key"); val != "" {
 		out := make([]config.CodexKey, 0, len(h.cfg.CodexKey))
@@ -707,6 +991,79 @@ func normalizeClaudeKey(entry *config.ClaudeKey) {
 	entry.Models = normalized
 }
 
+func normalizeCodexKey(entry *config.CodexKey) {
+	if entry == nil {
+		return
+	}
+	entry.APIKey = strings.TrimSpace(entry.APIKey)
+	entry.Prefix = strings.TrimSpace(entry.Prefix)
+	entry.BaseURL = strings.TrimSpace(entry.BaseURL)
+	entry.ProxyURL = strings.TrimSpace(entry.ProxyURL)
+	entry.Headers = config.NormalizeHeaders(entry.Headers)
+	entry.ExcludedModels = config.NormalizeExcludedModels(entry.ExcludedModels)
+	if len(entry.Models) == 0 {
+		return
+	}
+	normalized := make([]config.CodexModel, 0, len(entry.Models))
+	for i := range entry.Models {
+		model := entry.Models[i]
+		model.Name = strings.TrimSpace(model.Name)
+		model.Alias = strings.TrimSpace(model.Alias)
+		if model.Name == "" && model.Alias == "" {
+			continue
+		}
+		normalized = append(normalized, model)
+	}
+	entry.Models = normalized
+}
+
+func normalizeVertexCompatKey(entry *config.VertexCompatKey) {
+	if entry == nil {
+		return
+	}
+	entry.APIKey = strings.TrimSpace(entry.APIKey)
+	entry.Prefix = strings.TrimSpace(entry.Prefix)
+	entry.BaseURL = strings.TrimSpace(entry.BaseURL)
+	entry.ProxyURL = strings.TrimSpace(entry.ProxyURL)
+	entry.Headers = config.NormalizeHeaders(entry.Headers)
+	if len(entry.Models) == 0 {
+		return
+	}
+	normalized := make([]config.VertexCompatModel, 0, len(entry.Models))
+	for i := range entry.Models {
+		model := entry.Models[i]
+		model.Name = strings.TrimSpace(model.Name)
+		model.Alias = strings.TrimSpace(model.Alias)
+		if model.Name == "" || model.Alias == "" {
+			continue
+		}
+		normalized = append(normalized, model)
+	}
+	entry.Models = normalized
+}
+
+func sanitizedOAuthModelAlias(entries map[string][]config.OAuthModelAlias) map[string][]config.OAuthModelAlias {
+	if len(entries) == 0 {
+		return nil
+	}
+	copied := make(map[string][]config.OAuthModelAlias, len(entries))
+	for channel, aliases := range entries {
+		if len(aliases) == 0 {
+			continue
+		}
+		copied[channel] = append([]config.OAuthModelAlias(nil), aliases...)
+	}
+	if len(copied) == 0 {
+		return nil
+	}
+	cfg := config.Config{OAuthModelAlias: copied}
+	cfg.SanitizeOAuthModelAlias()
+	if len(cfg.OAuthModelAlias) == 0 {
+		return nil
+	}
+	return cfg.OAuthModelAlias
+}
+
 // GetAmpCode returns the complete ampcode configuration.
 func (h *Handler) GetAmpCode(c *gin.Context) {
 	if h == nil || h.cfg == nil {
@@ -858,3 +1215,151 @@ func (h *Handler) GetAmpForceModelMappings(c *gin.Context) {
 func (h *Handler) PutAmpForceModelMappings(c *gin.Context) {
 	h.updateBoolField(c, func(v bool) { h.cfg.AmpCode.ForceModelMappings = v })
 }
+
+// GetAmpUpstreamAPIKeys returns the ampcode upstream API keys mapping.
+func (h *Handler) GetAmpUpstreamAPIKeys(c *gin.Context) {
+	if h == nil || h.cfg == nil {
+		c.JSON(200, gin.H{"upstream-api-keys": []config.AmpUpstreamAPIKeyEntry{}})
+		return
+	}
+	c.JSON(200, gin.H{"upstream-api-keys": h.cfg.AmpCode.UpstreamAPIKeys})
+}
+
+// PutAmpUpstreamAPIKeys replaces all ampcode upstream API keys mappings.
+func (h *Handler) PutAmpUpstreamAPIKeys(c *gin.Context) {
+	var body struct {
+		Value []config.AmpUpstreamAPIKeyEntry `json:"value"`
+	}
+	if err := c.ShouldBindJSON(&body); err != nil {
+		c.JSON(400, gin.H{"error": "invalid body"})
+		return
+	}
+	// Normalize entries: trim whitespace, filter empty
+	normalized := normalizeAmpUpstreamAPIKeyEntries(body.Value)
+	h.cfg.AmpCode.UpstreamAPIKeys = normalized
+	h.persist(c)
+}
+
+// PatchAmpUpstreamAPIKeys adds or updates upstream API keys entries.
+// Matching is done by upstream-api-key value.
+func (h *Handler) PatchAmpUpstreamAPIKeys(c *gin.Context) {
+	var body struct {
+		Value []config.AmpUpstreamAPIKeyEntry `json:"value"`
+	}
+	if err := c.ShouldBindJSON(&body); err != nil {
+		c.JSON(400, gin.H{"error": "invalid body"})
+		return
+	}
+
+	existing := make(map[string]int)
+	for i, entry := range h.cfg.AmpCode.UpstreamAPIKeys {
+		existing[strings.TrimSpace(entry.UpstreamAPIKey)] = i
+	}
+
+	for _, newEntry := range body.Value {
+		upstreamKey := strings.TrimSpace(newEntry.UpstreamAPIKey)
+		if upstreamKey == "" {
+			continue
+		}
+		normalizedEntry := config.AmpUpstreamAPIKeyEntry{
+			UpstreamAPIKey: upstreamKey,
+			APIKeys:        normalizeAPIKeysList(newEntry.APIKeys),
+		}
+		if idx, ok := existing[upstreamKey]; ok {
+			h.cfg.AmpCode.UpstreamAPIKeys[idx] = normalizedEntry
+		} else {
+			h.cfg.AmpCode.UpstreamAPIKeys = append(h.cfg.AmpCode.UpstreamAPIKeys, normalizedEntry)
+			existing[upstreamKey] = len(h.cfg.AmpCode.UpstreamAPIKeys) - 1
+		}
+	}
+	h.persist(c)
+}
+
+// DeleteAmpUpstreamAPIKeys removes specified upstream API keys entries.
+// Body must be JSON: {"value": ["<upstream-api-key>", ...]}.
+// If "value" is an empty array, clears all entries.
+// If JSON is invalid or "value" is missing/null, returns 400 and does not persist any change.
+func (h *Handler) DeleteAmpUpstreamAPIKeys(c *gin.Context) {
+	var body struct {
+		Value []string `json:"value"`
+	}
+	if err := c.ShouldBindJSON(&body); err != nil {
+		c.JSON(400, gin.H{"error": "invalid body"})
+		return
+	}
+
+	if body.Value == nil {
+		c.JSON(400, gin.H{"error": "missing value"})
+		return
+	}
+
+	// Empty array means clear all
+	if len(body.Value) == 0 {
+		h.cfg.AmpCode.UpstreamAPIKeys = nil
+		h.persist(c)
+		return
+	}
+
+	toRemove := make(map[string]bool)
+	for _, key := range body.Value {
+		trimmed := strings.TrimSpace(key)
+		if trimmed == "" {
+			continue
+		}
+		toRemove[trimmed] = true
+	}
+	if len(toRemove) == 0 {
+		c.JSON(400, gin.H{"error": "empty value"})
+		return
+	}
+
+	newEntries := make([]config.AmpUpstreamAPIKeyEntry, 0, len(h.cfg.AmpCode.UpstreamAPIKeys))
+	for _, entry := range h.cfg.AmpCode.UpstreamAPIKeys {
+		if !toRemove[strings.TrimSpace(entry.UpstreamAPIKey)] {
+			newEntries = append(newEntries, entry)
+		}
+	}
+	h.cfg.AmpCode.UpstreamAPIKeys = newEntries
+	h.persist(c)
+}
+
+// normalizeAmpUpstreamAPIKeyEntries normalizes a list of upstream API key entries.
+func normalizeAmpUpstreamAPIKeyEntries(entries []config.AmpUpstreamAPIKeyEntry) []config.AmpUpstreamAPIKeyEntry {
+	if len(entries) == 0 {
+		return nil
+	}
+	out := make([]config.AmpUpstreamAPIKeyEntry, 0, len(entries))
+	for _, entry := range entries {
+		upstreamKey := strings.TrimSpace(entry.UpstreamAPIKey)
+		if upstreamKey == "" {
+			continue
+		}
+		apiKeys := normalizeAPIKeysList(entry.APIKeys)
+		out = append(out, config.AmpUpstreamAPIKeyEntry{
+			UpstreamAPIKey: upstreamKey,
+			APIKeys:        apiKeys,
+		})
+	}
+	if len(out) == 0 {
+		return nil
+	}
+	return out
+}
+
+// normalizeAPIKeysList trims and filters empty strings from a list of API keys.
+func normalizeAPIKeysList(keys []string) []string {
+	if len(keys) == 0 {
+		return nil
+	}
+	out := make([]string, 0, len(keys))
+	for _, k := range keys {
+		trimmed := strings.TrimSpace(k)
+		if trimmed != "" {
+			out = append(out, trimmed)
+		}
+	}
+	if len(out) == 0 {
+		return nil
+	}
+	return out
+}

internal/api/handlers/management/handler.go

@@ -24,8 +24,15 @@ import (
 type attemptInfo struct {
 	count        int
 	blockedUntil time.Time
+	lastActivity time.Time // track last activity for cleanup
 }
 
+// attemptCleanupInterval controls how often stale IP entries are purged
+const attemptCleanupInterval = 1 * time.Hour
+
+// attemptMaxIdleTime controls how long an IP can be idle before cleanup
+const attemptMaxIdleTime = 2 * time.Hour
+
 // Handler aggregates config reference, persistence path and helpers.
 type Handler struct {
 	cfg                 *config.Config
@@ -47,7 +54,7 @@ func NewHandler(cfg *config.Config, configFilePath string, manager *coreauth.Man
 	envSecret, _ := os.LookupEnv("MANAGEMENT_PASSWORD")
 	envSecret = strings.TrimSpace(envSecret)
 
-	return &Handler{
+	h := &Handler{
 		cfg:                 cfg,
 		configFilePath:      configFilePath,
 		failedAttempts:      make(map[string]*attemptInfo),
@@ -57,6 +64,43 @@ func NewHandler(cfg *config.Config, configFilePath string, manager *coreauth.Man
 		allowRemoteOverride: envSecret != "",
 		envSecret:           envSecret,
 	}
+	h.startAttemptCleanup()
+	return h
+}
+
+// startAttemptCleanup launches a background goroutine that periodically
+// removes stale IP entries from failedAttempts to prevent memory leaks.
+func (h *Handler) startAttemptCleanup() {
+	go func() {
+		ticker := time.NewTicker(attemptCleanupInterval)
+		defer ticker.Stop()
+		for range ticker.C {
+			h.purgeStaleAttempts()
+		}
+	}()
+}
+
+// purgeStaleAttempts removes IP entries that have been idle beyond attemptMaxIdleTime
+// and whose ban (if any) has expired.
+func (h *Handler) purgeStaleAttempts() {
+	now := time.Now()
+	h.attemptsMu.Lock()
+	defer h.attemptsMu.Unlock()
+	for ip, ai := range h.failedAttempts {
+		// Skip if still banned
+		if !ai.blockedUntil.IsZero() && now.Before(ai.blockedUntil) {
+			continue
+		}
+		// Remove if idle too long
+		if now.Sub(ai.lastActivity) > attemptMaxIdleTime {
+			delete(h.failedAttempts, ip)
+		}
+	}
+}
+
+// NewHandler creates a new management handler instance.
+func NewHandlerWithoutConfigFilePath(cfg *config.Config, manager *coreauth.Manager) *Handler {
+	return NewHandler(cfg, "", manager)
 }
 
 // SetConfig updates the in-memory config reference when the server hot-reloads.
@@ -144,6 +188,7 @@ func (h *Handler) Middleware() gin.HandlerFunc {
 					h.failedAttempts[clientIP] = aip
 				}
 				aip.count++
+				aip.lastActivity = time.Now()
 				if aip.count >= maxFailures {
 					aip.blockedUntil = time.Now().Add(banDuration)
 					aip.count = 0

internal/api/handlers/management/logs.go

@@ -13,7 +13,7 @@ import (
 	"time"
 
 	"github.com/gin-gonic/gin"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
 )
 
 const (
@@ -209,6 +209,94 @@ func (h *Handler) GetRequestErrorLogs(c *gin.Context) {
 	c.JSON(http.StatusOK, gin.H{"files": files})
 }
 
+// GetRequestLogByID finds and downloads a request log file by its request ID.
+// The ID is matched against the suffix of log file names (format: *-{requestID}.log).
+func (h *Handler) GetRequestLogByID(c *gin.Context) {
+	if h == nil {
+		c.JSON(http.StatusInternalServerError, gin.H{"error": "handler unavailable"})
+		return
+	}
+	if h.cfg == nil {
+		c.JSON(http.StatusServiceUnavailable, gin.H{"error": "configuration unavailable"})
+		return
+	}
+
+	dir := h.logDirectory()
+	if strings.TrimSpace(dir) == "" {
+		c.JSON(http.StatusInternalServerError, gin.H{"error": "log directory not configured"})
+		return
+	}
+
+	requestID := strings.TrimSpace(c.Param("id"))
+	if requestID == "" {
+		requestID = strings.TrimSpace(c.Query("id"))
+	}
+	if requestID == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "missing request ID"})
+		return
+	}
+	if strings.ContainsAny(requestID, "/\\") {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid request ID"})
+		return
+	}
+
+	entries, err := os.ReadDir(dir)
+	if err != nil {
+		if os.IsNotExist(err) {
+			c.JSON(http.StatusNotFound, gin.H{"error": "log directory not found"})
+			return
+		}
+		c.JSON(http.StatusInternalServerError, gin.H{"error": fmt.Sprintf("failed to list log directory: %v", err)})
+		return
+	}
+
+	suffix := "-" + requestID + ".log"
+	var matchedFile string
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		name := entry.Name()
+		if strings.HasSuffix(name, suffix) {
+			matchedFile = name
+			break
+		}
+	}
+
+	if matchedFile == "" {
+		c.JSON(http.StatusNotFound, gin.H{"error": "log file not found for the given request ID"})
+		return
+	}
+
+	dirAbs, errAbs := filepath.Abs(dir)
+	if errAbs != nil {
+		c.JSON(http.StatusInternalServerError, gin.H{"error": fmt.Sprintf("failed to resolve log directory: %v", errAbs)})
+		return
+	}
+	fullPath := filepath.Clean(filepath.Join(dirAbs, matchedFile))
+	prefix := dirAbs + string(os.PathSeparator)
+	if !strings.HasPrefix(fullPath, prefix) {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid log file path"})
+		return
+	}
+
+	info, errStat := os.Stat(fullPath)
+	if errStat != nil {
+		if os.IsNotExist(errStat) {
+			c.JSON(http.StatusNotFound, gin.H{"error": "log file not found"})
+			return
+		}
+		c.JSON(http.StatusInternalServerError, gin.H{"error": fmt.Sprintf("failed to read log file: %v", errStat)})
+		return
+	}
+	if info.IsDir() {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid log file"})
+		return
+	}
+
+	c.FileAttachment(fullPath, matchedFile)
+}
+
 // DownloadRequestErrorLog downloads a specific error request log file by name.
 func (h *Handler) DownloadRequestErrorLog(c *gin.Context) {
 	if h == nil {
@@ -272,16 +360,7 @@ func (h *Handler) logDirectory() string {
 	if h.logDir != "" {
 		return h.logDir
 	}
-	if base := util.WritablePath(); base != "" {
-		return filepath.Join(base, "logs")
-	}
-	if h.configFilePath != "" {
-		dir := filepath.Dir(h.configFilePath)
-		if dir != "" && dir != "." {
-			return filepath.Join(dir, "logs")
-		}
-	}
-	return "logs"
+	return logging.ResolveLogDirectory(h.cfg)
 }
 
 func (h *Handler) collectLogFiles(dir string) ([]string, error) {

internal/api/handlers/management/model_definitions.go

@@ -0,0 +1,33 @@
+package management
+
+import (
+	"net/http"
+	"strings"
+
+	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
+)
+
+// GetStaticModelDefinitions returns static model metadata for a given channel.
+// Channel is provided via path param (:channel) or query param (?channel=...).
+func (h *Handler) GetStaticModelDefinitions(c *gin.Context) {
+	channel := strings.TrimSpace(c.Param("channel"))
+	if channel == "" {
+		channel = strings.TrimSpace(c.Query("channel"))
+	}
+	if channel == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "channel is required"})
+		return
+	}
+
+	models := registry.GetStaticModelDefinitionsByChannel(channel)
+	if models == nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "unknown channel", "channel": channel})
+		return
+	}
+
+	c.JSON(http.StatusOK, gin.H{
+		"channel": strings.ToLower(strings.TrimSpace(channel)),
+		"models":  models,
+	})
+}

internal/api/handlers/management/oauth_callback.go

@@ -0,0 +1,100 @@
+package management
+
+import (
+	"errors"
+	"net/http"
+	"net/url"
+	"strings"
+
+	"github.com/gin-gonic/gin"
+)
+
+type oauthCallbackRequest struct {
+	Provider    string `json:"provider"`
+	RedirectURL string `json:"redirect_url"`
+	Code        string `json:"code"`
+	State       string `json:"state"`
+	Error       string `json:"error"`
+}
+
+func (h *Handler) PostOAuthCallback(c *gin.Context) {
+	if h == nil || h.cfg == nil {
+		c.JSON(http.StatusInternalServerError, gin.H{"status": "error", "error": "handler not initialized"})
+		return
+	}
+
+	var req oauthCallbackRequest
+	if err := c.ShouldBindJSON(&req); err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "invalid body"})
+		return
+	}
+
+	canonicalProvider, err := NormalizeOAuthProvider(req.Provider)
+	if err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "unsupported provider"})
+		return
+	}
+
+	state := strings.TrimSpace(req.State)
+	code := strings.TrimSpace(req.Code)
+	errMsg := strings.TrimSpace(req.Error)
+
+	if rawRedirect := strings.TrimSpace(req.RedirectURL); rawRedirect != "" {
+		u, errParse := url.Parse(rawRedirect)
+		if errParse != nil {
+			c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "invalid redirect_url"})
+			return
+		}
+		q := u.Query()
+		if state == "" {
+			state = strings.TrimSpace(q.Get("state"))
+		}
+		if code == "" {
+			code = strings.TrimSpace(q.Get("code"))
+		}
+		if errMsg == "" {
+			errMsg = strings.TrimSpace(q.Get("error"))
+			if errMsg == "" {
+				errMsg = strings.TrimSpace(q.Get("error_description"))
+			}
+		}
+	}
+
+	if state == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "state is required"})
+		return
+	}
+	if err := ValidateOAuthState(state); err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "invalid state"})
+		return
+	}
+	if code == "" && errMsg == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "code or error is required"})
+		return
+	}
+
+	sessionProvider, sessionStatus, ok := GetOAuthSession(state)
+	if !ok {
+		c.JSON(http.StatusNotFound, gin.H{"status": "error", "error": "unknown or expired state"})
+		return
+	}
+	if sessionStatus != "" {
+		c.JSON(http.StatusConflict, gin.H{"status": "error", "error": "oauth flow is not pending"})
+		return
+	}
+	if !strings.EqualFold(sessionProvider, canonicalProvider) {
+		c.JSON(http.StatusBadRequest, gin.H{"status": "error", "error": "provider does not match state"})
+		return
+	}
+
+	if _, errWrite := WriteOAuthCallbackFileForPendingSession(h.cfg.AuthDir, canonicalProvider, state, code, errMsg); errWrite != nil {
+		if errors.Is(errWrite, errOAuthSessionNotPending) {
+			c.JSON(http.StatusConflict, gin.H{"status": "error", "error": "oauth flow is not pending"})
+			return
+		}
+		c.JSON(http.StatusInternalServerError, gin.H{"status": "error", "error": "failed to persist oauth callback"})
+		return
+	}
+
+	c.JSON(http.StatusOK, gin.H{"status": "ok"})
+}

internal/api/handlers/management/oauth_sessions.go

@@ -0,0 +1,283 @@
+package management
+
+import (
+	"encoding/json"
+	"errors"
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+	"sync"
+	"time"
+)
+
+const (
+	oauthSessionTTL     = 10 * time.Minute
+	maxOAuthStateLength = 128
+)
+
+var (
+	errInvalidOAuthState      = errors.New("invalid oauth state")
+	errUnsupportedOAuthFlow   = errors.New("unsupported oauth provider")
+	errOAuthSessionNotPending = errors.New("oauth session is not pending")
+)
+
+type oauthSession struct {
+	Provider  string
+	Status    string
+	CreatedAt time.Time
+	ExpiresAt time.Time
+}
+
+type oauthSessionStore struct {
+	mu       sync.RWMutex
+	ttl      time.Duration
+	sessions map[string]oauthSession
+}
+
+func newOAuthSessionStore(ttl time.Duration) *oauthSessionStore {
+	if ttl <= 0 {
+		ttl = oauthSessionTTL
+	}
+	return &oauthSessionStore{
+		ttl:      ttl,
+		sessions: make(map[string]oauthSession),
+	}
+}
+
+func (s *oauthSessionStore) purgeExpiredLocked(now time.Time) {
+	for state, session := range s.sessions {
+		if !session.ExpiresAt.IsZero() && now.After(session.ExpiresAt) {
+			delete(s.sessions, state)
+		}
+	}
+}
+
+func (s *oauthSessionStore) Register(state, provider string) {
+	state = strings.TrimSpace(state)
+	provider = strings.ToLower(strings.TrimSpace(provider))
+	if state == "" || provider == "" {
+		return
+	}
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	s.sessions[state] = oauthSession{
+		Provider:  provider,
+		Status:    "",
+		CreatedAt: now,
+		ExpiresAt: now.Add(s.ttl),
+	}
+}
+
+func (s *oauthSessionStore) SetError(state, message string) {
+	state = strings.TrimSpace(state)
+	message = strings.TrimSpace(message)
+	if state == "" {
+		return
+	}
+	if message == "" {
+		message = "Authentication failed"
+	}
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	session, ok := s.sessions[state]
+	if !ok {
+		return
+	}
+	session.Status = message
+	session.ExpiresAt = now.Add(s.ttl)
+	s.sessions[state] = session
+}
+
+func (s *oauthSessionStore) Complete(state string) {
+	state = strings.TrimSpace(state)
+	if state == "" {
+		return
+	}
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	delete(s.sessions, state)
+}
+
+func (s *oauthSessionStore) CompleteProvider(provider string) int {
+	provider = strings.ToLower(strings.TrimSpace(provider))
+	if provider == "" {
+		return 0
+	}
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	removed := 0
+	for state, session := range s.sessions {
+		if strings.EqualFold(session.Provider, provider) {
+			delete(s.sessions, state)
+			removed++
+		}
+	}
+	return removed
+}
+
+func (s *oauthSessionStore) Get(state string) (oauthSession, bool) {
+	state = strings.TrimSpace(state)
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	session, ok := s.sessions[state]
+	return session, ok
+}
+
+func (s *oauthSessionStore) IsPending(state, provider string) bool {
+	state = strings.TrimSpace(state)
+	provider = strings.ToLower(strings.TrimSpace(provider))
+	now := time.Now()
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.purgeExpiredLocked(now)
+	session, ok := s.sessions[state]
+	if !ok {
+		return false
+	}
+	if session.Status != "" {
+		return false
+	}
+	if provider == "" {
+		return true
+	}
+	return strings.EqualFold(session.Provider, provider)
+}
+
+var oauthSessions = newOAuthSessionStore(oauthSessionTTL)
+
+func RegisterOAuthSession(state, provider string) { oauthSessions.Register(state, provider) }
+
+func SetOAuthSessionError(state, message string) { oauthSessions.SetError(state, message) }
+
+func CompleteOAuthSession(state string) { oauthSessions.Complete(state) }
+
+func CompleteOAuthSessionsByProvider(provider string) int {
+	return oauthSessions.CompleteProvider(provider)
+}
+
+func GetOAuthSession(state string) (provider string, status string, ok bool) {
+	session, ok := oauthSessions.Get(state)
+	if !ok {
+		return "", "", false
+	}
+	return session.Provider, session.Status, true
+}
+
+func IsOAuthSessionPending(state, provider string) bool {
+	return oauthSessions.IsPending(state, provider)
+}
+
+func ValidateOAuthState(state string) error {
+	trimmed := strings.TrimSpace(state)
+	if trimmed == "" {
+		return fmt.Errorf("%w: empty", errInvalidOAuthState)
+	}
+	if len(trimmed) > maxOAuthStateLength {
+		return fmt.Errorf("%w: too long", errInvalidOAuthState)
+	}
+	if strings.Contains(trimmed, "/") || strings.Contains(trimmed, "\\") {
+		return fmt.Errorf("%w: contains path separator", errInvalidOAuthState)
+	}
+	if strings.Contains(trimmed, "..") {
+		return fmt.Errorf("%w: contains '..'", errInvalidOAuthState)
+	}
+	for _, r := range trimmed {
+		switch {
+		case r >= 'a' && r <= 'z':
+		case r >= 'A' && r <= 'Z':
+		case r >= '0' && r <= '9':
+		case r == '-' || r == '_' || r == '.':
+		default:
+			return fmt.Errorf("%w: invalid character", errInvalidOAuthState)
+		}
+	}
+	return nil
+}
+
+func NormalizeOAuthProvider(provider string) (string, error) {
+	switch strings.ToLower(strings.TrimSpace(provider)) {
+	case "anthropic", "claude":
+		return "anthropic", nil
+	case "codex", "openai":
+		return "codex", nil
+	case "gemini", "google":
+		return "gemini", nil
+	case "iflow", "i-flow":
+		return "iflow", nil
+	case "antigravity", "anti-gravity":
+		return "antigravity", nil
+	case "qwen":
+		return "qwen", nil
+	default:
+		return "", errUnsupportedOAuthFlow
+	}
+}
+
+type oauthCallbackFilePayload struct {
+	Code  string `json:"code"`
+	State string `json:"state"`
+	Error string `json:"error"`
+}
+
+func WriteOAuthCallbackFile(authDir, provider, state, code, errorMessage string) (string, error) {
+	if strings.TrimSpace(authDir) == "" {
+		return "", fmt.Errorf("auth dir is empty")
+	}
+	canonicalProvider, err := NormalizeOAuthProvider(provider)
+	if err != nil {
+		return "", err
+	}
+	if err := ValidateOAuthState(state); err != nil {
+		return "", err
+	}
+
+	fileName := fmt.Sprintf(".oauth-%s-%s.oauth", canonicalProvider, state)
+	filePath := filepath.Join(authDir, fileName)
+	payload := oauthCallbackFilePayload{
+		Code:  strings.TrimSpace(code),
+		State: strings.TrimSpace(state),
+		Error: strings.TrimSpace(errorMessage),
+	}
+	data, err := json.Marshal(payload)
+	if err != nil {
+		return "", fmt.Errorf("marshal oauth callback payload: %w", err)
+	}
+	if err := os.WriteFile(filePath, data, 0o600); err != nil {
+		return "", fmt.Errorf("write oauth callback file: %w", err)
+	}
+	return filePath, nil
+}
+
+func WriteOAuthCallbackFileForPendingSession(authDir, provider, state, code, errorMessage string) (string, error) {
+	canonicalProvider, err := NormalizeOAuthProvider(provider)
+	if err != nil {
+		return "", err
+	}
+	if !IsOAuthSessionPending(state, canonicalProvider) {
+		return "", errOAuthSessionNotPending
+	}
+	return WriteOAuthCallbackFile(authDir, canonicalProvider, state, code, errorMessage)
+}

internal/api/handlers/management/usage.go

@@ -1,12 +1,25 @@
 package management
 
 import (
+	"encoding/json"
 	"net/http"
+	"time"
 
 	"github.com/gin-gonic/gin"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
 )
 
+type usageExportPayload struct {
+	Version    int                      `json:"version"`
+	ExportedAt time.Time                `json:"exported_at"`
+	Usage      usage.StatisticsSnapshot `json:"usage"`
+}
+
+type usageImportPayload struct {
+	Version int                      `json:"version"`
+	Usage   usage.StatisticsSnapshot `json:"usage"`
+}
+
 // GetUsageStatistics returns the in-memory request statistics snapshot.
 func (h *Handler) GetUsageStatistics(c *gin.Context) {
 	var snapshot usage.StatisticsSnapshot
@@ -18,3 +31,49 @@ func (h *Handler) GetUsageStatistics(c *gin.Context) {
 		"failed_requests": snapshot.FailureCount,
 	})
 }
+
+// ExportUsageStatistics returns a complete usage snapshot for backup/migration.
+func (h *Handler) ExportUsageStatistics(c *gin.Context) {
+	var snapshot usage.StatisticsSnapshot
+	if h != nil && h.usageStats != nil {
+		snapshot = h.usageStats.Snapshot()
+	}
+	c.JSON(http.StatusOK, usageExportPayload{
+		Version:    1,
+		ExportedAt: time.Now().UTC(),
+		Usage:      snapshot,
+	})
+}
+
+// ImportUsageStatistics merges a previously exported usage snapshot into memory.
+func (h *Handler) ImportUsageStatistics(c *gin.Context) {
+	if h == nil || h.usageStats == nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "usage statistics unavailable"})
+		return
+	}
+
+	data, err := c.GetRawData()
+	if err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "failed to read request body"})
+		return
+	}
+
+	var payload usageImportPayload
+	if err := json.Unmarshal(data, &payload); err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "invalid json"})
+		return
+	}
+	if payload.Version != 0 && payload.Version != 1 {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "unsupported version"})
+		return
+	}
+
+	result := h.usageStats.MergeSnapshot(payload.Usage)
+	snapshot := h.usageStats.Snapshot()
+	c.JSON(http.StatusOK, gin.H{
+		"added":           result.Added,
+		"skipped":         result.Skipped,
+		"total_requests":  snapshot.TotalRequests,
+		"failed_requests": snapshot.FailureCount,
+	})
+}

internal/api/middleware/request_logging.go

@@ -8,6 +8,7 @@ import (
 	"io"
 	"net/http"
 	"strings"
+	"time"
 
 	"github.com/gin-gonic/gin"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
@@ -102,6 +103,8 @@ func captureRequestInfo(c *gin.Context) (*RequestInfo, error) {
 		Method:    method,
 		Headers:   headers,
 		Body:      body,
+		RequestID: logging.GetGinRequestID(c),
+		Timestamp: time.Now(),
 	}, nil
 }
 

internal/api/middleware/response_writer.go

@@ -7,6 +7,7 @@ import (
 	"bytes"
 	"net/http"
 	"strings"
+	"time"
 
 	"github.com/gin-gonic/gin"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/interfaces"
@@ -19,6 +20,8 @@ type RequestInfo struct {
 	Method    string              // Method is the HTTP method (e.g., GET, POST).
 	Headers   map[string][]string // Headers contains the request headers.
 	Body      []byte              // Body is the raw request body.
+	RequestID string              // RequestID is the unique identifier for the request.
+	Timestamp time.Time           // Timestamp is when the request was received.
 }
 
 // ResponseWriterWrapper wraps the standard gin.ResponseWriter to intercept and log response data.
@@ -35,6 +38,7 @@ type ResponseWriterWrapper struct {
 	statusCode          int                        // statusCode stores the HTTP status code of the response.
 	headers             map[string][]string        // headers stores the response headers.
 	logOnErrorOnly      bool                       // logOnErrorOnly enables logging only when an error response is detected.
+	firstChunkTimestamp time.Time                  // firstChunkTimestamp captures TTFB for streaming responses.
 }
 
 // NewResponseWriterWrapper creates and initializes a new ResponseWriterWrapper.
@@ -71,22 +75,72 @@ func (w *ResponseWriterWrapper) Write(data []byte) (int, error) {
 	n, err := w.ResponseWriter.Write(data)
 
 	// THEN: Handle logging based on response type
-	if w.isStreaming {
+	if w.isStreaming && w.chunkChannel != nil {
+		// Capture TTFB on first chunk (synchronous, before async channel send)
+		if w.firstChunkTimestamp.IsZero() {
+			w.firstChunkTimestamp = time.Now()
+		}
 		// For streaming responses: Send to async logging channel (non-blocking)
-		if w.chunkChannel != nil {
 		select {
 		case w.chunkChannel <- append([]byte(nil), data...): // Non-blocking send with copy
 		default: // Channel full, skip logging to avoid blocking
 		}
+		return n, err
 	}
-	} else {
-		// For non-streaming responses: Buffer complete response
+
+	if w.shouldBufferResponseBody() {
 		w.body.Write(data)
 	}
 
 	return n, err
 }
 
+func (w *ResponseWriterWrapper) shouldBufferResponseBody() bool {
+	if w.logger != nil && w.logger.IsEnabled() {
+		return true
+	}
+	if !w.logOnErrorOnly {
+		return false
+	}
+	status := w.statusCode
+	if status == 0 {
+		if statusWriter, ok := w.ResponseWriter.(interface{ Status() int }); ok && statusWriter != nil {
+			status = statusWriter.Status()
+		} else {
+			status = http.StatusOK
+		}
+	}
+	return status >= http.StatusBadRequest
+}
+
+// WriteString wraps the underlying ResponseWriter's WriteString method to capture response data.
+// Some handlers (and fmt/io helpers) write via io.StringWriter; without this override, those writes
+// bypass Write() and would be missing from request logs.
+func (w *ResponseWriterWrapper) WriteString(data string) (int, error) {
+	w.ensureHeadersCaptured()
+
+	// CRITICAL: Write to client first (zero latency)
+	n, err := w.ResponseWriter.WriteString(data)
+
+	// THEN: Capture for logging
+	if w.isStreaming && w.chunkChannel != nil {
+		// Capture TTFB on first chunk (synchronous, before async channel send)
+		if w.firstChunkTimestamp.IsZero() {
+			w.firstChunkTimestamp = time.Now()
+		}
+		select {
+		case w.chunkChannel <- []byte(data):
+		default:
+		}
+		return n, err
+	}
+
+	if w.shouldBufferResponseBody() {
+		w.body.WriteString(data)
+	}
+	return n, err
+}
+
 // WriteHeader wraps the underlying ResponseWriter's WriteHeader method.
 // It captures the status code, detects if the response is streaming based on the Content-Type header,
 // and initializes the appropriate logging mechanism (standard or streaming).
@@ -107,6 +161,7 @@ func (w *ResponseWriterWrapper) WriteHeader(statusCode int) {
 			w.requestInfo.Method,
 			w.requestInfo.Headers,
 			w.requestInfo.Body,
+			w.requestInfo.RequestID,
 		)
 		if err == nil {
 			w.streamWriter = streamWriter
@@ -160,12 +215,16 @@ func (w *ResponseWriterWrapper) detectStreaming(contentType string) bool {
 		return true
 	}
 
-	// Check request body for streaming indicators
-	if w.requestInfo.Body != nil {
-		bodyStr := string(w.requestInfo.Body)
-		if strings.Contains(bodyStr, `"stream": true`) || strings.Contains(bodyStr, `"stream":true`) {
-			return true
+	// If a concrete Content-Type is already set (e.g., application/json for error responses),
+	// treat it as non-streaming instead of inferring from the request payload.
+	if strings.TrimSpace(contentType) != "" {
+		return false
 	}
+
+	// Only fall back to request payload hints when Content-Type is not set yet.
+	if w.requestInfo != nil && len(w.requestInfo.Body) > 0 {
+		bodyStr := string(w.requestInfo.Body)
+		return strings.Contains(bodyStr, `"stream": true`) || strings.Contains(bodyStr, `"stream":true`)
 	}
 
 	return false
@@ -221,7 +280,7 @@ func (w *ResponseWriterWrapper) Finalize(c *gin.Context) error {
 		return nil
 	}
 
-	if w.isStreaming {
+	if w.isStreaming && w.streamWriter != nil {
 		if w.chunkChannel != nil {
 			close(w.chunkChannel)
 			w.chunkChannel = nil
@@ -232,8 +291,9 @@ func (w *ResponseWriterWrapper) Finalize(c *gin.Context) error {
 			w.streamDone = nil
 		}
 
+		w.streamWriter.SetFirstChunkTimestamp(w.firstChunkTimestamp)
+
 		// Write API Request and Response to the streaming log before closing
-		if w.streamWriter != nil {
 		apiRequest := w.extractAPIRequest(c)
 		if len(apiRequest) > 0 {
 			_ = w.streamWriter.WriteAPIRequest(apiRequest)
@@ -247,14 +307,10 @@ func (w *ResponseWriterWrapper) Finalize(c *gin.Context) error {
 			return err
 		}
 		w.streamWriter = nil
-		}
-		if forceLog {
-			return w.logRequest(finalStatusCode, w.cloneHeaders(), w.body.Bytes(), w.extractAPIRequest(c), w.extractAPIResponse(c), slicesAPIResponseError, forceLog)
-		}
 		return nil
 	}
 
-	return w.logRequest(finalStatusCode, w.cloneHeaders(), w.body.Bytes(), w.extractAPIRequest(c), w.extractAPIResponse(c), slicesAPIResponseError, forceLog)
+	return w.logRequest(finalStatusCode, w.cloneHeaders(), w.body.Bytes(), w.extractAPIRequest(c), w.extractAPIResponse(c), w.extractAPIResponseTimestamp(c), slicesAPIResponseError, forceLog)
 }
 
 func (w *ResponseWriterWrapper) cloneHeaders() map[string][]string {
@@ -294,7 +350,18 @@ func (w *ResponseWriterWrapper) extractAPIResponse(c *gin.Context) []byte {
 	return data
 }
 
-func (w *ResponseWriterWrapper) logRequest(statusCode int, headers map[string][]string, body []byte, apiRequestBody, apiResponseBody []byte, apiResponseErrors []*interfaces.ErrorMessage, forceLog bool) error {
+func (w *ResponseWriterWrapper) extractAPIResponseTimestamp(c *gin.Context) time.Time {
+	ts, isExist := c.Get("API_RESPONSE_TIMESTAMP")
+	if !isExist {
+		return time.Time{}
+	}
+	if t, ok := ts.(time.Time); ok {
+		return t
+	}
+	return time.Time{}
+}
+
+func (w *ResponseWriterWrapper) logRequest(statusCode int, headers map[string][]string, body []byte, apiRequestBody, apiResponseBody []byte, apiResponseTimestamp time.Time, apiResponseErrors []*interfaces.ErrorMessage, forceLog bool) error {
 	if w.requestInfo == nil {
 		return nil
 	}
@@ -305,7 +372,7 @@ func (w *ResponseWriterWrapper) logRequest(statusCode int, headers map[string][]
 	}
 
 	if loggerWithOptions, ok := w.logger.(interface {
-		LogRequestWithOptions(string, string, map[string][]string, []byte, int, map[string][]string, []byte, []byte, []byte, []*interfaces.ErrorMessage, bool) error
+		LogRequestWithOptions(string, string, map[string][]string, []byte, int, map[string][]string, []byte, []byte, []byte, []*interfaces.ErrorMessage, bool, string, time.Time, time.Time) error
 	}); ok {
 		return loggerWithOptions.LogRequestWithOptions(
 			w.requestInfo.URL,
@@ -319,6 +386,9 @@ func (w *ResponseWriterWrapper) logRequest(statusCode int, headers map[string][]
 			apiResponseBody,
 			apiResponseErrors,
 			forceLog,
+			w.requestInfo.RequestID,
+			w.requestInfo.Timestamp,
+			apiResponseTimestamp,
 		)
 	}
 
@@ -333,28 +403,8 @@ func (w *ResponseWriterWrapper) logRequest(statusCode int, headers map[string][]
 		apiRequestBody,
 		apiResponseBody,
 		apiResponseErrors,
+		w.requestInfo.RequestID,
+		w.requestInfo.Timestamp,
+		apiResponseTimestamp,
 	)
 }
-
-// Status returns the HTTP response status code captured by the wrapper.
-// It defaults to 200 if WriteHeader has not been called.
-func (w *ResponseWriterWrapper) Status() int {
-	if w.statusCode == 0 {
-		return 200 // Default status code
-	}
-	return w.statusCode
-}
-
-// Size returns the size of the response body in bytes for non-streaming responses.
-// For streaming responses, it returns -1, as the total size is unknown.
-func (w *ResponseWriterWrapper) Size() int {
-	if w.isStreaming {
-		return -1 // Unknown size for streaming responses
-	}
-	return w.body.Len()
-}
-
-// Written returns true if the response header has been written (i.e., a status code has been set).
-func (w *ResponseWriterWrapper) Written() bool {
-	return w.statusCode != 0
-}

internal/api/modules/amp/amp.go

@@ -125,6 +125,8 @@ func (m *AmpModule) Register(ctx modules.Context) error {
 	m.registerOnce.Do(func() {
 		// Initialize model mapper from config (for routing unavailable models to alternatives)
 		m.modelMapper = NewModelMapper(settings.ModelMappings)
+		// Load oauth-model-alias for provider lookup via aliases
+		m.modelMapper.UpdateOAuthModelAlias(ctx.Config.OAuthModelAlias)
 
 		// Store initial config for partial reload comparison
 		settingsCopy := settings
@@ -137,7 +139,8 @@ func (m *AmpModule) Register(ctx modules.Context) error {
 		m.registerProviderAliases(ctx.Engine, ctx.BaseHandler, auth)
 
 		// Register management proxy routes once; middleware will gate access when upstream is unavailable.
-		m.registerManagementRoutes(ctx.Engine, ctx.BaseHandler)
+		// Pass auth middleware to require valid API key for all management routes.
+		m.registerManagementRoutes(ctx.Engine, ctx.BaseHandler, auth)
 
 		// If no upstream URL, skip proxy routes but provider aliases are still available
 		if upstreamURL == "" {
@@ -187,9 +190,6 @@ func (m *AmpModule) OnConfigUpdated(cfg *config.Config) error {
 
 	if oldSettings != nil && oldSettings.RestrictManagementToLocalhost != newSettings.RestrictManagementToLocalhost {
 		m.setRestrictToLocalhost(newSettings.RestrictManagementToLocalhost)
-		if !newSettings.RestrictManagementToLocalhost {
-			log.Warnf("amp management routes now accessible from any IP - this is insecure!")
-		}
 	}
 
 	newUpstreamURL := strings.TrimSpace(newSettings.UpstreamURL)
@@ -214,6 +214,11 @@ func (m *AmpModule) OnConfigUpdated(cfg *config.Config) error {
 		}
 	}
 
+	// Always update oauth-model-alias for model mapper (used for provider lookup)
+	if m.modelMapper != nil {
+		m.modelMapper.UpdateOAuthModelAlias(cfg.OAuthModelAlias)
+	}
+
 	if m.enabled {
 		// Check upstream URL change - now supports hot-reload
 		if newUpstreamURL == "" && oldUpstreamURL != "" {
@@ -229,11 +234,20 @@ func (m *AmpModule) OnConfigUpdated(cfg *config.Config) error {
 			}
 		}
 
-		// Check API key change
+		// Check API key change (both default and per-client mappings)
 		apiKeyChanged := m.hasAPIKeyChanged(oldSettings, &newSettings)
-		if apiKeyChanged {
+		upstreamAPIKeysChanged := m.hasUpstreamAPIKeysChanged(oldSettings, &newSettings)
+		if apiKeyChanged || upstreamAPIKeysChanged {
 			if m.secretSource != nil {
-				if ms, ok := m.secretSource.(*MultiSourceSecret); ok {
+				if ms, ok := m.secretSource.(*MappedSecretSource); ok {
+					if apiKeyChanged {
+						ms.UpdateDefaultExplicitKey(newSettings.UpstreamAPIKey)
+						ms.InvalidateCache()
+					}
+					if upstreamAPIKeysChanged {
+						ms.UpdateMappings(newSettings.UpstreamAPIKeys)
+					}
+				} else if ms, ok := m.secretSource.(*MultiSourceSecret); ok {
 					ms.UpdateExplicitKey(newSettings.UpstreamAPIKey)
 					ms.InvalidateCache()
 				}
@@ -253,10 +267,22 @@ func (m *AmpModule) OnConfigUpdated(cfg *config.Config) error {
 
 func (m *AmpModule) enableUpstreamProxy(upstreamURL string, settings *config.AmpCode) error {
 	if m.secretSource == nil {
-		m.secretSource = NewMultiSourceSecret(settings.UpstreamAPIKey, 0 /* default 5min */)
+		// Create MultiSourceSecret as the default source, then wrap with MappedSecretSource
+		defaultSource := NewMultiSourceSecret(settings.UpstreamAPIKey, 0 /* default 5min */)
+		mappedSource := NewMappedSecretSource(defaultSource)
+		mappedSource.UpdateMappings(settings.UpstreamAPIKeys)
+		m.secretSource = mappedSource
+	} else if ms, ok := m.secretSource.(*MappedSecretSource); ok {
+		ms.UpdateDefaultExplicitKey(settings.UpstreamAPIKey)
+		ms.InvalidateCache()
+		ms.UpdateMappings(settings.UpstreamAPIKeys)
 	} else if ms, ok := m.secretSource.(*MultiSourceSecret); ok {
+		// Legacy path: wrap existing MultiSourceSecret with MappedSecretSource
 		ms.UpdateExplicitKey(settings.UpstreamAPIKey)
 		ms.InvalidateCache()
+		mappedSource := NewMappedSecretSource(ms)
+		mappedSource.UpdateMappings(settings.UpstreamAPIKeys)
+		m.secretSource = mappedSource
 	}
 
 	proxy, err := createReverseProxy(upstreamURL, m.secretSource)
@@ -281,16 +307,23 @@ func (m *AmpModule) hasModelMappingsChanged(old *config.AmpCode, new *config.Amp
 		return true
 	}
 
-	// Build map for efficient comparison
-	oldMap := make(map[string]string, len(old.ModelMappings))
+	// Build map for efficient and robust comparison
+	type mappingInfo struct {
+		to    string
+		regex bool
+	}
+	oldMap := make(map[string]mappingInfo, len(old.ModelMappings))
 	for _, mapping := range old.ModelMappings {
-		oldMap[strings.TrimSpace(mapping.From)] = strings.TrimSpace(mapping.To)
+		oldMap[strings.TrimSpace(mapping.From)] = mappingInfo{
+			to:    strings.TrimSpace(mapping.To),
+			regex: mapping.Regex,
+		}
 	}
 
 	for _, mapping := range new.ModelMappings {
 		from := strings.TrimSpace(mapping.From)
 		to := strings.TrimSpace(mapping.To)
-		if oldTo, exists := oldMap[from]; !exists || oldTo != to {
+		if oldVal, exists := oldMap[from]; !exists || oldVal.to != to || oldVal.regex != mapping.Regex {
 			return true
 		}
 	}
@@ -308,6 +341,66 @@ func (m *AmpModule) hasAPIKeyChanged(old *config.AmpCode, new *config.AmpCode) b
 	return oldKey != newKey
 }
 
+// hasUpstreamAPIKeysChanged compares old and new per-client upstream API key mappings.
+func (m *AmpModule) hasUpstreamAPIKeysChanged(old *config.AmpCode, new *config.AmpCode) bool {
+	if old == nil {
+		return len(new.UpstreamAPIKeys) > 0
+	}
+
+	if len(old.UpstreamAPIKeys) != len(new.UpstreamAPIKeys) {
+		return true
+	}
+
+	// Build map for comparison: upstreamKey -> set of clientKeys
+	type entryInfo struct {
+		upstreamKey string
+		clientKeys  map[string]struct{}
+	}
+	oldEntries := make([]entryInfo, len(old.UpstreamAPIKeys))
+	for i, entry := range old.UpstreamAPIKeys {
+		clientKeys := make(map[string]struct{}, len(entry.APIKeys))
+		for _, k := range entry.APIKeys {
+			trimmed := strings.TrimSpace(k)
+			if trimmed == "" {
+				continue
+			}
+			clientKeys[trimmed] = struct{}{}
+		}
+		oldEntries[i] = entryInfo{
+			upstreamKey: strings.TrimSpace(entry.UpstreamAPIKey),
+			clientKeys:  clientKeys,
+		}
+	}
+
+	for i, newEntry := range new.UpstreamAPIKeys {
+		if i >= len(oldEntries) {
+			return true
+		}
+		oldE := oldEntries[i]
+		if strings.TrimSpace(newEntry.UpstreamAPIKey) != oldE.upstreamKey {
+			return true
+		}
+		newKeys := make(map[string]struct{}, len(newEntry.APIKeys))
+		for _, k := range newEntry.APIKeys {
+			trimmed := strings.TrimSpace(k)
+			if trimmed == "" {
+				continue
+			}
+			newKeys[trimmed] = struct{}{}
+		}
+		if len(newKeys) != len(oldE.clientKeys) {
+			return true
+		}
+		for k := range newKeys {
+			if _, ok := oldE.clientKeys[k]; !ok {
+				return true
+			}
+		}
+	}
+
+	return false
+}
+
 // GetModelMapper returns the model mapper instance (for testing/debugging).
 func (m *AmpModule) GetModelMapper() *DefaultModelMapper {
 	return m.modelMapper

internal/api/modules/amp/amp_test.go

@@ -146,6 +146,9 @@ func TestAmpModule_OnConfigUpdated_CacheInvalidation(t *testing.T) {
 	m := &AmpModule{enabled: true}
 	ms := NewMultiSourceSecretWithPath("", p, time.Minute)
 	m.secretSource = ms
+	m.lastConfig = &config.AmpCode{
+		UpstreamAPIKey: "old-key",
+	}
 
 	// Warm the cache
 	if _, err := ms.Get(context.Background()); err != nil {
@@ -157,7 +160,7 @@ func TestAmpModule_OnConfigUpdated_CacheInvalidation(t *testing.T) {
 	}
 
 	// Update config - should invalidate cache
-	if err := m.OnConfigUpdated(&config.Config{AmpCode: config.AmpCode{UpstreamURL: "http://x"}}); err != nil {
+	if err := m.OnConfigUpdated(&config.Config{AmpCode: config.AmpCode{UpstreamURL: "http://x", UpstreamAPIKey: "new-key"}}); err != nil {
 		t.Fatal(err)
 	}
 
@@ -309,3 +312,41 @@ func TestAmpModule_ProviderAliasesAlwaysRegistered(t *testing.T) {
 		})
 	}
 }
+
+func TestAmpModule_hasUpstreamAPIKeysChanged_DetectsRemovedKeyWithDuplicateInput(t *testing.T) {
+	m := &AmpModule{}
+
+	oldCfg := &config.AmpCode{
+		UpstreamAPIKeys: []config.AmpUpstreamAPIKeyEntry{
+			{UpstreamAPIKey: "u1", APIKeys: []string{"k1", "k2"}},
+		},
+	}
+	newCfg := &config.AmpCode{
+		UpstreamAPIKeys: []config.AmpUpstreamAPIKeyEntry{
+			{UpstreamAPIKey: "u1", APIKeys: []string{"k1", "k1"}},
+		},
+	}
+
+	if !m.hasUpstreamAPIKeysChanged(oldCfg, newCfg) {
+		t.Fatal("expected change to be detected when k2 is removed but new list contains duplicates")
+	}
+}
+
+func TestAmpModule_hasUpstreamAPIKeysChanged_IgnoresEmptyAndWhitespaceKeys(t *testing.T) {
+	m := &AmpModule{}
+
+	oldCfg := &config.AmpCode{
+		UpstreamAPIKeys: []config.AmpUpstreamAPIKeyEntry{
+			{UpstreamAPIKey: "u1", APIKeys: []string{"k1", "k2"}},
+		},
+	}
+	newCfg := &config.AmpCode{
+		UpstreamAPIKeys: []config.AmpUpstreamAPIKeyEntry{
+			{UpstreamAPIKey: "u1", APIKeys: []string{"  k1  ", "", "k2", "   "}},
+		},
+	}
+
+	if m.hasUpstreamAPIKeysChanged(oldCfg, newCfg) {
+		t.Fatal("expected no change when only whitespace/empty entries differ")
+	}
+}

internal/api/modules/amp/fallback_handlers.go

@@ -2,12 +2,16 @@ package amp
 
 import (
 	"bytes"
+	"errors"
 	"io"
+	"net/http"
 	"net/http/httputil"
 	"strings"
 	"time"
 
 	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/routing/ctxkeys"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	log "github.com/sirupsen/logrus"
 	"github.com/tidwall/gjson"
@@ -29,7 +33,13 @@ const (
 )
 
 // MappedModelContextKey is the Gin context key for passing mapped model names.
-const MappedModelContextKey = "mapped_model"
+// Deprecated: Use ctxkeys.MappedModel instead.
+const MappedModelContextKey = string(ctxkeys.MappedModel)
+
+// FallbackModelsContextKey is the Gin context key for passing fallback model names.
+// When the primary mapped model fails (e.g., quota exceeded), these models can be tried.
+// Deprecated: Use ctxkeys.FallbackModels instead.
+const FallbackModelsContextKey = string(ctxkeys.FallbackModels)
 
 // logAmpRouting logs the routing decision for an Amp request with structured fields
 func logAmpRouting(routeType AmpRouteType, requestedModel, resolvedModel, provider, path string) {
@@ -64,7 +74,7 @@ func logAmpRouting(routeType AmpRouteType, requestedModel, resolvedModel, provid
 		fields["cost"] = "amp_credits"
 		fields["source"] = "ampcode.com"
 		fields["model_id"] = requestedModel // Explicit model_id for easy config reference
-		log.WithFields(fields).Warnf("forwarding to ampcode.com (uses amp credits) - model_id: %s | To use local proxy, add to config: amp-model-mappings: [{from: \"%s\", to: \"<your-local-model>\"}]", requestedModel, requestedModel)
+		log.WithFields(fields).Warnf("forwarding to ampcode.com (uses amp credits) - model_id: %s | To use local provider, add to config: ampcode.model-mappings: [{from: \"%s\", to: \"<your-local-model>\"}]", requestedModel, requestedModel)
 
 	case RouteTypeNoProvider:
 		fields["cost"] = "none"
@@ -76,6 +86,10 @@ func logAmpRouting(routeType AmpRouteType, requestedModel, resolvedModel, provid
 
 // FallbackHandler wraps a standard handler with fallback logic to ampcode.com
 // when the model's provider is not available in CLIProxyAPI
+//
+// Deprecated: FallbackHandler is deprecated in favor of routing.ModelRoutingWrapper.
+// Use routing.NewModelRoutingWrapper() instead for unified routing logic.
+// This type is kept for backward compatibility and test purposes.
 type FallbackHandler struct {
 	getProxy           func() *httputil.ReverseProxy
 	modelMapper        ModelMapper
@@ -84,6 +98,8 @@ type FallbackHandler struct {
 
 // NewFallbackHandler creates a new fallback handler wrapper
 // The getProxy function allows lazy evaluation of the proxy (useful when proxy is created after routes)
+//
+// Deprecated: Use routing.NewModelRoutingWrapper() instead.
 func NewFallbackHandler(getProxy func() *httputil.ReverseProxy) *FallbackHandler {
 	return &FallbackHandler{
 		getProxy:           getProxy,
@@ -92,6 +108,8 @@ func NewFallbackHandler(getProxy func() *httputil.ReverseProxy) *FallbackHandler
 }
 
 // NewFallbackHandlerWithMapper creates a new fallback handler with model mapping support
+//
+// Deprecated: Use routing.NewModelRoutingWrapper() instead.
 func NewFallbackHandlerWithMapper(getProxy func() *httputil.ReverseProxy, mapper ModelMapper, forceModelMappings func() bool) *FallbackHandler {
 	if forceModelMappings == nil {
 		forceModelMappings = func() bool { return false }
@@ -112,6 +130,20 @@ func (fh *FallbackHandler) SetModelMapper(mapper ModelMapper) {
 // If the model's provider is not configured in CLIProxyAPI, it forwards to ampcode.com
 func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc {
 	return func(c *gin.Context) {
+		// Swallow ErrAbortHandler panics from ReverseProxy to avoid noisy stack traces.
+		// ReverseProxy raises this panic when the client connection is closed prematurely
+		// (e.g., user cancels request, network disconnect) or when ServeHTTP is called
+		// with a ResponseWriter that doesn't implement http.CloseNotifier.
+		// This is an expected error condition, not a bug, so we handle it gracefully.
+		defer func() {
+			if rec := recover(); rec != nil {
+				if err, ok := rec.(error); ok && errors.Is(err, http.ErrAbortHandler) {
+					return
+				}
+				panic(rec)
+			}
+		}()
+
 		requestPath := c.Request.URL.Path
 
 		// Read the request body to extract the model name
@@ -134,34 +166,92 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
 		}
 
 		// Normalize model (handles dynamic thinking suffixes)
-		normalizedModel, _ := util.NormalizeThinkingModel(modelName)
+		suffixResult := thinking.ParseSuffix(modelName)
+		normalizedModel := suffixResult.ModelName
+		thinkingSuffix := ""
+		if suffixResult.HasSuffix {
+			thinkingSuffix = "(" + suffixResult.RawSuffix + ")"
+		}
+
+		// resolveMappedModels returns all mapped models (primary + fallbacks) and providers for the first one.
+		resolveMappedModels := func() ([]string, []string) {
+			if fh.modelMapper == nil {
+				return nil, nil
+			}
+
+			mapper, ok := fh.modelMapper.(*DefaultModelMapper)
+			if !ok {
+				// Fallback to single model for non-DefaultModelMapper
+				mappedModel := fh.modelMapper.MapModel(modelName)
+				if mappedModel == "" {
+					mappedModel = fh.modelMapper.MapModel(normalizedModel)
+				}
+				if mappedModel == "" {
+					return nil, nil
+				}
+				mappedBaseModel := thinking.ParseSuffix(mappedModel).ModelName
+				mappedProviders := util.GetProviderName(mappedBaseModel)
+				if len(mappedProviders) == 0 {
+					return nil, nil
+				}
+				return []string{mappedModel}, mappedProviders
+			}
+
+			// Use MapModelWithFallbacks for DefaultModelMapper
+			mappedModels := mapper.MapModelWithFallbacks(modelName)
+			if len(mappedModels) == 0 {
+				mappedModels = mapper.MapModelWithFallbacks(normalizedModel)
+			}
+			if len(mappedModels) == 0 {
+				return nil, nil
+			}
+
+			// Apply thinking suffix if needed
+			for i, model := range mappedModels {
+				if thinkingSuffix != "" {
+					suffixResult := thinking.ParseSuffix(model)
+					if !suffixResult.HasSuffix {
+						mappedModels[i] = model + thinkingSuffix
+					}
+				}
+			}
+
+			// Get providers for the first model
+			firstBaseModel := thinking.ParseSuffix(mappedModels[0]).ModelName
+			providers := util.GetProviderName(firstBaseModel)
+			if len(providers) == 0 {
+				return nil, nil
+			}
+
+			return mappedModels, providers
+		}
 
 		// Track resolved model for logging (may change if mapping is applied)
 		resolvedModel := normalizedModel
 		usedMapping := false
 		var providers []string
 
+		// Helper to apply model mapping and update state
+		applyMapping := func(mappedModels []string, mappedProviders []string) {
+			bodyBytes = rewriteModelInRequest(bodyBytes, mappedModels[0])
+			c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
+			c.Set(string(ctxkeys.MappedModel), mappedModels[0])
+			if len(mappedModels) > 1 {
+				c.Set(string(ctxkeys.FallbackModels), mappedModels[1:])
+			}
+			resolvedModel = mappedModels[0]
+			usedMapping = true
+			providers = mappedProviders
+		}
+
 		// Check if model mappings should be forced ahead of local API keys
 		forceMappings := fh.forceModelMappings != nil && fh.forceModelMappings()
 
 		if forceMappings {
 			// FORCE MODE: Check model mappings FIRST (takes precedence over local API keys)
 			// This allows users to route Amp requests to their preferred OAuth providers
-			if fh.modelMapper != nil {
-				if mappedModel := fh.modelMapper.MapModel(normalizedModel); mappedModel != "" {
-					// Mapping found - check if we have a provider for the mapped model
-					mappedProviders := util.GetProviderName(mappedModel)
-					if len(mappedProviders) > 0 {
-						// Mapping found and provider available - rewrite the model in request body
-						bodyBytes = rewriteModelInRequest(bodyBytes, mappedModel)
-						c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
-						// Store mapped model in context for handlers that check it (like gemini bridge)
-						c.Set(MappedModelContextKey, mappedModel)
-						resolvedModel = mappedModel
-						usedMapping = true
-						providers = mappedProviders
-					}
-				}
+			if mappedModels, mappedProviders := resolveMappedModels(); len(mappedModels) > 0 {
+				applyMapping(mappedModels, mappedProviders)
 			}
 
 			// If no mapping applied, check for local providers
@@ -174,21 +264,8 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
 
 			if len(providers) == 0 {
 				// No providers configured - check if we have a model mapping
-				if fh.modelMapper != nil {
-					if mappedModel := fh.modelMapper.MapModel(normalizedModel); mappedModel != "" {
-						// Mapping found - check if we have a provider for the mapped model
-						mappedProviders := util.GetProviderName(mappedModel)
-						if len(mappedProviders) > 0 {
-							// Mapping found and provider available - rewrite the model in request body
-							bodyBytes = rewriteModelInRequest(bodyBytes, mappedModel)
-							c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
-							// Store mapped model in context for handlers that check it (like gemini bridge)
-							c.Set(MappedModelContextKey, mappedModel)
-							resolvedModel = mappedModel
-							usedMapping = true
-							providers = mappedProviders
-						}
-					}
+				if mappedModels, mappedProviders := resolveMappedModels(); len(mappedModels) > 0 {
+					applyMapping(mappedModels, mappedProviders)
 				}
 			}
 		}
@@ -222,14 +299,14 @@ func (fh *FallbackHandler) WrapHandler(handler gin.HandlerFunc) gin.HandlerFunc
 			// Log: Model was mapped to another model
 			log.Debugf("amp model mapping: request %s -> %s", normalizedModel, resolvedModel)
 			logAmpRouting(RouteTypeModelMapping, modelName, resolvedModel, providerName, requestPath)
-			rewriter := NewResponseRewriter(c.Writer, normalizedModel)
+			rewriter := NewResponseRewriter(c.Writer, modelName)
 			c.Writer = rewriter
 			// Filter Anthropic-Beta header only for local handling paths
 			filterAntropicBetaHeader(c)
 			c.Request.Body = io.NopCloser(bytes.NewReader(bodyBytes))
 			handler(c)
 			rewriter.Flush()
-			log.Debugf("amp model mapping: response %s -> %s", resolvedModel, normalizedModel)
+			log.Debugf("amp model mapping: response %s -> %s", resolvedModel, modelName)
 		} else if len(providers) > 0 {
 			// Log: Using local provider (free)
 			logAmpRouting(RouteTypeLocalProvider, modelName, resolvedModel, providerName, requestPath)

internal/api/modules/amp/fallback_handlers_characterization_test.go

@@ -0,0 +1,326 @@
+package amp
+
+import (
+	"bytes"
+	"net/http"
+	"net/http/httptest"
+	"net/http/httputil"
+	"net/url"
+	"strings"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/routing/testutil"
+	"github.com/stretchr/testify/assert"
+)
+
+// Characterization tests for fallback_handlers.go using testutil recorders
+// These tests capture existing behavior before refactoring to routing layer
+
+func TestCharacterization_LocalProvider(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the test model
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("char-test-local", "anthropic", []*registry.ModelInfo{
+		{ID: "test-model-local"},
+	})
+	defer reg.UnregisterClient("char-test-local")
+
+	// Setup recorders
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create gin context
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	body := `{"model": "test-model-local", "messages": [{"role": "user", "content": "hello"}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/anthropic/v1/messages", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Create fallback handler with proxy recorder
+	// Create a test server to act as the proxy target
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		// Create a reverse proxy that forwards to our test server
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	})
+
+	// Execute
+	wrapped := fh.WrapHandler(handlerRecorder.GinHandler())
+	wrapped(c)
+
+	// Assert: proxy NOT called
+	assert.False(t, proxyRecorder.Called, "proxy should NOT be called for local provider")
+
+	// Assert: local handler called once
+	assert.True(t, handlerRecorder.WasCalled(), "local handler should be called")
+	assert.Equal(t, 1, handlerRecorder.GetCallCount(), "local handler should be called exactly once")
+
+	// Assert: request body model unchanged
+	assert.Contains(t, string(handlerRecorder.RequestBody), "test-model-local", "request body model should be unchanged")
+}
+
+func TestCharacterization_ModelMapping(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the TARGET model (the mapped-to model)
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("char-test-mapped", "openai", []*registry.ModelInfo{
+		{ID: "gpt-4-local"},
+	})
+	defer reg.UnregisterClient("char-test-mapped")
+
+	// Setup recorders
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create model mapper with a mapping
+	mapper := NewModelMapper([]config.AmpModelMapping{
+		{From: "gpt-4-turbo", To: "gpt-4-local"},
+	})
+
+	// Create gin context
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	// Request with original model that gets mapped
+	body := `{"model": "gpt-4-turbo", "messages": [{"role": "user", "content": "hello"}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/openai/v1/chat/completions", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Create fallback handler with mapper
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	fh := NewFallbackHandlerWithMapper(func() *httputil.ReverseProxy {
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	}, mapper, func() bool { return false })
+
+	// Execute - use handler that returns model in response for rewriter to work
+	wrapped := fh.WrapHandler(handlerRecorder.GinHandlerWithModel())
+	wrapped(c)
+
+	// Assert: proxy NOT called
+	assert.False(t, proxyRecorder.Called, "proxy should NOT be called for model mapping")
+
+	// Assert: local handler called once
+	assert.True(t, handlerRecorder.WasCalled(), "local handler should be called")
+	assert.Equal(t, 1, handlerRecorder.GetCallCount(), "local handler should be called exactly once")
+
+	// Assert: request body model was rewritten to mapped model
+	assert.Contains(t, string(handlerRecorder.RequestBody), "gpt-4-local", "request body model should be rewritten to mapped model")
+	assert.NotContains(t, string(handlerRecorder.RequestBody), "gpt-4-turbo", "request body should NOT contain original model")
+
+	// Assert: context has mapped_model key set
+	mappedModel, exists := handlerRecorder.GetContextKey("mapped_model")
+	assert.True(t, exists, "context should have mapped_model key")
+	assert.Equal(t, "gpt-4-local", mappedModel, "mapped_model should be the target model")
+
+	// Assert: response body model rewritten back to original
+	// The response writer should rewrite model names in the response
+	responseBody := w.Body.String()
+	assert.Contains(t, responseBody, "gpt-4-turbo", "response should have original model name")
+}
+
+func TestCharacterization_AmpCreditsProxy(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Setup recorders - NO local provider registered, NO mapping configured
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create gin context with CloseNotifier support (required for ReverseProxy)
+	w := testutil.NewCloseNotifierRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	// Request with a model that has no local provider and no mapping
+	body := `{"model": "unknown-model-no-provider", "messages": [{"role": "user", "content": "hello"}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/openai/v1/chat/completions", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Create fallback handler
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	})
+
+	// Execute
+	wrapped := fh.WrapHandler(handlerRecorder.GinHandler())
+	wrapped(c)
+
+	// Assert: proxy called once
+	assert.True(t, proxyRecorder.Called, "proxy should be called when no local provider and no mapping")
+	assert.Equal(t, 1, proxyRecorder.GetCallCount(), "proxy should be called exactly once")
+
+	// Assert: local handler NOT called
+	assert.False(t, handlerRecorder.WasCalled(), "local handler should NOT be called when falling back to proxy")
+
+	// Assert: body forwarded to proxy is original (no rewrite)
+	assert.Contains(t, string(proxyRecorder.RequestBody), "unknown-model-no-provider", "request body model should be unchanged when proxying")
+}
+
+func TestCharacterization_BodyRestore(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the test model
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("char-test-body", "anthropic", []*registry.ModelInfo{
+		{ID: "test-model-body"},
+	})
+	defer reg.UnregisterClient("char-test-body")
+
+	// Setup recorders
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create gin context
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	// Create a complex request body that will be read by the wrapper for model extraction
+	originalBody := `{"model": "test-model-body", "messages": [{"role": "user", "content": "hello"}], "temperature": 0.7, "stream": true}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/anthropic/v1/messages", bytes.NewReader([]byte(originalBody)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Create fallback handler with proxy recorder
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	})
+
+	// Execute
+	wrapped := fh.WrapHandler(handlerRecorder.GinHandler())
+	wrapped(c)
+
+	// Assert: local handler called (not proxy, since we have a local provider)
+	assert.True(t, handlerRecorder.WasCalled(), "local handler should be called")
+	assert.False(t, proxyRecorder.Called, "proxy should NOT be called for local provider")
+
+	// Assert: handler receives complete original body
+	// This verifies that the body was properly restored after the wrapper read it for model extraction
+	assert.Equal(t, originalBody, string(handlerRecorder.RequestBody), "handler should receive complete original body after wrapper reads it for model extraction")
+}
+
+// TestCharacterization_GeminiV1Beta1_PostModels tests that POST requests with /models/ path use Gemini bridge handler
+// This is a characterization test for the route gating logic in routes.go
+func TestCharacterization_GeminiV1Beta1_PostModels(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the test model (Gemini format uses path-based model extraction)
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("char-test-gemini", "google", []*registry.ModelInfo{
+		{ID: "gemini-pro"},
+	})
+	defer reg.UnregisterClient("char-test-gemini")
+
+	// Setup recorders
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create a test server for the proxy
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	// Create fallback handler
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	})
+
+	// Create the Gemini bridge handler (simulating what routes.go does)
+	geminiBridge := createGeminiBridgeHandler(handlerRecorder.GinHandler())
+	geminiV1Beta1Handler := fh.WrapHandler(geminiBridge)
+
+	// Create router with the same gating logic as routes.go
+	r := gin.New()
+	r.Any("/api/provider/google/v1beta1/*path", func(c *gin.Context) {
+		if c.Request.Method == "POST" {
+			if path := c.Param("path"); strings.Contains(path, "/models/") {
+				// POST with /models/ path -> use Gemini bridge with fallback handler
+				geminiV1Beta1Handler(c)
+				return
+			}
+		}
+		// Non-POST or no /models/ in path -> proxy upstream
+		proxyRecorder.ServeHTTP(c.Writer, c.Request)
+	})
+
+	// Execute: POST request with /models/ in path
+	body := `{"contents": [{"role": "user", "parts": [{"text": "hello"}]}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/google/v1beta1/publishers/google/models/gemini-pro:generateContent", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	w := httptest.NewRecorder()
+	r.ServeHTTP(w, req)
+
+	// Assert: local Gemini handler called
+	assert.True(t, handlerRecorder.WasCalled(), "local Gemini handler should be called for POST /models/")
+
+	// Assert: proxy NOT called
+	assert.False(t, proxyRecorder.Called, "proxy should NOT be called for POST /models/ path")
+}
+
+// TestCharacterization_GeminiV1Beta1_GetProxies tests that GET requests to Gemini v1beta1 always use proxy
+// This is a characterization test for the route gating logic in routes.go
+func TestCharacterization_GeminiV1Beta1_GetProxies(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Setup recorders
+	proxyRecorder := testutil.NewFakeProxyRecorder()
+	handlerRecorder := testutil.NewFakeHandlerRecorder()
+
+	// Create a test server for the proxy
+	proxyServer := httptest.NewServer(proxyRecorder.ToHandler())
+	defer proxyServer.Close()
+
+	// Create fallback handler
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		targetURL, _ := url.Parse(proxyServer.URL)
+		return httputil.NewSingleHostReverseProxy(targetURL)
+	})
+
+	// Create the Gemini bridge handler
+	geminiBridge := createGeminiBridgeHandler(handlerRecorder.GinHandler())
+	geminiV1Beta1Handler := fh.WrapHandler(geminiBridge)
+
+	// Create router with the same gating logic as routes.go
+	r := gin.New()
+	r.Any("/api/provider/google/v1beta1/*path", func(c *gin.Context) {
+		if c.Request.Method == "POST" {
+			if path := c.Param("path"); strings.Contains(path, "/models/") {
+				geminiV1Beta1Handler(c)
+				return
+			}
+		}
+		proxyRecorder.ServeHTTP(c.Writer, c.Request)
+	})
+
+	// Execute: GET request (even with /models/ in path)
+	req := httptest.NewRequest(http.MethodGet, "/api/provider/google/v1beta1/publishers/google/models/gemini-pro", nil)
+	w := httptest.NewRecorder()
+	r.ServeHTTP(w, req)
+
+	// Assert: proxy called
+	assert.True(t, proxyRecorder.Called, "proxy should be called for GET requests")
+	assert.Equal(t, 1, proxyRecorder.GetCallCount(), "proxy should be called exactly once")
+
+	// Assert: local handler NOT called
+	assert.False(t, handlerRecorder.WasCalled(), "local handler should NOT be called for GET requests")
+}

internal/api/modules/amp/fallback_handlers_test.go

@@ -0,0 +1,148 @@
+package amp
+
+import (
+	"bytes"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"net/http/httputil"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/registry"
+	"github.com/stretchr/testify/assert"
+)
+
+// Characterization tests for fallback_handlers.go
+// These tests capture existing behavior before refactoring to routing layer
+
+func TestFallbackHandler_WrapHandler_LocalProvider_NoMapping(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Setup: model that has local providers (gemini-2.5-pro is registered)
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	body := `{"model": "gemini-2.5-pro", "messages": [{"role": "user", "content": "hello"}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/anthropic/v1/messages", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Handler that should be called (not proxy)
+	handlerCalled := false
+	handler := func(c *gin.Context) {
+		handlerCalled = true
+		c.JSON(200, gin.H{"status": "ok"})
+	}
+
+	// Create fallback handler
+	fh := NewFallbackHandler(func() *httputil.ReverseProxy {
+		return nil // no proxy
+	})
+
+	// Execute
+	wrapped := fh.WrapHandler(handler)
+	wrapped(c)
+
+	// Assert: handler should be called directly (no mapping needed)
+	assert.True(t, handlerCalled, "handler should be called for local provider")
+	assert.Equal(t, 200, w.Code)
+}
+
+func TestFallbackHandler_WrapHandler_MappingApplied(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the target model
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client", "anthropic", []*registry.ModelInfo{
+		{ID: "claude-opus-4-5-thinking"},
+	})
+
+	// Setup: model that needs mapping
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	body := `{"model": "claude-opus-4-5-20251101", "messages": [{"role": "user", "content": "hello"}]}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/anthropic/v1/messages", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	// Handler to capture rewritten body
+	var capturedBody []byte
+	handler := func(c *gin.Context) {
+		capturedBody, _ = io.ReadAll(c.Request.Body)
+		c.JSON(200, gin.H{"status": "ok"})
+	}
+
+	// Create fallback handler with mapper
+	mapper := NewModelMapper([]config.AmpModelMapping{
+		{From: "claude-opus-4-5-20251101", To: "claude-opus-4-5-thinking"},
+	})
+
+	fh := NewFallbackHandlerWithMapper(
+		func() *httputil.ReverseProxy { return nil },
+		mapper,
+		func() bool { return false },
+	)
+
+	// Execute
+	wrapped := fh.WrapHandler(handler)
+	wrapped(c)
+
+	// Assert: body should be rewritten
+	assert.Contains(t, string(capturedBody), "claude-opus-4-5-thinking")
+
+	// Assert: context should have mapped model
+	mappedModel, exists := c.Get(MappedModelContextKey)
+	assert.True(t, exists, "MappedModelContextKey should be set")
+	assert.NotEmpty(t, mappedModel)
+}
+
+func TestFallbackHandler_WrapHandler_ThinkingSuffixPreserved(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	// Register a mock provider for the target model
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client-2", "anthropic", []*registry.ModelInfo{
+		{ID: "claude-opus-4-5-thinking"},
+	})
+
+	w := httptest.NewRecorder()
+	c, _ := gin.CreateTestContext(w)
+
+	// Model with thinking suffix
+	body := `{"model": "claude-opus-4-5-20251101(xhigh)", "messages": []}`
+	req := httptest.NewRequest(http.MethodPost, "/api/provider/anthropic/v1/messages", bytes.NewReader([]byte(body)))
+	req.Header.Set("Content-Type", "application/json")
+	c.Request = req
+
+	var capturedBody []byte
+	handler := func(c *gin.Context) {
+		capturedBody, _ = io.ReadAll(c.Request.Body)
+		c.JSON(200, gin.H{"status": "ok"})
+	}
+
+	mapper := NewModelMapper([]config.AmpModelMapping{
+		{From: "claude-opus-4-5-20251101", To: "claude-opus-4-5-thinking"},
+	})
+
+	fh := NewFallbackHandlerWithMapper(
+		func() *httputil.ReverseProxy { return nil },
+		mapper,
+		func() bool { return false },
+	)
+
+	wrapped := fh.WrapHandler(handler)
+	wrapped(c)
+
+	// Assert: thinking suffix should be preserved
+	assert.Contains(t, string(capturedBody), "(xhigh)")
+}
+
+func TestFallbackHandler_WrapHandler_NoProvider_NoMapping_ProxyEnabled(t *testing.T) {
+	// Skip: httptest.ResponseRecorder doesn't implement http.CloseNotifier
+	// which is required by httputil.ReverseProxy. This test requires a real
+	// HTTP server and client to properly test proxy behavior.
+	t.Skip("requires real HTTP server for proxy testing")
+}

internal/api/modules/amp/model_mapping.go

@@ -3,10 +3,12 @@
 package amp
 
 import (
+	"regexp"
 	"strings"
 	"sync"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	log "github.com/sirupsen/logrus"
 )
@@ -26,47 +28,195 @@ type ModelMapper interface {
 // DefaultModelMapper implements ModelMapper with thread-safe mapping storage.
 type DefaultModelMapper struct {
 	mu       sync.RWMutex
-	mappings map[string]string // from -> to (normalized lowercase keys)
+	mappings map[string]string // exact: from -> to (normalized lowercase keys)
+	regexps  []regexMapping    // regex rules evaluated in order
+
+	// oauthAliasForward maps channel -> name (lower) -> []alias for oauth-model-alias lookup.
+	// This allows model-mappings targets to find providers via their aliases.
+	oauthAliasForward map[string]map[string][]string
 }
 
 // NewModelMapper creates a new model mapper with the given initial mappings.
 func NewModelMapper(mappings []config.AmpModelMapping) *DefaultModelMapper {
 	m := &DefaultModelMapper{
 		mappings:          make(map[string]string),
+		regexps:           nil,
+		oauthAliasForward: nil,
 	}
 	m.UpdateMappings(mappings)
 	return m
 }
 
+// UpdateOAuthModelAlias updates the oauth-model-alias lookup table.
+// This is called during initialization and on config hot-reload.
+func (m *DefaultModelMapper) UpdateOAuthModelAlias(aliases map[string][]config.OAuthModelAlias) {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	if len(aliases) == 0 {
+		m.oauthAliasForward = nil
+		return
+	}
+
+	forward := make(map[string]map[string][]string, len(aliases))
+	for rawChannel, entries := range aliases {
+		channel := strings.ToLower(strings.TrimSpace(rawChannel))
+		if channel == "" || len(entries) == 0 {
+			continue
+		}
+		channelMap := make(map[string][]string)
+		for _, entry := range entries {
+			name := strings.TrimSpace(entry.Name)
+			alias := strings.TrimSpace(entry.Alias)
+			if name == "" || alias == "" {
+				continue
+			}
+			if strings.EqualFold(name, alias) {
+				continue
+			}
+			nameKey := strings.ToLower(name)
+			channelMap[nameKey] = append(channelMap[nameKey], alias)
+		}
+		if len(channelMap) > 0 {
+			forward[channel] = channelMap
+		}
+	}
+	if len(forward) == 0 {
+		m.oauthAliasForward = nil
+		return
+	}
+	m.oauthAliasForward = forward
+	log.Debugf("amp model mapping: loaded oauth-model-alias for %d channel(s)", len(forward))
+}
+
+// findAllAliasesWithProviders returns all oauth-model-alias aliases for targetModel
+// that have available providers. Useful for fallback when one alias is quota-exceeded.
+func (m *DefaultModelMapper) findAllAliasesWithProviders(targetModel string) []string {
+	if m.oauthAliasForward == nil {
+		return nil
+	}
+
+	targetKey := strings.ToLower(strings.TrimSpace(targetModel))
+	if targetKey == "" {
+		return nil
+	}
+
+	var result []string
+	seen := make(map[string]struct{})
+
+	// Check all channels for this model name
+	for _, channelMap := range m.oauthAliasForward {
+		aliases := channelMap[targetKey]
+		for _, alias := range aliases {
+			aliasLower := strings.ToLower(alias)
+			if _, exists := seen[aliasLower]; exists {
+				continue
+			}
+			providers := util.GetProviderName(alias)
+			if len(providers) > 0 {
+				result = append(result, alias)
+				seen[aliasLower] = struct{}{}
+			}
+		}
+	}
+	return result
+}
+
 // MapModel checks if a mapping exists for the requested model and if the
 // target model has available local providers. Returns the mapped model name
 // or empty string if no valid mapping exists.
+//
+// If the requested model contains a thinking suffix (e.g., "g25p(8192)"),
+// the suffix is preserved in the returned model name (e.g., "gemini-2.5-pro(8192)").
+// However, if the mapping target already contains a suffix, the config suffix
+// takes priority over the user's suffix.
 func (m *DefaultModelMapper) MapModel(requestedModel string) string {
-	if requestedModel == "" {
+	models := m.MapModelWithFallbacks(requestedModel)
+	if len(models) == 0 {
 		return ""
 	}
+	return models[0]
+}
+
+// MapModelWithFallbacks returns all possible target models for the requested model,
+// including fallback aliases from oauth-model-alias. The first model is the primary target,
+// and subsequent models are fallbacks to try if the primary is unavailable (e.g., quota exceeded).
+func (m *DefaultModelMapper) MapModelWithFallbacks(requestedModel string) []string {
+	if requestedModel == "" {
+		return nil
+	}
 
 	m.mu.RLock()
 	defer m.mu.RUnlock()
 
-	// Normalize the requested model for lookup
-	normalizedRequest := strings.ToLower(strings.TrimSpace(requestedModel))
+	// Extract thinking suffix from requested model using ParseSuffix
+	requestResult := thinking.ParseSuffix(requestedModel)
+	baseModel := requestResult.ModelName
 
-	// Check for direct mapping
-	targetModel, exists := m.mappings[normalizedRequest]
+	// Normalize the base model for lookup (case-insensitive)
+	normalizedBase := strings.ToLower(strings.TrimSpace(baseModel))
+
+	// Check for direct mapping using base model name
+	targetModel, exists := m.mappings[normalizedBase]
 	if !exists {
-		return ""
+		// Try regex mappings in order using base model only
+		// (suffix is handled separately via ParseSuffix)
+		for _, rm := range m.regexps {
+			if rm.re.MatchString(baseModel) {
+				targetModel = rm.to
+				exists = true
+				break
+			}
+		}
+		if !exists {
+			return nil
+		}
 	}
 
-	// Verify target model has available providers
-	providers := util.GetProviderName(targetModel)
-	if len(providers) == 0 {
+	// Check if target model already has a thinking suffix (config priority)
+	targetResult := thinking.ParseSuffix(targetModel)
+	targetBase := targetResult.ModelName
+
+	// Helper to apply suffix to a model
+	applySuffix := func(model string) string {
+		modelResult := thinking.ParseSuffix(model)
+		if modelResult.HasSuffix {
+			return model
+		}
+		if requestResult.HasSuffix && requestResult.RawSuffix != "" {
+			return model + "(" + requestResult.RawSuffix + ")"
+		}
+		return model
+	}
+
+	// Verify target model has available providers (use base model for lookup)
+	providers := util.GetProviderName(targetBase)
+
+	// If direct provider available, return it as primary
+	if len(providers) > 0 {
+		return []string{applySuffix(targetModel)}
+	}
+
+	// No direct providers - check oauth-model-alias for all aliases that have providers
+	allAliases := m.findAllAliasesWithProviders(targetBase)
+	if len(allAliases) == 0 {
 		log.Debugf("amp model mapping: target model %s has no available providers, skipping mapping", targetModel)
-		return ""
+		return nil
 	}
 
-	// Note: Detailed routing log is handled by logAmpRouting in fallback_handlers.go
-	return targetModel
+	// Log resolution
+	if len(allAliases) == 1 {
+		log.Debugf("amp model mapping: resolved %s -> %s via oauth-model-alias", targetModel, allAliases[0])
+	} else {
+		log.Debugf("amp model mapping: resolved %s -> %v via oauth-model-alias (%d fallbacks)", targetModel, allAliases, len(allAliases)-1)
+	}
+
+	// Apply suffix to all aliases
+	result := make([]string, len(allAliases))
+	for i, alias := range allAliases {
+		result[i] = applySuffix(alias)
+	}
+	return result
 }
 
 // UpdateMappings refreshes the mapping configuration from config.
@@ -77,6 +227,7 @@ func (m *DefaultModelMapper) UpdateMappings(mappings []config.AmpModelMapping) {
 
 	// Clear and rebuild mappings
 	m.mappings = make(map[string]string, len(mappings))
+	m.regexps = make([]regexMapping, 0, len(mappings))
 
 	for _, mapping := range mappings {
 		from := strings.TrimSpace(mapping.From)
@@ -87,16 +238,30 @@ func (m *DefaultModelMapper) UpdateMappings(mappings []config.AmpModelMapping) {
 			continue
 		}
 
+		if mapping.Regex {
+			// Compile case-insensitive regex; wrap with (?i) to match behavior of exact lookups
+			pattern := "(?i)" + from
+			re, err := regexp.Compile(pattern)
+			if err != nil {
+				log.Warnf("amp model mapping: invalid regex %q: %v", from, err)
+				continue
+			}
+			m.regexps = append(m.regexps, regexMapping{re: re, to: to})
+			log.Debugf("amp model regex mapping registered: /%s/ -> %s", from, to)
+		} else {
 			// Store with normalized lowercase key for case-insensitive lookup
 			normalizedFrom := strings.ToLower(from)
 			m.mappings[normalizedFrom] = to
-
 			log.Debugf("amp model mapping registered: %s -> %s", from, to)
 		}
+	}
 
 	if len(m.mappings) > 0 {
 		log.Infof("amp model mapping: loaded %d mapping(s)", len(m.mappings))
 	}
+	if n := len(m.regexps); n > 0 {
+		log.Infof("amp model mapping: loaded %d regex mapping(s)", n)
+	}
 }
 
 // GetMappings returns a copy of current mappings (for debugging/status).
@@ -110,3 +275,24 @@ func (m *DefaultModelMapper) GetMappings() map[string]string {
 	}
 	return result
 }
+
+// GetMappingsAsConfig returns the current model mappings as config.AmpModelMapping slice.
+// Safe for concurrent use.
+func (m *DefaultModelMapper) GetMappingsAsConfig() []config.AmpModelMapping {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+
+	result := make([]config.AmpModelMapping, 0, len(m.mappings))
+	for from, to := range m.mappings {
+		result = append(result, config.AmpModelMapping{
+			From: from,
+			To:   to,
+		})
+	}
+	return result
+}
+
+type regexMapping struct {
+	re *regexp.Regexp
+	to string
+}

internal/api/modules/amp/model_mapping_test.go

@@ -71,6 +71,25 @@ func TestModelMapper_MapModel_WithProvider(t *testing.T) {
 	}
 }
 
+func TestModelMapper_MapModel_TargetWithThinkingSuffix(t *testing.T) {
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client-thinking", "codex", []*registry.ModelInfo{
+		{ID: "gpt-5.2", OwnedBy: "openai", Type: "codex"},
+	})
+	defer reg.UnregisterClient("test-client-thinking")
+
+	mappings := []config.AmpModelMapping{
+		{From: "gpt-5.2-alias", To: "gpt-5.2(xhigh)"},
+	}
+
+	mapper := NewModelMapper(mappings)
+
+	result := mapper.MapModel("gpt-5.2-alias")
+	if result != "gpt-5.2(xhigh)" {
+		t.Errorf("Expected gpt-5.2(xhigh), got %s", result)
+	}
+}
+
 func TestModelMapper_MapModel_CaseInsensitive(t *testing.T) {
 	reg := registry.GetGlobalRegistry()
 	reg.RegisterClient("test-client2", "claude", []*registry.ModelInfo{
@@ -184,3 +203,173 @@ func TestModelMapper_GetMappings_ReturnsCopy(t *testing.T) {
 		t.Error("Original map was modified")
 	}
 }
+
+func TestModelMapper_Regex_MatchBaseWithoutParens(t *testing.T) {
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client-regex-1", "gemini", []*registry.ModelInfo{
+		{ID: "gemini-2.5-pro", OwnedBy: "google", Type: "gemini"},
+	})
+	defer reg.UnregisterClient("test-client-regex-1")
+
+	mappings := []config.AmpModelMapping{
+		{From: "^gpt-5$", To: "gemini-2.5-pro", Regex: true},
+	}
+
+	mapper := NewModelMapper(mappings)
+
+	// Incoming model has reasoning suffix, regex matches base, suffix is preserved
+	result := mapper.MapModel("gpt-5(high)")
+	if result != "gemini-2.5-pro(high)" {
+		t.Errorf("Expected gemini-2.5-pro(high), got %s", result)
+	}
+}
+
+func TestModelMapper_Regex_ExactPrecedence(t *testing.T) {
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client-regex-2", "claude", []*registry.ModelInfo{
+		{ID: "claude-sonnet-4", OwnedBy: "anthropic", Type: "claude"},
+	})
+	reg.RegisterClient("test-client-regex-3", "gemini", []*registry.ModelInfo{
+		{ID: "gemini-2.5-pro", OwnedBy: "google", Type: "gemini"},
+	})
+	defer reg.UnregisterClient("test-client-regex-2")
+	defer reg.UnregisterClient("test-client-regex-3")
+
+	mappings := []config.AmpModelMapping{
+		{From: "gpt-5", To: "claude-sonnet-4"},                 // exact
+		{From: "^gpt-5.*$", To: "gemini-2.5-pro", Regex: true}, // regex
+	}
+
+	mapper := NewModelMapper(mappings)
+
+	// Exact match should win over regex
+	result := mapper.MapModel("gpt-5")
+	if result != "claude-sonnet-4" {
+		t.Errorf("Expected claude-sonnet-4, got %s", result)
+	}
+}
+
+func TestModelMapper_Regex_InvalidPattern_Skipped(t *testing.T) {
+	// Invalid regex should be skipped and not cause panic
+	mappings := []config.AmpModelMapping{
+		{From: "(", To: "target", Regex: true},
+	}
+
+	mapper := NewModelMapper(mappings)
+
+	result := mapper.MapModel("anything")
+	if result != "" {
+		t.Errorf("Expected empty result due to invalid regex, got %s", result)
+	}
+}
+
+func TestModelMapper_Regex_CaseInsensitive(t *testing.T) {
+	reg := registry.GetGlobalRegistry()
+	reg.RegisterClient("test-client-regex-4", "claude", []*registry.ModelInfo{
+		{ID: "claude-sonnet-4", OwnedBy: "anthropic", Type: "claude"},
+	})
+	defer reg.UnregisterClient("test-client-regex-4")
+
+	mappings := []config.AmpModelMapping{
+		{From: "^CLAUDE-OPUS-.*$", To: "claude-sonnet-4", Regex: true},
+	}
+
+	mapper := NewModelMapper(mappings)
+
+	result := mapper.MapModel("claude-opus-4.5")
+	if result != "claude-sonnet-4" {
+		t.Errorf("Expected claude-sonnet-4, got %s", result)
+	}
+}
+
+func TestModelMapper_SuffixPreservation(t *testing.T) {
+	reg := registry.GetGlobalRegistry()
+
+	// Register test models
+	reg.RegisterClient("test-client-suffix", "gemini", []*registry.ModelInfo{
+		{ID: "gemini-2.5-pro", OwnedBy: "google", Type: "gemini"},
+	})
+	reg.RegisterClient("test-client-suffix-2", "claude", []*registry.ModelInfo{
+		{ID: "claude-sonnet-4", OwnedBy: "anthropic", Type: "claude"},
+	})
+	defer reg.UnregisterClient("test-client-suffix")
+	defer reg.UnregisterClient("test-client-suffix-2")
+
+	tests := []struct {
+		name     string
+		mappings []config.AmpModelMapping
+		input    string
+		want     string
+	}{
+		{
+			name:     "numeric suffix preserved",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p(8192)",
+			want:     "gemini-2.5-pro(8192)",
+		},
+		{
+			name:     "level suffix preserved",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p(high)",
+			want:     "gemini-2.5-pro(high)",
+		},
+		{
+			name:     "no suffix unchanged",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p",
+			want:     "gemini-2.5-pro",
+		},
+		{
+			name:     "config suffix takes priority",
+			mappings: []config.AmpModelMapping{{From: "alias", To: "gemini-2.5-pro(medium)"}},
+			input:    "alias(high)",
+			want:     "gemini-2.5-pro(medium)",
+		},
+		{
+			name:     "regex with suffix preserved",
+			mappings: []config.AmpModelMapping{{From: "^g25.*", To: "gemini-2.5-pro", Regex: true}},
+			input:    "g25p(8192)",
+			want:     "gemini-2.5-pro(8192)",
+		},
+		{
+			name:     "auto suffix preserved",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p(auto)",
+			want:     "gemini-2.5-pro(auto)",
+		},
+		{
+			name:     "none suffix preserved",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p(none)",
+			want:     "gemini-2.5-pro(none)",
+		},
+		{
+			name:     "case insensitive base lookup with suffix",
+			mappings: []config.AmpModelMapping{{From: "G25P", To: "gemini-2.5-pro"}},
+			input:    "g25p(high)",
+			want:     "gemini-2.5-pro(high)",
+		},
+		{
+			name:     "empty suffix filtered out",
+			mappings: []config.AmpModelMapping{{From: "g25p", To: "gemini-2.5-pro"}},
+			input:    "g25p()",
+			want:     "gemini-2.5-pro",
+		},
+		{
+			name:     "incomplete suffix treated as no suffix",
+			mappings: []config.AmpModelMapping{{From: "g25p(high", To: "gemini-2.5-pro"}},
+			input:    "g25p(high",
+			want:     "gemini-2.5-pro",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			mapper := NewModelMapper(tt.mappings)
+			got := mapper.MapModel(tt.input)
+			if got != tt.want {
+				t.Errorf("MapModel(%q) = %q, want %q", tt.input, got, tt.want)
+			}
+		})
+	}
+}

internal/api/modules/amp/proxy.go

@@ -15,6 +15,33 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+func removeQueryValuesMatching(req *http.Request, key string, match string) {
+	if req == nil || req.URL == nil || match == "" {
+		return
+	}
+
+	q := req.URL.Query()
+	values, ok := q[key]
+	if !ok || len(values) == 0 {
+		return
+	}
+
+	kept := make([]string, 0, len(values))
+	for _, v := range values {
+		if v == match {
+			continue
+		}
+		kept = append(kept, v)
+	}
+
+	if len(kept) == 0 {
+		q.Del(key)
+	} else {
+		q[key] = kept
+	}
+	req.URL.RawQuery = q.Encode()
+}
+
 // readCloser wraps a reader and forwards Close to a separate closer.
 // Used to restore peeked bytes while preserving upstream body Close behavior.
 type readCloser struct {
@@ -41,6 +68,19 @@ func createReverseProxy(upstreamURL string, secretSource SecretSource) (*httputi
 		originalDirector(req)
 		req.Host = parsed.Host
 
+		// Remove client's Authorization header - it was only used for CLI Proxy API authentication
+		// We will set our own Authorization using the configured upstream-api-key
+		req.Header.Del("Authorization")
+		req.Header.Del("X-Api-Key")
+		req.Header.Del("X-Goog-Api-Key")
+
+		// Remove query-based credentials if they match the authenticated client API key.
+		// This prevents leaking client auth material to the Amp upstream while avoiding
+		// breaking unrelated upstream query parameters.
+		clientKey := getClientAPIKeyFromContext(req.Context())
+		removeQueryValuesMatching(req, "key", clientKey)
+		removeQueryValuesMatching(req, "auth_token", clientKey)
+
 		// Preserve correlation headers for debugging
 		if req.Header.Get("X-Request-ID") == "" {
 			// Could generate one here if needed
@@ -50,7 +90,7 @@ func createReverseProxy(upstreamURL string, secretSource SecretSource) (*httputi
 		// Users going through ampcode.com proxy are paying for the service and should get all features
 		// including 1M context window (context-1m-2025-08-07)
 
-		// Inject API key from secret source (precedence: config > env > file)
+		// Inject API key from secret source (only uses upstream-api-key from config)
 		if key, err := secretSource.Get(req.Context()); err == nil && key != "" {
 			req.Header.Set("X-Api-Key", key)
 			req.Header.Set("Authorization", fmt.Sprintf("Bearer %s", key))

internal/api/modules/amp/proxy_test.go

@@ -3,11 +3,15 @@ package amp
 import (
 	"bytes"
 	"compress/gzip"
+	"context"
 	"fmt"
 	"io"
 	"net/http"
 	"net/http/httptest"
+	"strings"
 	"testing"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 )
 
 // Helper: compress data with gzip
@@ -306,6 +310,159 @@ func TestReverseProxy_EmptySecret(t *testing.T) {
 	}
 }
 
+func TestReverseProxy_StripsClientCredentialsFromHeadersAndQuery(t *testing.T) {
+	type captured struct {
+		headers http.Header
+		query   string
+	}
+	got := make(chan captured, 1)
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		got <- captured{headers: r.Header.Clone(), query: r.URL.RawQuery}
+		w.WriteHeader(200)
+		w.Write([]byte(`ok`))
+	}))
+	defer upstream.Close()
+
+	proxy, err := createReverseProxy(upstream.URL, NewStaticSecretSource("upstream"))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		// Simulate clientAPIKeyMiddleware injection (per-request)
+		ctx := context.WithValue(r.Context(), clientAPIKeyContextKey{}, "client-key")
+		proxy.ServeHTTP(w, r.WithContext(ctx))
+	}))
+	defer srv.Close()
+
+	req, err := http.NewRequest(http.MethodGet, srv.URL+"/test?key=client-key&key=keep&auth_token=client-key&foo=bar", nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	req.Header.Set("Authorization", "Bearer client-key")
+	req.Header.Set("X-Api-Key", "client-key")
+	req.Header.Set("X-Goog-Api-Key", "client-key")
+
+	res, err := http.DefaultClient.Do(req)
+	if err != nil {
+		t.Fatal(err)
+	}
+	res.Body.Close()
+
+	c := <-got
+
+	// These are client-provided credentials and must not reach the upstream.
+	if v := c.headers.Get("X-Goog-Api-Key"); v != "" {
+		t.Fatalf("X-Goog-Api-Key should be stripped, got: %q", v)
+	}
+
+	// We inject upstream Authorization/X-Api-Key, so the client auth must not survive.
+	if v := c.headers.Get("Authorization"); v != "Bearer upstream" {
+		t.Fatalf("Authorization should be upstream-injected, got: %q", v)
+	}
+	if v := c.headers.Get("X-Api-Key"); v != "upstream" {
+		t.Fatalf("X-Api-Key should be upstream-injected, got: %q", v)
+	}
+
+	// Query-based credentials should be stripped only when they match the authenticated client key.
+	// Should keep unrelated values and parameters.
+	if strings.Contains(c.query, "auth_token=client-key") || strings.Contains(c.query, "key=client-key") {
+		t.Fatalf("query credentials should be stripped, got raw query: %q", c.query)
+	}
+	if !strings.Contains(c.query, "key=keep") || !strings.Contains(c.query, "foo=bar") {
+		t.Fatalf("expected query to keep non-credential params, got raw query: %q", c.query)
+	}
+}
+
+func TestReverseProxy_InjectsMappedSecret_FromRequestContext(t *testing.T) {
+	gotHeaders := make(chan http.Header, 1)
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		gotHeaders <- r.Header.Clone()
+		w.WriteHeader(200)
+		w.Write([]byte(`ok`))
+	}))
+	defer upstream.Close()
+
+	defaultSource := NewStaticSecretSource("default")
+	mapped := NewMappedSecretSource(defaultSource)
+	mapped.UpdateMappings([]config.AmpUpstreamAPIKeyEntry{
+		{
+			UpstreamAPIKey: "u1",
+			APIKeys:        []string{"k1"},
+		},
+	})
+
+	proxy, err := createReverseProxy(upstream.URL, mapped)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		// Simulate clientAPIKeyMiddleware injection (per-request)
+		ctx := context.WithValue(r.Context(), clientAPIKeyContextKey{}, "k1")
+		proxy.ServeHTTP(w, r.WithContext(ctx))
+	}))
+	defer srv.Close()
+
+	res, err := http.Get(srv.URL + "/test")
+	if err != nil {
+		t.Fatal(err)
+	}
+	res.Body.Close()
+
+	hdr := <-gotHeaders
+	if hdr.Get("X-Api-Key") != "u1" {
+		t.Fatalf("X-Api-Key missing or wrong, got: %q", hdr.Get("X-Api-Key"))
+	}
+	if hdr.Get("Authorization") != "Bearer u1" {
+		t.Fatalf("Authorization missing or wrong, got: %q", hdr.Get("Authorization"))
+	}
+}
+
+func TestReverseProxy_MappedSecret_FallsBackToDefault(t *testing.T) {
+	gotHeaders := make(chan http.Header, 1)
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		gotHeaders <- r.Header.Clone()
+		w.WriteHeader(200)
+		w.Write([]byte(`ok`))
+	}))
+	defer upstream.Close()
+
+	defaultSource := NewStaticSecretSource("default")
+	mapped := NewMappedSecretSource(defaultSource)
+	mapped.UpdateMappings([]config.AmpUpstreamAPIKeyEntry{
+		{
+			UpstreamAPIKey: "u1",
+			APIKeys:        []string{"k1"},
+		},
+	})
+
+	proxy, err := createReverseProxy(upstream.URL, mapped)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		ctx := context.WithValue(r.Context(), clientAPIKeyContextKey{}, "k2")
+		proxy.ServeHTTP(w, r.WithContext(ctx))
+	}))
+	defer srv.Close()
+
+	res, err := http.Get(srv.URL + "/test")
+	if err != nil {
+		t.Fatal(err)
+	}
+	res.Body.Close()
+
+	hdr := <-gotHeaders
+	if hdr.Get("X-Api-Key") != "default" {
+		t.Fatalf("X-Api-Key fallback missing or wrong, got: %q", hdr.Get("X-Api-Key"))
+	}
+	if hdr.Get("Authorization") != "Bearer default" {
+		t.Fatalf("Authorization fallback missing or wrong, got: %q", hdr.Get("Authorization"))
+	}
+}
+
 func TestReverseProxy_ErrorHandler(t *testing.T) {
 	// Point proxy to a non-routable address to trigger error
 	proxy, err := createReverseProxy("http://127.0.0.1:1", NewStaticSecretSource(""))

internal/api/modules/amp/response_rewriter.go

@@ -39,7 +39,13 @@ func (rw *ResponseRewriter) Write(data []byte) (int, error) {
 	}
 
 	if rw.isStreaming {
-		return rw.ResponseWriter.Write(rw.rewriteStreamChunk(data))
+		n, err := rw.ResponseWriter.Write(rw.rewriteStreamChunk(data))
+		if err == nil {
+			if flusher, ok := rw.ResponseWriter.(http.Flusher); ok {
+				flusher.Flush()
+			}
+		}
+		return n, err
 	}
 	return rw.body.Write(data)
 }
@@ -63,7 +69,30 @@ func (rw *ResponseRewriter) Flush() {
 var modelFieldPaths = []string{"model", "modelVersion", "response.modelVersion", "message.model"}
 
 // rewriteModelInResponse replaces all occurrences of the mapped model with the original model in JSON
+// It also suppresses "thinking" blocks if "tool_use" is present to ensure Amp client compatibility
 func (rw *ResponseRewriter) rewriteModelInResponse(data []byte) []byte {
+	// 1. Amp Compatibility: Suppress thinking blocks if tool use is detected
+	// The Amp client struggles when both thinking and tool_use blocks are present
+	if gjson.GetBytes(data, `content.#(type=="tool_use")`).Exists() {
+		filtered := gjson.GetBytes(data, `content.#(type!="thinking")#`)
+		if filtered.Exists() {
+			originalCount := gjson.GetBytes(data, "content.#").Int()
+			filteredCount := filtered.Get("#").Int()
+
+			if originalCount > filteredCount {
+				var err error
+				data, err = sjson.SetBytes(data, "content", filtered.Value())
+				if err != nil {
+					log.Warnf("Amp ResponseRewriter: failed to suppress thinking blocks: %v", err)
+				} else {
+					log.Debugf("Amp ResponseRewriter: Suppressed %d thinking blocks due to tool usage", originalCount-filteredCount)
+					// Log the result for verification
+					log.Debugf("Amp ResponseRewriter: Resulting content: %s", gjson.GetBytes(data, "content").String())
+				}
+			}
+		}
+	}
+
 	if rw.originalModel == "" {
 		return data
 	}

internal/api/modules/amp/routes.go

@@ -1,14 +1,16 @@
 package amp
 
 import (
+	"context"
 	"errors"
 	"net"
 	"net/http"
-	"net/http/httputil"
 	"strings"
 
 	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/routing"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers/claude"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers/gemini"
@@ -16,6 +18,37 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+// clientAPIKeyContextKey is the context key used to pass the client API key
+// from gin.Context to the request context for SecretSource lookup.
+type clientAPIKeyContextKey struct{}
+
+// clientAPIKeyMiddleware injects the authenticated client API key from gin.Context["apiKey"]
+// into the request context so that SecretSource can look it up for per-client upstream routing.
+func clientAPIKeyMiddleware() gin.HandlerFunc {
+	return func(c *gin.Context) {
+		// Extract the client API key from gin context (set by AuthMiddleware)
+		if apiKey, exists := c.Get("apiKey"); exists {
+			if keyStr, ok := apiKey.(string); ok && keyStr != "" {
+				// Inject into request context for SecretSource.Get(ctx) to read
+				ctx := context.WithValue(c.Request.Context(), clientAPIKeyContextKey{}, keyStr)
+				c.Request = c.Request.WithContext(ctx)
+			}
+		}
+		c.Next()
+	}
+}
+
+// getClientAPIKeyFromContext retrieves the client API key from request context.
+// Returns empty string if not present.
+func getClientAPIKeyFromContext(ctx context.Context) string {
+	if val := ctx.Value(clientAPIKeyContextKey{}); val != nil {
+		if keyStr, ok := val.(string); ok {
+			return keyStr
+		}
+	}
+	return ""
+}
+
 // localhostOnlyMiddleware returns a middleware that dynamically checks the module's
 // localhost restriction setting. This allows hot-reload of the restriction without restarting.
 func (m *AmpModule) localhostOnlyMiddleware() gin.HandlerFunc {
@@ -95,10 +128,25 @@ func (m *AmpModule) managementAvailabilityMiddleware() gin.HandlerFunc {
 	}
 }
 
+// wrapManagementAuth skips auth for selected management paths while keeping authentication elsewhere.
+func wrapManagementAuth(auth gin.HandlerFunc, prefixes ...string) gin.HandlerFunc {
+	return func(c *gin.Context) {
+		path := c.Request.URL.Path
+		for _, prefix := range prefixes {
+			if strings.HasPrefix(path, prefix) && (len(path) == len(prefix) || path[len(prefix)] == '/') {
+				c.Next()
+				return
+			}
+		}
+		auth(c)
+	}
+}
+
 // registerManagementRoutes registers Amp management proxy routes
 // These routes proxy through to the Amp control plane for OAuth, user management, etc.
 // Uses dynamic middleware and proxy getter for hot-reload support.
-func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *handlers.BaseAPIHandler) {
+// The auth middleware validates Authorization header against configured API keys.
+func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *handlers.BaseAPIHandler, auth gin.HandlerFunc) {
 	ampAPI := engine.Group("/api")
 
 	// Always disable CORS for management routes to prevent browser-based attacks
@@ -107,10 +155,16 @@ func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *ha
 	// Apply dynamic localhost-only restriction (hot-reloadable via m.IsRestrictedToLocalhost())
 	ampAPI.Use(m.localhostOnlyMiddleware())
 
-	if !m.IsRestrictedToLocalhost() {
-		log.Warn("amp management routes are NOT restricted to localhost - this is insecure!")
+	// Apply authentication middleware - requires valid API key in Authorization header
+	var authWithBypass gin.HandlerFunc
+	if auth != nil {
+		ampAPI.Use(auth)
+		authWithBypass = wrapManagementAuth(auth, "/threads", "/auth", "/docs", "/settings")
 	}
 
+	// Inject client API key into request context for per-client upstream routing
+	ampAPI.Use(clientAPIKeyMiddleware())
+
 	// Dynamic proxy handler that uses m.getProxy() for hot-reload support
 	proxyHandler := func(c *gin.Context) {
 		// Swallow ErrAbortHandler panics from ReverseProxy copyResponse to avoid noisy stack traces
@@ -154,7 +208,18 @@ func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *ha
 	// Root-level routes that AMP CLI expects without /api prefix
 	// These need the same security middleware as the /api/* routes (dynamic for hot-reload)
 	rootMiddleware := []gin.HandlerFunc{m.managementAvailabilityMiddleware(), noCORSMiddleware(), m.localhostOnlyMiddleware()}
+	if authWithBypass != nil {
+		rootMiddleware = append(rootMiddleware, authWithBypass)
+	}
+	// Add clientAPIKeyMiddleware after auth for per-client upstream routing
+	rootMiddleware = append(rootMiddleware, clientAPIKeyMiddleware())
+	engine.GET("/threads", append(rootMiddleware, proxyHandler)...)
 	engine.GET("/threads/*path", append(rootMiddleware, proxyHandler)...)
+	engine.GET("/docs", append(rootMiddleware, proxyHandler)...)
+	engine.GET("/docs/*path", append(rootMiddleware, proxyHandler)...)
+	engine.GET("/settings", append(rootMiddleware, proxyHandler)...)
+	engine.GET("/settings/*path", append(rootMiddleware, proxyHandler)...)
+
 	engine.GET("/threads.rss", append(rootMiddleware, proxyHandler)...)
 	engine.GET("/news.rss", append(rootMiddleware, proxyHandler)...)
 
@@ -170,19 +235,20 @@ func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *ha
 	// If no local OAuth is available, falls back to ampcode.com proxy.
 	geminiHandlers := gemini.NewGeminiAPIHandler(baseHandler)
 	geminiBridge := createGeminiBridgeHandler(geminiHandlers.GeminiHandler)
-	geminiV1Beta1Fallback := NewFallbackHandlerWithMapper(func() *httputil.ReverseProxy {
-		return m.getProxy()
-	}, m.modelMapper, m.forceModelMappings)
-	geminiV1Beta1Handler := geminiV1Beta1Fallback.WrapHandler(geminiBridge)
 
-	// Route POST model calls through Gemini bridge with FallbackHandler.
-	// FallbackHandler checks provider -> mapping -> proxy fallback automatically.
+	// T-025: Migrated Gemini v1beta1 bridge to use ModelRoutingWrapper
+	// Create a dedicated routing wrapper for the Gemini bridge
+	geminiBridgeWrapper := m.createModelRoutingWrapper()
+	geminiV1Beta1Handler := geminiBridgeWrapper.Wrap(geminiBridge)
+
+	// Route POST model calls through Gemini bridge with ModelRoutingWrapper.
+	// ModelRoutingWrapper checks provider -> mapping -> proxy fallback automatically.
 	// All other methods (e.g., GET model listing) always proxy to upstream to preserve Amp CLI behavior.
 	ampAPI.Any("/provider/google/v1beta1/*path", func(c *gin.Context) {
 		if c.Request.Method == "POST" {
 			if path := c.Param("path"); strings.Contains(path, "/models/") {
-				// POST with /models/ path -> use Gemini bridge with fallback handler
-				// FallbackHandler will check provider/mapping and proxy if needed
+				// POST with /models/ path -> use Gemini bridge with unified routing wrapper
+				// ModelRoutingWrapper will check provider/mapping and proxy if needed
 				geminiV1Beta1Handler(c)
 				return
 			}
@@ -192,6 +258,41 @@ func (m *AmpModule) registerManagementRoutes(engine *gin.Engine, baseHandler *ha
 	})
 }
 
+// createModelRoutingWrapper creates a new ModelRoutingWrapper for unified routing.
+// This is used for testing the new routing implementation (T-021 onwards).
+func (m *AmpModule) createModelRoutingWrapper() *routing.ModelRoutingWrapper {
+	// Create a registry - in production this would be populated with actual providers
+	registry := routing.NewRegistry()
+
+	// Create a minimal config with just AmpCode settings
+	// The Router only needs AmpCode.ModelMappings and OAuthModelAlias
+	cfg := &config.Config{
+		AmpCode: func() config.AmpCode {
+			if m.modelMapper != nil {
+				return config.AmpCode{
+					ModelMappings: m.modelMapper.GetMappingsAsConfig(),
+				}
+			}
+			return config.AmpCode{}
+		}(),
+	}
+
+	// Create router with registry and config
+	router := routing.NewRouter(registry, cfg)
+
+	// Create wrapper with proxy function
+	proxyFunc := func(c *gin.Context) {
+		proxy := m.getProxy()
+		if proxy != nil {
+			proxy.ServeHTTP(c.Writer, c.Request)
+		} else {
+			c.JSON(503, gin.H{"error": "amp upstream proxy not available"})
+		}
+	}
+
+	return routing.NewModelRoutingWrapper(router, nil, nil, proxyFunc)
+}
+
 // registerProviderAliases registers /api/provider/{provider}/... routes
 // These allow Amp CLI to route requests like:
 //
@@ -205,18 +306,17 @@ func (m *AmpModule) registerProviderAliases(engine *gin.Engine, baseHandler *han
 	claudeCodeHandlers := claude.NewClaudeCodeAPIHandler(baseHandler)
 	openaiResponsesHandlers := openai.NewOpenAIResponsesAPIHandler(baseHandler)
 
-	// Create fallback handler wrapper that forwards to ampcode.com when provider not found
-	// Uses m.getProxy() for hot-reload support (proxy can be updated at runtime)
-	// Also includes model mapping support for routing unavailable models to alternatives
-	fallbackHandler := NewFallbackHandlerWithMapper(func() *httputil.ReverseProxy {
-		return m.getProxy()
-	}, m.modelMapper, m.forceModelMappings)
+	// Create unified routing wrapper (T-021 onwards)
+	// Replaces FallbackHandler with Router-based unified routing
+	routingWrapper := m.createModelRoutingWrapper()
 
 	// Provider-specific routes under /api/provider/:provider
 	ampProviders := engine.Group("/api/provider")
 	if auth != nil {
 		ampProviders.Use(auth)
 	}
+	// Inject client API key into request context for per-client upstream routing
+	ampProviders.Use(clientAPIKeyMiddleware())
 
 	provider := ampProviders.Group("/:provider")
 
@@ -236,33 +336,36 @@ func (m *AmpModule) registerProviderAliases(engine *gin.Engine, baseHandler *han
 	}
 
 	// Root-level routes (for providers that omit /v1, like groq/cerebras)
-	// Wrap handlers with fallback logic to forward to ampcode.com when provider not found
+	// T-022: Migrated all OpenAI routes to use ModelRoutingWrapper for unified routing
 	provider.GET("/models", ampModelsHandler) // Models endpoint doesn't need fallback (no body to check)
-	provider.POST("/chat/completions", fallbackHandler.WrapHandler(openaiHandlers.ChatCompletions))
-	provider.POST("/completions", fallbackHandler.WrapHandler(openaiHandlers.Completions))
-	provider.POST("/responses", fallbackHandler.WrapHandler(openaiResponsesHandlers.Responses))
+	provider.POST("/chat/completions", routingWrapper.Wrap(openaiHandlers.ChatCompletions))
+	provider.POST("/completions", routingWrapper.Wrap(openaiHandlers.Completions))
+	provider.POST("/responses", routingWrapper.Wrap(openaiResponsesHandlers.Responses))
 
 	// /v1 routes (OpenAI/Claude-compatible endpoints)
 	v1Amp := provider.Group("/v1")
 	{
 		v1Amp.GET("/models", ampModelsHandler) // Models endpoint doesn't need fallback
 
-		// OpenAI-compatible endpoints with fallback
-		v1Amp.POST("/chat/completions", fallbackHandler.WrapHandler(openaiHandlers.ChatCompletions))
-		v1Amp.POST("/completions", fallbackHandler.WrapHandler(openaiHandlers.Completions))
-		v1Amp.POST("/responses", fallbackHandler.WrapHandler(openaiResponsesHandlers.Responses))
+		// OpenAI-compatible endpoints with ModelRoutingWrapper
+		// T-021, T-022: Migrated to unified routing wrapper
+		v1Amp.POST("/chat/completions", routingWrapper.Wrap(openaiHandlers.ChatCompletions))
+		v1Amp.POST("/completions", routingWrapper.Wrap(openaiHandlers.Completions))
+		v1Amp.POST("/responses", routingWrapper.Wrap(openaiResponsesHandlers.Responses))
 
-		// Claude/Anthropic-compatible endpoints with fallback
-		v1Amp.POST("/messages", fallbackHandler.WrapHandler(claudeCodeHandlers.ClaudeMessages))
-		v1Amp.POST("/messages/count_tokens", fallbackHandler.WrapHandler(claudeCodeHandlers.ClaudeCountTokens))
+		// Claude/Anthropic-compatible endpoints with ModelRoutingWrapper
+		// T-023: Migrated Claude routes to unified routing wrapper
+		v1Amp.POST("/messages", routingWrapper.Wrap(claudeCodeHandlers.ClaudeMessages))
+		v1Amp.POST("/messages/count_tokens", routingWrapper.Wrap(claudeCodeHandlers.ClaudeCountTokens))
 	}
 
 	// /v1beta routes (Gemini native API)
 	// Note: Gemini handler extracts model from URL path, so fallback logic needs special handling
+	// T-024: Migrated Gemini v1beta routes to unified routing wrapper
 	v1betaAmp := provider.Group("/v1beta")
 	{
 		v1betaAmp.GET("/models", geminiHandlers.GeminiModels)
-		v1betaAmp.POST("/models/:action", fallbackHandler.WrapHandler(geminiHandlers.GeminiHandler))
-		v1betaAmp.GET("/models/:action", geminiHandlers.GeminiGetHandler)
+		v1betaAmp.POST("/models/*action", routingWrapper.Wrap(geminiHandlers.GeminiHandler))
+		v1betaAmp.GET("/models/*action", geminiHandlers.GeminiGetHandler)
 	}
 }

internal/api/modules/amp/routes_test.go

@@ -32,7 +32,9 @@ func TestRegisterManagementRoutes(t *testing.T) {
 	m.setProxy(proxy)
 
 	base := &handlers.BaseAPIHandler{}
-	m.registerManagementRoutes(r, base)
+	m.registerManagementRoutes(r, base, nil)
+	srv := httptest.NewServer(r)
+	defer srv.Close()
 
 	managementPaths := []struct {
 		path   string
@@ -63,11 +65,17 @@ func TestRegisterManagementRoutes(t *testing.T) {
 	for _, path := range managementPaths {
 		t.Run(path.path, func(t *testing.T) {
 			proxyCalled = false
-			req := httptest.NewRequest(path.method, path.path, nil)
-			w := httptest.NewRecorder()
-			r.ServeHTTP(w, req)
+			req, err := http.NewRequest(path.method, srv.URL+path.path, nil)
+			if err != nil {
+				t.Fatalf("failed to build request: %v", err)
+			}
+			resp, err := http.DefaultClient.Do(req)
+			if err != nil {
+				t.Fatalf("request failed: %v", err)
+			}
+			defer resp.Body.Close()
 
-			if w.Code == http.StatusNotFound {
+			if resp.StatusCode == http.StatusNotFound {
 				t.Fatalf("route %s not registered", path.path)
 			}
 			if !proxyCalled {

internal/api/modules/amp/secret.go

@@ -9,6 +9,9 @@ import (
 	"strings"
 	"sync"
 	"time"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	log "github.com/sirupsen/logrus"
 )
 
 // SecretSource provides Amp API keys with configurable precedence and caching
@@ -164,3 +167,82 @@ func NewStaticSecretSource(key string) *StaticSecretSource {
 func (s *StaticSecretSource) Get(ctx context.Context) (string, error) {
 	return s.key, nil
 }
+
+// MappedSecretSource wraps a default SecretSource and adds per-client API key mapping.
+// When a request context contains a client API key that matches a configured mapping,
+// the corresponding upstream key is returned. Otherwise, falls back to the default source.
+type MappedSecretSource struct {
+	defaultSource SecretSource
+	mu            sync.RWMutex
+	lookup        map[string]string // clientKey -> upstreamKey
+}
+
+// NewMappedSecretSource creates a MappedSecretSource wrapping the given default source.
+func NewMappedSecretSource(defaultSource SecretSource) *MappedSecretSource {
+	return &MappedSecretSource{
+		defaultSource: defaultSource,
+		lookup:        make(map[string]string),
+	}
+}
+
+// Get retrieves the Amp API key, checking per-client mappings first.
+// If the request context contains a client API key that matches a configured mapping,
+// returns the corresponding upstream key. Otherwise, falls back to the default source.
+func (s *MappedSecretSource) Get(ctx context.Context) (string, error) {
+	// Try to get client API key from request context
+	clientKey := getClientAPIKeyFromContext(ctx)
+	if clientKey != "" {
+		s.mu.RLock()
+		if upstreamKey, ok := s.lookup[clientKey]; ok && upstreamKey != "" {
+			s.mu.RUnlock()
+			return upstreamKey, nil
+		}
+		s.mu.RUnlock()
+	}
+
+	// Fall back to default source
+	return s.defaultSource.Get(ctx)
+}
+
+// UpdateMappings rebuilds the client-to-upstream key mapping from configuration entries.
+// If the same client key appears in multiple entries, logs a warning and uses the first one.
+func (s *MappedSecretSource) UpdateMappings(entries []config.AmpUpstreamAPIKeyEntry) {
+	newLookup := make(map[string]string)
+
+	for _, entry := range entries {
+		upstreamKey := strings.TrimSpace(entry.UpstreamAPIKey)
+		if upstreamKey == "" {
+			continue
+		}
+		for _, clientKey := range entry.APIKeys {
+			trimmedKey := strings.TrimSpace(clientKey)
+			if trimmedKey == "" {
+				continue
+			}
+			if _, exists := newLookup[trimmedKey]; exists {
+				// Log warning for duplicate client key, first one wins
+				log.Warnf("amp upstream-api-keys: client API key appears in multiple entries; using first mapping.")
+				continue
+			}
+			newLookup[trimmedKey] = upstreamKey
+		}
+	}
+
+	s.mu.Lock()
+	s.lookup = newLookup
+	s.mu.Unlock()
+}
+
+// UpdateDefaultExplicitKey updates the explicit key on the underlying MultiSourceSecret (if applicable).
+func (s *MappedSecretSource) UpdateDefaultExplicitKey(key string) {
+	if ms, ok := s.defaultSource.(*MultiSourceSecret); ok {
+		ms.UpdateExplicitKey(key)
+	}
+}
+
+// InvalidateCache invalidates cache on the underlying MultiSourceSecret (if applicable).
+func (s *MappedSecretSource) InvalidateCache() {
+	if ms, ok := s.defaultSource.(*MultiSourceSecret); ok {
+		ms.InvalidateCache()
+	}
+}

internal/api/modules/amp/secret_test.go

@@ -8,6 +8,10 @@ import (
 	"sync"
 	"testing"
 	"time"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	log "github.com/sirupsen/logrus"
+	"github.com/sirupsen/logrus/hooks/test"
 )
 
 func TestMultiSourceSecret_PrecedenceOrder(t *testing.T) {
@@ -278,3 +282,85 @@ func TestMultiSourceSecret_CacheEmptyResult(t *testing.T) {
 		t.Fatalf("after cache expiry, expected new-value, got %q", got3)
 	}
 }
+
+func TestMappedSecretSource_UsesMappingFromContext(t *testing.T) {
+	defaultSource := NewStaticSecretSource("default")
+	s := NewMappedSecretSource(defaultSource)
+	s.UpdateMappings([]config.AmpUpstreamAPIKeyEntry{
+		{
+			UpstreamAPIKey: "u1",
+			APIKeys:        []string{"k1"},
+		},
+	})
+
+	ctx := context.WithValue(context.Background(), clientAPIKeyContextKey{}, "k1")
+	got, err := s.Get(ctx)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if got != "u1" {
+		t.Fatalf("want u1, got %q", got)
+	}
+
+	ctx = context.WithValue(context.Background(), clientAPIKeyContextKey{}, "k2")
+	got, err = s.Get(ctx)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if got != "default" {
+		t.Fatalf("want default fallback, got %q", got)
+	}
+}
+
+func TestMappedSecretSource_DuplicateClientKey_FirstWins(t *testing.T) {
+	defaultSource := NewStaticSecretSource("default")
+	s := NewMappedSecretSource(defaultSource)
+	s.UpdateMappings([]config.AmpUpstreamAPIKeyEntry{
+		{
+			UpstreamAPIKey: "u1",
+			APIKeys:        []string{"k1"},
+		},
+		{
+			UpstreamAPIKey: "u2",
+			APIKeys:        []string{"k1"},
+		},
+	})
+
+	ctx := context.WithValue(context.Background(), clientAPIKeyContextKey{}, "k1")
+	got, err := s.Get(ctx)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if got != "u1" {
+		t.Fatalf("want u1 (first wins), got %q", got)
+	}
+}
+
+func TestMappedSecretSource_DuplicateClientKey_LogsWarning(t *testing.T) {
+	hook := test.NewLocal(log.StandardLogger())
+	defer hook.Reset()
+
+	defaultSource := NewStaticSecretSource("default")
+	s := NewMappedSecretSource(defaultSource)
+	s.UpdateMappings([]config.AmpUpstreamAPIKeyEntry{
+		{
+			UpstreamAPIKey: "u1",
+			APIKeys:        []string{"k1"},
+		},
+		{
+			UpstreamAPIKey: "u2",
+			APIKeys:        []string{"k1"},
+		},
+	})
+
+	foundWarning := false
+	for _, entry := range hook.AllEntries() {
+		if entry.Level == log.WarnLevel && entry.Message == "amp upstream-api-keys: client API key appears in multiple entries; using first mapping." {
+			foundWarning = true
+			break
+		}
+	}
+	if !foundWarning {
+		t.Fatal("expected warning log for duplicate client key, but none was found")
+	}
+}

internal/api/server.go

@@ -12,6 +12,7 @@ import (
 	"net/http"
 	"os"
 	"path/filepath"
+	"reflect"
 	"strings"
 	"sync"
 	"sync/atomic"
@@ -26,6 +27,7 @@ import (
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/logging"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/managementasset"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/misc"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/usage"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	sdkaccess "github.com/router-for-me/CLIProxyAPI/v6/sdk/access"
@@ -33,6 +35,7 @@ import (
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers/claude"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers/gemini"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/api/handlers/openai"
+	sdkAuth "github.com/router-for-me/CLIProxyAPI/v6/sdk/auth"
 	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
 	log "github.com/sirupsen/logrus"
 	"gopkg.in/yaml.v3"
@@ -57,9 +60,9 @@ type ServerOption func(*serverOptionConfig)
 func defaultRequestLoggerFactory(cfg *config.Config, configPath string) logging.RequestLogger {
 	configDir := filepath.Dir(configPath)
 	if base := util.WritablePath(); base != "" {
-		return logging.NewFileRequestLogger(cfg.RequestLog, filepath.Join(base, "logs"), configDir)
+		return logging.NewFileRequestLogger(cfg.RequestLog, filepath.Join(base, "logs"), configDir, cfg.ErrorLogsMaxFiles)
 	}
-	return logging.NewFileRequestLogger(cfg.RequestLog, "logs", configDir)
+	return logging.NewFileRequestLogger(cfg.RequestLog, "logs", configDir, cfg.ErrorLogsMaxFiles)
 }
 
 // WithMiddleware appends additional Gin middleware during server construction.
@@ -209,6 +212,7 @@ func NewServer(cfg *config.Config, authManager *auth.Manager, accessManager *sdk
 	// Resolve logs directory relative to the configuration file directory.
 	var requestLogger logging.RequestLogger
 	var toggle func(bool)
+	if !cfg.CommercialMode {
 		if optionState.requestLoggerFactory != nil {
 			requestLogger = optionState.requestLoggerFactory(cfg, configFilePath)
 		}
@@ -218,6 +222,7 @@ func NewServer(cfg *config.Config, authManager *auth.Manager, accessManager *sdk
 				toggle = setter.SetEnabled
 			}
 		}
+	}
 
 	engine.Use(corsMiddleware())
 	wd, err := os.Getwd()
@@ -230,13 +235,9 @@ func NewServer(cfg *config.Config, authManager *auth.Manager, accessManager *sdk
 	envManagementSecret := envAdminPasswordSet && envAdminPassword != ""
 
 	// Create server instance
-	providerNames := make([]string, 0, len(cfg.OpenAICompatibility))
-	for _, p := range cfg.OpenAICompatibility {
-		providerNames = append(providerNames, p.Name)
-	}
 	s := &Server{
 		engine:              engine,
-		handlers:            handlers.NewBaseAPIHandlers(&cfg.SDKConfig, authManager, providerNames),
+		handlers:            handlers.NewBaseAPIHandlers(&cfg.SDKConfig, authManager),
 		cfg:                 cfg,
 		accessManager:       accessManager,
 		requestLogger:       requestLogger,
@@ -255,15 +256,13 @@ func NewServer(cfg *config.Config, authManager *auth.Manager, accessManager *sdk
 	}
 	managementasset.SetCurrentConfig(cfg)
 	auth.SetQuotaCooldownDisabled(cfg.DisableCooling)
+	misc.SetCodexInstructionsEnabled(cfg.CodexInstructionsEnabled)
 	// Initialize management handler
 	s.mgmt = managementHandlers.NewHandler(cfg, configFilePath, authManager)
 	if optionState.localPassword != "" {
 		s.mgmt.SetLocalPassword(optionState.localPassword)
 	}
-	logDir := filepath.Join(s.currentPath, "logs")
-	if base := util.WritablePath(); base != "" {
-		logDir = filepath.Join(base, "logs")
-	}
+	logDir := logging.ResolveLogDirectory(cfg)
 	s.mgmt.SetLogDirectory(logDir)
 	s.localPassword = optionState.localPassword
 
@@ -327,6 +326,7 @@ func (s *Server) setupRoutes() {
 		v1.POST("/messages", claudeCodeHandlers.ClaudeMessages)
 		v1.POST("/messages/count_tokens", claudeCodeHandlers.ClaudeCountTokens)
 		v1.POST("/responses", openaiResponsesHandlers.Responses)
+		v1.POST("/responses/compact", openaiResponsesHandlers.Compact)
 	}
 
 	// Gemini compatible API routes
@@ -334,8 +334,8 @@ func (s *Server) setupRoutes() {
 	v1beta.Use(AuthMiddleware(s.accessManager))
 	{
 		v1beta.GET("/models", geminiHandlers.GeminiModels)
-		v1beta.POST("/models/:action", geminiHandlers.GeminiHandler)
-		v1beta.GET("/models/:action", geminiHandlers.GeminiGetHandler)
+		v1beta.POST("/models/*action", geminiHandlers.GeminiHandler)
+		v1beta.GET("/models/*action", geminiHandlers.GeminiGetHandler)
 	}
 
 	// Root endpoint
@@ -358,10 +358,11 @@ func (s *Server) setupRoutes() {
 		code := c.Query("code")
 		state := c.Query("state")
 		errStr := c.Query("error")
-		// Persist to a temporary file keyed by state
+		if errStr == "" {
+			errStr = c.Query("error_description")
+		}
 		if state != "" {
-			file := fmt.Sprintf("%s/.oauth-anthropic-%s.oauth", s.cfg.AuthDir, state)
-			_ = os.WriteFile(file, []byte(fmt.Sprintf(`{"code":"%s","state":"%s","error":"%s"}`, code, state, errStr)), 0o600)
+			_, _ = managementHandlers.WriteOAuthCallbackFileForPendingSession(s.cfg.AuthDir, "anthropic", state, code, errStr)
 		}
 		c.Header("Content-Type", "text/html; charset=utf-8")
 		c.String(http.StatusOK, oauthCallbackSuccessHTML)
@@ -371,9 +372,11 @@ func (s *Server) setupRoutes() {
 		code := c.Query("code")
 		state := c.Query("state")
 		errStr := c.Query("error")
+		if errStr == "" {
+			errStr = c.Query("error_description")
+		}
 		if state != "" {
-			file := fmt.Sprintf("%s/.oauth-codex-%s.oauth", s.cfg.AuthDir, state)
-			_ = os.WriteFile(file, []byte(fmt.Sprintf(`{"code":"%s","state":"%s","error":"%s"}`, code, state, errStr)), 0o600)
+			_, _ = managementHandlers.WriteOAuthCallbackFileForPendingSession(s.cfg.AuthDir, "codex", state, code, errStr)
 		}
 		c.Header("Content-Type", "text/html; charset=utf-8")
 		c.String(http.StatusOK, oauthCallbackSuccessHTML)
@@ -383,9 +386,11 @@ func (s *Server) setupRoutes() {
 		code := c.Query("code")
 		state := c.Query("state")
 		errStr := c.Query("error")
+		if errStr == "" {
+			errStr = c.Query("error_description")
+		}
 		if state != "" {
-			file := fmt.Sprintf("%s/.oauth-gemini-%s.oauth", s.cfg.AuthDir, state)
-			_ = os.WriteFile(file, []byte(fmt.Sprintf(`{"code":"%s","state":"%s","error":"%s"}`, code, state, errStr)), 0o600)
+			_, _ = managementHandlers.WriteOAuthCallbackFileForPendingSession(s.cfg.AuthDir, "gemini", state, code, errStr)
 		}
 		c.Header("Content-Type", "text/html; charset=utf-8")
 		c.String(http.StatusOK, oauthCallbackSuccessHTML)
@@ -395,9 +400,11 @@ func (s *Server) setupRoutes() {
 		code := c.Query("code")
 		state := c.Query("state")
 		errStr := c.Query("error")
+		if errStr == "" {
+			errStr = c.Query("error_description")
+		}
 		if state != "" {
-			file := fmt.Sprintf("%s/.oauth-iflow-%s.oauth", s.cfg.AuthDir, state)
-			_ = os.WriteFile(file, []byte(fmt.Sprintf(`{"code":"%s","state":"%s","error":"%s"}`, code, state, errStr)), 0o600)
+			_, _ = managementHandlers.WriteOAuthCallbackFileForPendingSession(s.cfg.AuthDir, "iflow", state, code, errStr)
 		}
 		c.Header("Content-Type", "text/html; charset=utf-8")
 		c.String(http.StatusOK, oauthCallbackSuccessHTML)
@@ -407,9 +414,11 @@ func (s *Server) setupRoutes() {
 		code := c.Query("code")
 		state := c.Query("state")
 		errStr := c.Query("error")
+		if errStr == "" {
+			errStr = c.Query("error_description")
+		}
 		if state != "" {
-			file := fmt.Sprintf("%s/.oauth-antigravity-%s.oauth", s.cfg.AuthDir, state)
-			_ = os.WriteFile(file, []byte(fmt.Sprintf(`{"code":"%s","state":"%s","error":"%s"}`, code, state, errStr)), 0o600)
+			_, _ = managementHandlers.WriteOAuthCallbackFileForPendingSession(s.cfg.AuthDir, "antigravity", state, code, errStr)
 		}
 		c.Header("Content-Type", "text/html; charset=utf-8")
 		c.String(http.StatusOK, oauthCallbackSuccessHTML)
@@ -469,6 +478,8 @@ func (s *Server) registerManagementRoutes() {
 	mgmt.Use(s.managementAvailabilityMiddleware(), s.mgmt.Middleware())
 	{
 		mgmt.GET("/usage", s.mgmt.GetUsageStatistics)
+		mgmt.GET("/usage/export", s.mgmt.ExportUsageStatistics)
+		mgmt.POST("/usage/import", s.mgmt.ImportUsageStatistics)
 		mgmt.GET("/config", s.mgmt.GetConfig)
 		mgmt.GET("/config.yaml", s.mgmt.GetConfigYAML)
 		mgmt.PUT("/config.yaml", s.mgmt.PutConfigYAML)
@@ -482,6 +493,14 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.PUT("/logging-to-file", s.mgmt.PutLoggingToFile)
 		mgmt.PATCH("/logging-to-file", s.mgmt.PutLoggingToFile)
 
+		mgmt.GET("/logs-max-total-size-mb", s.mgmt.GetLogsMaxTotalSizeMB)
+		mgmt.PUT("/logs-max-total-size-mb", s.mgmt.PutLogsMaxTotalSizeMB)
+		mgmt.PATCH("/logs-max-total-size-mb", s.mgmt.PutLogsMaxTotalSizeMB)
+
+		mgmt.GET("/error-logs-max-files", s.mgmt.GetErrorLogsMaxFiles)
+		mgmt.PUT("/error-logs-max-files", s.mgmt.PutErrorLogsMaxFiles)
+		mgmt.PATCH("/error-logs-max-files", s.mgmt.PutErrorLogsMaxFiles)
+
 		mgmt.GET("/usage-statistics-enabled", s.mgmt.GetUsageStatisticsEnabled)
 		mgmt.PUT("/usage-statistics-enabled", s.mgmt.PutUsageStatisticsEnabled)
 		mgmt.PATCH("/usage-statistics-enabled", s.mgmt.PutUsageStatisticsEnabled)
@@ -491,6 +510,8 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.PATCH("/proxy-url", s.mgmt.PutProxyURL)
 		mgmt.DELETE("/proxy-url", s.mgmt.DeleteProxyURL)
 
+		mgmt.POST("/api-call", s.mgmt.APICall)
+
 		mgmt.GET("/quota-exceeded/switch-project", s.mgmt.GetSwitchProject)
 		mgmt.PUT("/quota-exceeded/switch-project", s.mgmt.PutSwitchProject)
 		mgmt.PATCH("/quota-exceeded/switch-project", s.mgmt.PutSwitchProject)
@@ -513,6 +534,7 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.DELETE("/logs", s.mgmt.DeleteLogs)
 		mgmt.GET("/request-error-logs", s.mgmt.GetRequestErrorLogs)
 		mgmt.GET("/request-error-logs/:name", s.mgmt.DownloadRequestErrorLog)
+		mgmt.GET("/request-log-by-id/:id", s.mgmt.GetRequestLogByID)
 		mgmt.GET("/request-log", s.mgmt.GetRequestLog)
 		mgmt.PUT("/request-log", s.mgmt.PutRequestLog)
 		mgmt.PATCH("/request-log", s.mgmt.PutRequestLog)
@@ -539,6 +561,10 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.GET("/ampcode/force-model-mappings", s.mgmt.GetAmpForceModelMappings)
 		mgmt.PUT("/ampcode/force-model-mappings", s.mgmt.PutAmpForceModelMappings)
 		mgmt.PATCH("/ampcode/force-model-mappings", s.mgmt.PutAmpForceModelMappings)
+		mgmt.GET("/ampcode/upstream-api-keys", s.mgmt.GetAmpUpstreamAPIKeys)
+		mgmt.PUT("/ampcode/upstream-api-keys", s.mgmt.PutAmpUpstreamAPIKeys)
+		mgmt.PATCH("/ampcode/upstream-api-keys", s.mgmt.PatchAmpUpstreamAPIKeys)
+		mgmt.DELETE("/ampcode/upstream-api-keys", s.mgmt.DeleteAmpUpstreamAPIKeys)
 
 		mgmt.GET("/request-retry", s.mgmt.GetRequestRetry)
 		mgmt.PUT("/request-retry", s.mgmt.PutRequestRetry)
@@ -547,6 +573,14 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.PUT("/max-retry-interval", s.mgmt.PutMaxRetryInterval)
 		mgmt.PATCH("/max-retry-interval", s.mgmt.PutMaxRetryInterval)
 
+		mgmt.GET("/force-model-prefix", s.mgmt.GetForceModelPrefix)
+		mgmt.PUT("/force-model-prefix", s.mgmt.PutForceModelPrefix)
+		mgmt.PATCH("/force-model-prefix", s.mgmt.PutForceModelPrefix)
+
+		mgmt.GET("/routing/strategy", s.mgmt.GetRoutingStrategy)
+		mgmt.PUT("/routing/strategy", s.mgmt.PutRoutingStrategy)
+		mgmt.PATCH("/routing/strategy", s.mgmt.PutRoutingStrategy)
+
 		mgmt.GET("/claude-api-key", s.mgmt.GetClaudeKeys)
 		mgmt.PUT("/claude-api-key", s.mgmt.PutClaudeKeys)
 		mgmt.PATCH("/claude-api-key", s.mgmt.PatchClaudeKey)
@@ -562,15 +596,28 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.PATCH("/openai-compatibility", s.mgmt.PatchOpenAICompat)
 		mgmt.DELETE("/openai-compatibility", s.mgmt.DeleteOpenAICompat)
 
+		mgmt.GET("/vertex-api-key", s.mgmt.GetVertexCompatKeys)
+		mgmt.PUT("/vertex-api-key", s.mgmt.PutVertexCompatKeys)
+		mgmt.PATCH("/vertex-api-key", s.mgmt.PatchVertexCompatKey)
+		mgmt.DELETE("/vertex-api-key", s.mgmt.DeleteVertexCompatKey)
+
 		mgmt.GET("/oauth-excluded-models", s.mgmt.GetOAuthExcludedModels)
 		mgmt.PUT("/oauth-excluded-models", s.mgmt.PutOAuthExcludedModels)
 		mgmt.PATCH("/oauth-excluded-models", s.mgmt.PatchOAuthExcludedModels)
 		mgmt.DELETE("/oauth-excluded-models", s.mgmt.DeleteOAuthExcludedModels)
 
+		mgmt.GET("/oauth-model-alias", s.mgmt.GetOAuthModelAlias)
+		mgmt.PUT("/oauth-model-alias", s.mgmt.PutOAuthModelAlias)
+		mgmt.PATCH("/oauth-model-alias", s.mgmt.PatchOAuthModelAlias)
+		mgmt.DELETE("/oauth-model-alias", s.mgmt.DeleteOAuthModelAlias)
+
 		mgmt.GET("/auth-files", s.mgmt.ListAuthFiles)
+		mgmt.GET("/auth-files/models", s.mgmt.GetAuthFileModels)
+		mgmt.GET("/model-definitions/:channel", s.mgmt.GetStaticModelDefinitions)
 		mgmt.GET("/auth-files/download", s.mgmt.DownloadAuthFile)
 		mgmt.POST("/auth-files", s.mgmt.UploadAuthFile)
 		mgmt.DELETE("/auth-files", s.mgmt.DeleteAuthFile)
+		mgmt.PATCH("/auth-files/status", s.mgmt.PatchAuthFileStatus)
 		mgmt.POST("/vertex/import", s.mgmt.ImportVertexCredential)
 
 		mgmt.GET("/anthropic-auth-url", s.mgmt.RequestAnthropicToken)
@@ -580,6 +627,7 @@ func (s *Server) registerManagementRoutes() {
 		mgmt.GET("/qwen-auth-url", s.mgmt.RequestQwenToken)
 		mgmt.GET("/iflow-auth-url", s.mgmt.RequestIFlowToken)
 		mgmt.POST("/iflow-auth-url", s.mgmt.RequestIFlowCookieToken)
+		mgmt.POST("/oauth-callback", s.mgmt.PostOAuthCallback)
 		mgmt.GET("/get-auth-status", s.mgmt.GetAuthStatus)
 	}
 }
@@ -608,7 +656,7 @@ func (s *Server) serveManagementControlPanel(c *gin.Context) {
 
 	if _, err := os.Stat(filePath); err != nil {
 		if os.IsNotExist(err) {
-			go managementasset.EnsureLatestManagementHTML(context.Background(), managementasset.StaticDir(s.configFilePath), cfg.ProxyURL)
+			go managementasset.EnsureLatestManagementHTML(context.Background(), managementasset.StaticDir(s.configFilePath), cfg.ProxyURL, cfg.RemoteManagement.PanelGitHubRepository)
 			c.AbortWithStatus(http.StatusNotFound)
 			return
 		}
@@ -837,12 +885,21 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 		}
 	}
 
-	if oldCfg != nil && oldCfg.LoggingToFile != cfg.LoggingToFile {
-		if err := logging.ConfigureLogOutput(cfg.LoggingToFile); err != nil {
+	if oldCfg == nil || oldCfg.LoggingToFile != cfg.LoggingToFile || oldCfg.LogsMaxTotalSizeMB != cfg.LogsMaxTotalSizeMB {
+		if err := logging.ConfigureLogOutput(cfg); err != nil {
 			log.Errorf("failed to reconfigure log output: %v", err)
 		} else {
+			if oldCfg == nil {
+				log.Debug("log output configuration refreshed")
+			} else {
+				if oldCfg.LoggingToFile != cfg.LoggingToFile {
 					log.Debugf("logging_to_file updated from %t to %t", oldCfg.LoggingToFile, cfg.LoggingToFile)
 				}
+				if oldCfg.LogsMaxTotalSizeMB != cfg.LogsMaxTotalSizeMB {
+					log.Debugf("logs_max_total_size_mb updated from %d to %d", oldCfg.LogsMaxTotalSizeMB, cfg.LogsMaxTotalSizeMB)
+				}
+			}
+		}
 	}
 
 	if oldCfg == nil || oldCfg.UsageStatisticsEnabled != cfg.UsageStatisticsEnabled {
@@ -854,6 +911,15 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 		}
 	}
 
+	if s.requestLogger != nil && (oldCfg == nil || oldCfg.ErrorLogsMaxFiles != cfg.ErrorLogsMaxFiles) {
+		if setter, ok := s.requestLogger.(interface{ SetErrorLogsMaxFiles(int) }); ok {
+			setter.SetErrorLogsMaxFiles(cfg.ErrorLogsMaxFiles)
+		}
+		if oldCfg != nil {
+			log.Debugf("error_logs_max_files updated from %d to %d", oldCfg.ErrorLogsMaxFiles, cfg.ErrorLogsMaxFiles)
+		}
+	}
+
 	if oldCfg == nil || oldCfg.DisableCooling != cfg.DisableCooling {
 		auth.SetQuotaCooldownDisabled(cfg.DisableCooling)
 		if oldCfg != nil {
@@ -862,6 +928,16 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 			log.Debugf("disable_cooling toggled to %t", cfg.DisableCooling)
 		}
 	}
+
+	if oldCfg == nil || oldCfg.CodexInstructionsEnabled != cfg.CodexInstructionsEnabled {
+		misc.SetCodexInstructionsEnabled(cfg.CodexInstructionsEnabled)
+		if oldCfg != nil {
+			log.Debugf("codex_instructions_enabled updated from %t to %t", oldCfg.CodexInstructionsEnabled, cfg.CodexInstructionsEnabled)
+		} else {
+			log.Debugf("codex_instructions_enabled toggled to %t", cfg.CodexInstructionsEnabled)
+		}
+	}
+
 	if s.handlers != nil && s.handlers.AuthManager != nil {
 		s.handlers.AuthManager.SetRetryConfig(cfg.RequestRetry, time.Duration(cfg.MaxRetryInterval)*time.Second)
 	}
@@ -918,24 +994,20 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 	// Save YAML snapshot for next comparison
 	s.oldConfigYaml, _ = yaml.Marshal(cfg)
 
-	providerNames := make([]string, 0, len(cfg.OpenAICompatibility))
-	for _, p := range cfg.OpenAICompatibility {
-		providerNames = append(providerNames, p.Name)
-	}
-	s.handlers.OpenAICompatProviders = providerNames
-
 	s.handlers.UpdateClients(&cfg.SDKConfig)
 
 	if !cfg.RemoteManagement.DisableControlPanel {
 		staticDir := managementasset.StaticDir(s.configFilePath)
-		go managementasset.EnsureLatestManagementHTML(context.Background(), staticDir, cfg.ProxyURL)
+		go managementasset.EnsureLatestManagementHTML(context.Background(), staticDir, cfg.ProxyURL, cfg.RemoteManagement.PanelGitHubRepository)
 	}
 	if s.mgmt != nil {
 		s.mgmt.SetConfig(cfg)
 		s.mgmt.SetAuthManager(s.handlers.AuthManager)
 	}
 
-	// Notify Amp module of config changes (for model mapping hot-reload)
+	// Notify Amp module when Amp config or OAuth model aliases have changed.
+	ampConfigChanged := oldCfg == nil || !reflect.DeepEqual(oldCfg.AmpCode, cfg.AmpCode) || !reflect.DeepEqual(oldCfg.OAuthModelAlias, cfg.OAuthModelAlias)
+	if ampConfigChanged {
 		if s.ampModule != nil {
 			log.Debugf("triggering amp module config update")
 			if err := s.ampModule.OnConfigUpdated(cfg); err != nil {
@@ -944,9 +1016,14 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 		} else {
 			log.Warnf("amp module is nil, skipping config update")
 		}
+	}
 
-	// Count client sources from configuration and auth directory
-	authFiles := util.CountAuthFiles(cfg.AuthDir)
+	// Count client sources from configuration and auth store.
+	tokenStore := sdkAuth.GetTokenStore()
+	if dirSetter, ok := tokenStore.(interface{ SetBaseDir(string) }); ok {
+		dirSetter.SetBaseDir(cfg.AuthDir)
+	}
+	authEntries := util.CountAuthFiles(context.Background(), tokenStore)
 	geminiAPIKeyCount := len(cfg.GeminiKey)
 	claudeAPIKeyCount := len(cfg.ClaudeKey)
 	codexAPIKeyCount := len(cfg.CodexKey)
@@ -957,10 +1034,10 @@ func (s *Server) UpdateClients(cfg *config.Config) {
 		openAICompatCount += len(entry.APIKeyEntries)
 	}
 
-	total := authFiles + geminiAPIKeyCount + claudeAPIKeyCount + codexAPIKeyCount + vertexAICompatCount + openAICompatCount
-	fmt.Printf("server clients and configuration updated: %d clients (%d auth files + %d Gemini API keys + %d Claude API keys + %d Codex keys + %d Vertex-compat + %d OpenAI-compat)\n",
+	total := authEntries + geminiAPIKeyCount + claudeAPIKeyCount + codexAPIKeyCount + vertexAICompatCount + openAICompatCount
+	fmt.Printf("server clients and configuration updated: %d clients (%d auth entries + %d Gemini API keys + %d Claude API keys + %d Codex keys + %d Vertex-compat + %d OpenAI-compat)\n",
 		total,
-		authFiles,
+		authEntries,
 		geminiAPIKeyCount,
 		claudeAPIKeyCount,
 		codexAPIKeyCount,

internal/auth/antigravity/auth.go

@@ -0,0 +1,344 @@
+// Package antigravity provides OAuth2 authentication functionality for the Antigravity provider.
+package antigravity
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"strings"
+	"time"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
+	log "github.com/sirupsen/logrus"
+)
+
+// TokenResponse represents OAuth token response from Google
+type TokenResponse struct {
+	AccessToken  string `json:"access_token"`
+	RefreshToken string `json:"refresh_token"`
+	ExpiresIn    int64  `json:"expires_in"`
+	TokenType    string `json:"token_type"`
+}
+
+// userInfo represents Google user profile
+type userInfo struct {
+	Email string `json:"email"`
+}
+
+// AntigravityAuth handles Antigravity OAuth authentication
+type AntigravityAuth struct {
+	httpClient *http.Client
+}
+
+// NewAntigravityAuth creates a new Antigravity auth service.
+func NewAntigravityAuth(cfg *config.Config, httpClient *http.Client) *AntigravityAuth {
+	if httpClient != nil {
+		return &AntigravityAuth{httpClient: httpClient}
+	}
+	if cfg == nil {
+		cfg = &config.Config{}
+	}
+	return &AntigravityAuth{
+		httpClient: util.SetProxy(&cfg.SDKConfig, &http.Client{}),
+	}
+}
+
+// BuildAuthURL generates the OAuth authorization URL.
+func (o *AntigravityAuth) BuildAuthURL(state, redirectURI string) string {
+	if strings.TrimSpace(redirectURI) == "" {
+		redirectURI = fmt.Sprintf("http://localhost:%d/oauth-callback", CallbackPort)
+	}
+	params := url.Values{}
+	params.Set("access_type", "offline")
+	params.Set("client_id", ClientID)
+	params.Set("prompt", "consent")
+	params.Set("redirect_uri", redirectURI)
+	params.Set("response_type", "code")
+	params.Set("scope", strings.Join(Scopes, " "))
+	params.Set("state", state)
+	return AuthEndpoint + "?" + params.Encode()
+}
+
+// ExchangeCodeForTokens exchanges authorization code for access and refresh tokens
+func (o *AntigravityAuth) ExchangeCodeForTokens(ctx context.Context, code, redirectURI string) (*TokenResponse, error) {
+	data := url.Values{}
+	data.Set("code", code)
+	data.Set("client_id", ClientID)
+	data.Set("client_secret", ClientSecret)
+	data.Set("redirect_uri", redirectURI)
+	data.Set("grant_type", "authorization_code")
+
+	req, err := http.NewRequestWithContext(ctx, http.MethodPost, TokenEndpoint, strings.NewReader(data.Encode()))
+	if err != nil {
+		return nil, fmt.Errorf("antigravity token exchange: create request: %w", err)
+	}
+	req.Header.Set("Content-Type", "application/x-www-form-urlencoded")
+
+	resp, errDo := o.httpClient.Do(req)
+	if errDo != nil {
+		return nil, fmt.Errorf("antigravity token exchange: execute request: %w", errDo)
+	}
+	defer func() {
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("antigravity token exchange: close body error: %v", errClose)
+		}
+	}()
+
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
+		bodyBytes, errRead := io.ReadAll(io.LimitReader(resp.Body, 8<<10))
+		if errRead != nil {
+			return nil, fmt.Errorf("antigravity token exchange: read response: %w", errRead)
+		}
+		body := strings.TrimSpace(string(bodyBytes))
+		if body == "" {
+			return nil, fmt.Errorf("antigravity token exchange: request failed: status %d", resp.StatusCode)
+		}
+		return nil, fmt.Errorf("antigravity token exchange: request failed: status %d: %s", resp.StatusCode, body)
+	}
+
+	var token TokenResponse
+	if errDecode := json.NewDecoder(resp.Body).Decode(&token); errDecode != nil {
+		return nil, fmt.Errorf("antigravity token exchange: decode response: %w", errDecode)
+	}
+	return &token, nil
+}
+
+// FetchUserInfo retrieves user email from Google
+func (o *AntigravityAuth) FetchUserInfo(ctx context.Context, accessToken string) (string, error) {
+	accessToken = strings.TrimSpace(accessToken)
+	if accessToken == "" {
+		return "", fmt.Errorf("antigravity userinfo: missing access token")
+	}
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, UserInfoEndpoint, nil)
+	if err != nil {
+		return "", fmt.Errorf("antigravity userinfo: create request: %w", err)
+	}
+	req.Header.Set("Authorization", "Bearer "+accessToken)
+
+	resp, errDo := o.httpClient.Do(req)
+	if errDo != nil {
+		return "", fmt.Errorf("antigravity userinfo: execute request: %w", errDo)
+	}
+	defer func() {
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("antigravity userinfo: close body error: %v", errClose)
+		}
+	}()
+
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
+		bodyBytes, errRead := io.ReadAll(io.LimitReader(resp.Body, 8<<10))
+		if errRead != nil {
+			return "", fmt.Errorf("antigravity userinfo: read response: %w", errRead)
+		}
+		body := strings.TrimSpace(string(bodyBytes))
+		if body == "" {
+			return "", fmt.Errorf("antigravity userinfo: request failed: status %d", resp.StatusCode)
+		}
+		return "", fmt.Errorf("antigravity userinfo: request failed: status %d: %s", resp.StatusCode, body)
+	}
+	var info userInfo
+	if errDecode := json.NewDecoder(resp.Body).Decode(&info); errDecode != nil {
+		return "", fmt.Errorf("antigravity userinfo: decode response: %w", errDecode)
+	}
+	email := strings.TrimSpace(info.Email)
+	if email == "" {
+		return "", fmt.Errorf("antigravity userinfo: response missing email")
+	}
+	return email, nil
+}
+
+// FetchProjectID retrieves the project ID for the authenticated user via loadCodeAssist
+func (o *AntigravityAuth) FetchProjectID(ctx context.Context, accessToken string) (string, error) {
+	loadReqBody := map[string]any{
+		"metadata": map[string]string{
+			"ideType":    "ANTIGRAVITY",
+			"platform":   "PLATFORM_UNSPECIFIED",
+			"pluginType": "GEMINI",
+		},
+	}
+
+	rawBody, errMarshal := json.Marshal(loadReqBody)
+	if errMarshal != nil {
+		return "", fmt.Errorf("marshal request body: %w", errMarshal)
+	}
+
+	endpointURL := fmt.Sprintf("%s/%s:loadCodeAssist", APIEndpoint, APIVersion)
+	req, err := http.NewRequestWithContext(ctx, http.MethodPost, endpointURL, strings.NewReader(string(rawBody)))
+	if err != nil {
+		return "", fmt.Errorf("create request: %w", err)
+	}
+	req.Header.Set("Authorization", "Bearer "+accessToken)
+	req.Header.Set("Content-Type", "application/json")
+	req.Header.Set("User-Agent", APIUserAgent)
+	req.Header.Set("X-Goog-Api-Client", APIClient)
+	req.Header.Set("Client-Metadata", ClientMetadata)
+
+	resp, errDo := o.httpClient.Do(req)
+	if errDo != nil {
+		return "", fmt.Errorf("execute request: %w", errDo)
+	}
+	defer func() {
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("antigravity loadCodeAssist: close body error: %v", errClose)
+		}
+	}()
+
+	bodyBytes, errRead := io.ReadAll(resp.Body)
+	if errRead != nil {
+		return "", fmt.Errorf("read response: %w", errRead)
+	}
+
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
+		return "", fmt.Errorf("request failed with status %d: %s", resp.StatusCode, strings.TrimSpace(string(bodyBytes)))
+	}
+
+	var loadResp map[string]any
+	if errDecode := json.Unmarshal(bodyBytes, &loadResp); errDecode != nil {
+		return "", fmt.Errorf("decode response: %w", errDecode)
+	}
+
+	// Extract projectID from response
+	projectID := ""
+	if id, ok := loadResp["cloudaicompanionProject"].(string); ok {
+		projectID = strings.TrimSpace(id)
+	}
+	if projectID == "" {
+		if projectMap, ok := loadResp["cloudaicompanionProject"].(map[string]any); ok {
+			if id, okID := projectMap["id"].(string); okID {
+				projectID = strings.TrimSpace(id)
+			}
+		}
+	}
+
+	if projectID == "" {
+		tierID := "legacy-tier"
+		if tiers, okTiers := loadResp["allowedTiers"].([]any); okTiers {
+			for _, rawTier := range tiers {
+				tier, okTier := rawTier.(map[string]any)
+				if !okTier {
+					continue
+				}
+				if isDefault, okDefault := tier["isDefault"].(bool); okDefault && isDefault {
+					if id, okID := tier["id"].(string); okID && strings.TrimSpace(id) != "" {
+						tierID = strings.TrimSpace(id)
+						break
+					}
+				}
+			}
+		}
+
+		projectID, err = o.OnboardUser(ctx, accessToken, tierID)
+		if err != nil {
+			return "", err
+		}
+		return projectID, nil
+	}
+
+	return projectID, nil
+}
+
+// OnboardUser attempts to fetch the project ID via onboardUser by polling for completion
+func (o *AntigravityAuth) OnboardUser(ctx context.Context, accessToken, tierID string) (string, error) {
+	log.Infof("Antigravity: onboarding user with tier: %s", tierID)
+	requestBody := map[string]any{
+		"tierId": tierID,
+		"metadata": map[string]string{
+			"ideType":    "ANTIGRAVITY",
+			"platform":   "PLATFORM_UNSPECIFIED",
+			"pluginType": "GEMINI",
+		},
+	}
+
+	rawBody, errMarshal := json.Marshal(requestBody)
+	if errMarshal != nil {
+		return "", fmt.Errorf("marshal request body: %w", errMarshal)
+	}
+
+	maxAttempts := 5
+	for attempt := 1; attempt <= maxAttempts; attempt++ {
+		log.Debugf("Polling attempt %d/%d", attempt, maxAttempts)
+
+		reqCtx := ctx
+		var cancel context.CancelFunc
+		if reqCtx == nil {
+			reqCtx = context.Background()
+		}
+		reqCtx, cancel = context.WithTimeout(reqCtx, 30*time.Second)
+
+		endpointURL := fmt.Sprintf("%s/%s:onboardUser", APIEndpoint, APIVersion)
+		req, errRequest := http.NewRequestWithContext(reqCtx, http.MethodPost, endpointURL, strings.NewReader(string(rawBody)))
+		if errRequest != nil {
+			cancel()
+			return "", fmt.Errorf("create request: %w", errRequest)
+		}
+		req.Header.Set("Authorization", "Bearer "+accessToken)
+		req.Header.Set("Content-Type", "application/json")
+		req.Header.Set("User-Agent", APIUserAgent)
+		req.Header.Set("X-Goog-Api-Client", APIClient)
+		req.Header.Set("Client-Metadata", ClientMetadata)
+
+		resp, errDo := o.httpClient.Do(req)
+		if errDo != nil {
+			cancel()
+			return "", fmt.Errorf("execute request: %w", errDo)
+		}
+
+		bodyBytes, errRead := io.ReadAll(resp.Body)
+		if errClose := resp.Body.Close(); errClose != nil {
+			log.Errorf("close body error: %v", errClose)
+		}
+		cancel()
+
+		if errRead != nil {
+			return "", fmt.Errorf("read response: %w", errRead)
+		}
+
+		if resp.StatusCode == http.StatusOK {
+			var data map[string]any
+			if errDecode := json.Unmarshal(bodyBytes, &data); errDecode != nil {
+				return "", fmt.Errorf("decode response: %w", errDecode)
+			}
+
+			if done, okDone := data["done"].(bool); okDone && done {
+				projectID := ""
+				if responseData, okResp := data["response"].(map[string]any); okResp {
+					switch projectValue := responseData["cloudaicompanionProject"].(type) {
+					case map[string]any:
+						if id, okID := projectValue["id"].(string); okID {
+							projectID = strings.TrimSpace(id)
+						}
+					case string:
+						projectID = strings.TrimSpace(projectValue)
+					}
+				}
+
+				if projectID != "" {
+					log.Infof("Successfully fetched project_id: %s", projectID)
+					return projectID, nil
+				}
+
+				return "", fmt.Errorf("no project_id in response")
+			}
+
+			time.Sleep(2 * time.Second)
+			continue
+		}
+
+		responsePreview := strings.TrimSpace(string(bodyBytes))
+		if len(responsePreview) > 500 {
+			responsePreview = responsePreview[:500]
+		}
+
+		responseErr := responsePreview
+		if len(responseErr) > 200 {
+			responseErr = responseErr[:200]
+		}
+		return "", fmt.Errorf("http %d: %s", resp.StatusCode, responseErr)
+	}
+
+	return "", nil
+}

internal/auth/antigravity/constants.go

@@ -0,0 +1,34 @@
+// Package antigravity provides OAuth2 authentication functionality for the Antigravity provider.
+package antigravity
+
+// OAuth client credentials and configuration
+const (
+	ClientID     = "1071006060591-tmhssin2h21lcre235vtolojh4g403ep.apps.googleusercontent.com"
+	ClientSecret = "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"
+	CallbackPort = 51121
+)
+
+// Scopes defines the OAuth scopes required for Antigravity authentication
+var Scopes = []string{
+	"https://www.googleapis.com/auth/cloud-platform",
+	"https://www.googleapis.com/auth/userinfo.email",
+	"https://www.googleapis.com/auth/userinfo.profile",
+	"https://www.googleapis.com/auth/cclog",
+	"https://www.googleapis.com/auth/experimentsandconfigs",
+}
+
+// OAuth2 endpoints for Google authentication
+const (
+	TokenEndpoint    = "https://oauth2.googleapis.com/token"
+	AuthEndpoint     = "https://accounts.google.com/o/oauth2/v2/auth"
+	UserInfoEndpoint = "https://www.googleapis.com/oauth2/v1/userinfo?alt=json"
+)
+
+// Antigravity API configuration
+const (
+	APIEndpoint    = "https://cloudcode-pa.googleapis.com"
+	APIVersion     = "v1internal"
+	APIUserAgent   = "google-api-nodejs-client/9.15.1"
+	APIClient      = "google-cloud-sdk vscode_cloudshelleditor/0.1"
+	ClientMetadata = `{"ideType":"IDE_UNSPECIFIED","platform":"PLATFORM_UNSPECIFIED","pluginType":"GEMINI"}`
+)

internal/auth/antigravity/filename.go

@@ -0,0 +1,16 @@
+package antigravity
+
+import (
+	"fmt"
+	"strings"
+)
+
+// CredentialFileName returns the filename used to persist Antigravity credentials.
+// It uses the email as a suffix to disambiguate accounts.
+func CredentialFileName(email string) string {
+	email = strings.TrimSpace(email)
+	if email == "" {
+		return "antigravity.json"
+	}
+	return fmt.Sprintf("antigravity-%s.json", email)
+}

internal/auth/claude/anthropic_auth.go

@@ -14,15 +14,15 @@ import (
 	"time"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
-	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	log "github.com/sirupsen/logrus"
 )
 
+// OAuth configuration constants for Claude/Anthropic
 const (
-	anthropicAuthURL  = "https://claude.ai/oauth/authorize"
-	anthropicTokenURL = "https://console.anthropic.com/v1/oauth/token"
-	anthropicClientID = "9d1c250a-e61b-44d9-88ed-5944d1962f5e"
-	redirectURI       = "http://localhost:54545/callback"
+	AuthURL     = "https://claude.ai/oauth/authorize"
+	TokenURL    = "https://console.anthropic.com/v1/oauth/token"
+	ClientID    = "9d1c250a-e61b-44d9-88ed-5944d1962f5e"
+	RedirectURI = "http://localhost:54545/callback"
 )
 
 // tokenResponse represents the response structure from Anthropic's OAuth token endpoint.
@@ -50,7 +50,8 @@ type ClaudeAuth struct {
 }
 
 // NewClaudeAuth creates a new Anthropic authentication service.
-// It initializes the HTTP client with proxy settings from the configuration.
+// It initializes the HTTP client with a custom TLS transport that uses Firefox
+// fingerprint to bypass Cloudflare's TLS fingerprinting on Anthropic domains.
 //
 // Parameters:
 //   - cfg: The application configuration containing proxy settings
@@ -58,8 +59,10 @@ type ClaudeAuth struct {
 // Returns:
 //   - *ClaudeAuth: A new Claude authentication service instance
 func NewClaudeAuth(cfg *config.Config) *ClaudeAuth {
+	// Use custom HTTP client with Firefox TLS fingerprint to bypass
+	// Cloudflare's bot detection on Anthropic domains
 	return &ClaudeAuth{
-		httpClient: util.SetProxy(&cfg.SDKConfig, &http.Client{}),
+		httpClient: NewAnthropicHttpClient(&cfg.SDKConfig),
 	}
 }
 
@@ -82,16 +85,16 @@ func (o *ClaudeAuth) GenerateAuthURL(state string, pkceCodes *PKCECodes) (string
 
 	params := url.Values{
 		"code":                  {"true"},
-		"client_id":             {anthropicClientID},
+		"client_id":             {ClientID},
 		"response_type":         {"code"},
-		"redirect_uri":          {redirectURI},
+		"redirect_uri":          {RedirectURI},
 		"scope":                 {"org:create_api_key user:profile user:inference"},
 		"code_challenge":        {pkceCodes.CodeChallenge},
 		"code_challenge_method": {"S256"},
 		"state":                 {state},
 	}
 
-	authURL := fmt.Sprintf("%s?%s", anthropicAuthURL, params.Encode())
+	authURL := fmt.Sprintf("%s?%s", AuthURL, params.Encode())
 	return authURL, state, nil
 }
 
@@ -137,8 +140,8 @@ func (o *ClaudeAuth) ExchangeCodeForTokens(ctx context.Context, code, state stri
 		"code":          newCode,
 		"state":         state,
 		"grant_type":    "authorization_code",
-		"client_id":     anthropicClientID,
-		"redirect_uri":  redirectURI,
+		"client_id":     ClientID,
+		"redirect_uri":  RedirectURI,
 		"code_verifier": pkceCodes.CodeVerifier,
 	}
 
@@ -154,7 +157,7 @@ func (o *ClaudeAuth) ExchangeCodeForTokens(ctx context.Context, code, state stri
 
 	// log.Debugf("Token exchange request: %s", string(jsonBody))
 
-	req, err := http.NewRequestWithContext(ctx, "POST", anthropicTokenURL, strings.NewReader(string(jsonBody)))
+	req, err := http.NewRequestWithContext(ctx, "POST", TokenURL, strings.NewReader(string(jsonBody)))
 	if err != nil {
 		return nil, fmt.Errorf("failed to create token request: %w", err)
 	}
@@ -221,7 +224,7 @@ func (o *ClaudeAuth) RefreshTokens(ctx context.Context, refreshToken string) (*C
 	}
 
 	reqBody := map[string]interface{}{
-		"client_id":     anthropicClientID,
+		"client_id":     ClientID,
 		"grant_type":    "refresh_token",
 		"refresh_token": refreshToken,
 	}
@@ -231,7 +234,7 @@ func (o *ClaudeAuth) RefreshTokens(ctx context.Context, refreshToken string) (*C
 		return nil, fmt.Errorf("failed to marshal request body: %w", err)
 	}
 
-	req, err := http.NewRequestWithContext(ctx, "POST", anthropicTokenURL, strings.NewReader(string(jsonBody)))
+	req, err := http.NewRequestWithContext(ctx, "POST", TokenURL, strings.NewReader(string(jsonBody)))
 	if err != nil {
 		return nil, fmt.Errorf("failed to create refresh request: %w", err)
 	}

internal/auth/claude/utls_transport.go

@@ -0,0 +1,165 @@
+// Package claude provides authentication functionality for Anthropic's Claude API.
+// This file implements a custom HTTP transport using utls to bypass TLS fingerprinting.
+package claude
+
+import (
+	"net/http"
+	"net/url"
+	"strings"
+	"sync"
+
+	tls "github.com/refraction-networking/utls"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/net/http2"
+	"golang.org/x/net/proxy"
+)
+
+// utlsRoundTripper implements http.RoundTripper using utls with Firefox fingerprint
+// to bypass Cloudflare's TLS fingerprinting on Anthropic domains.
+type utlsRoundTripper struct {
+	// mu protects the connections map and pending map
+	mu sync.Mutex
+	// connections caches HTTP/2 client connections per host
+	connections map[string]*http2.ClientConn
+	// pending tracks hosts that are currently being connected to (prevents race condition)
+	pending map[string]*sync.Cond
+	// dialer is used to create network connections, supporting proxies
+	dialer proxy.Dialer
+}
+
+// newUtlsRoundTripper creates a new utls-based round tripper with optional proxy support
+func newUtlsRoundTripper(cfg *config.SDKConfig) *utlsRoundTripper {
+	var dialer proxy.Dialer = proxy.Direct
+	if cfg != nil && cfg.ProxyURL != "" {
+		proxyURL, err := url.Parse(cfg.ProxyURL)
+		if err != nil {
+			log.Errorf("failed to parse proxy URL %q: %v", cfg.ProxyURL, err)
+		} else {
+			pDialer, err := proxy.FromURL(proxyURL, proxy.Direct)
+			if err != nil {
+				log.Errorf("failed to create proxy dialer for %q: %v", cfg.ProxyURL, err)
+			} else {
+				dialer = pDialer
+			}
+		}
+	}
+
+	return &utlsRoundTripper{
+		connections: make(map[string]*http2.ClientConn),
+		pending:     make(map[string]*sync.Cond),
+		dialer:      dialer,
+	}
+}
+
+// getOrCreateConnection gets an existing connection or creates a new one.
+// It uses a per-host locking mechanism to prevent multiple goroutines from
+// creating connections to the same host simultaneously.
+func (t *utlsRoundTripper) getOrCreateConnection(host, addr string) (*http2.ClientConn, error) {
+	t.mu.Lock()
+
+	// Check if connection exists and is usable
+	if h2Conn, ok := t.connections[host]; ok && h2Conn.CanTakeNewRequest() {
+		t.mu.Unlock()
+		return h2Conn, nil
+	}
+
+	// Check if another goroutine is already creating a connection
+	if cond, ok := t.pending[host]; ok {
+		// Wait for the other goroutine to finish
+		cond.Wait()
+		// Check if connection is now available
+		if h2Conn, ok := t.connections[host]; ok && h2Conn.CanTakeNewRequest() {
+			t.mu.Unlock()
+			return h2Conn, nil
+		}
+		// Connection still not available, we'll create one
+	}
+
+	// Mark this host as pending
+	cond := sync.NewCond(&t.mu)
+	t.pending[host] = cond
+	t.mu.Unlock()
+
+	// Create connection outside the lock
+	h2Conn, err := t.createConnection(host, addr)
+
+	t.mu.Lock()
+	defer t.mu.Unlock()
+
+	// Remove pending marker and wake up waiting goroutines
+	delete(t.pending, host)
+	cond.Broadcast()
+
+	if err != nil {
+		return nil, err
+	}
+
+	// Store the new connection
+	t.connections[host] = h2Conn
+	return h2Conn, nil
+}
+
+// createConnection creates a new HTTP/2 connection with Firefox TLS fingerprint
+func (t *utlsRoundTripper) createConnection(host, addr string) (*http2.ClientConn, error) {
+	conn, err := t.dialer.Dial("tcp", addr)
+	if err != nil {
+		return nil, err
+	}
+
+	tlsConfig := &tls.Config{ServerName: host}
+	tlsConn := tls.UClient(conn, tlsConfig, tls.HelloFirefox_Auto)
+
+	if err := tlsConn.Handshake(); err != nil {
+		conn.Close()
+		return nil, err
+	}
+
+	tr := &http2.Transport{}
+	h2Conn, err := tr.NewClientConn(tlsConn)
+	if err != nil {
+		tlsConn.Close()
+		return nil, err
+	}
+
+	return h2Conn, nil
+}
+
+// RoundTrip implements http.RoundTripper
+func (t *utlsRoundTripper) RoundTrip(req *http.Request) (*http.Response, error) {
+	host := req.URL.Host
+	addr := host
+	if !strings.Contains(addr, ":") {
+		addr += ":443"
+	}
+
+	// Get hostname without port for TLS ServerName
+	hostname := req.URL.Hostname()
+
+	h2Conn, err := t.getOrCreateConnection(hostname, addr)
+	if err != nil {
+		return nil, err
+	}
+
+	resp, err := h2Conn.RoundTrip(req)
+	if err != nil {
+		// Connection failed, remove it from cache
+		t.mu.Lock()
+		if cached, ok := t.connections[hostname]; ok && cached == h2Conn {
+			delete(t.connections, hostname)
+		}
+		t.mu.Unlock()
+		return nil, err
+	}
+
+	return resp, nil
+}
+
+// NewAnthropicHttpClient creates an HTTP client that bypasses TLS fingerprinting
+// for Anthropic domains by using utls with Firefox fingerprint.
+// It accepts optional SDK configuration for proxy settings.
+func NewAnthropicHttpClient(cfg *config.SDKConfig) *http.Client {
+	return &http.Client{
+		Transport: newUtlsRoundTripper(cfg),
+	}
+}

internal/auth/codex/filename.go

@@ -0,0 +1,46 @@
+package codex
+
+import (
+	"fmt"
+	"strings"
+	"unicode"
+)
+
+// CredentialFileName returns the filename used to persist Codex OAuth credentials.
+// When planType is available (e.g. "plus", "team"), it is appended after the email
+// as a suffix to disambiguate subscriptions.
+func CredentialFileName(email, planType, hashAccountID string, includeProviderPrefix bool) string {
+	email = strings.TrimSpace(email)
+	plan := normalizePlanTypeForFilename(planType)
+
+	prefix := ""
+	if includeProviderPrefix {
+		prefix = "codex"
+	}
+
+	if plan == "" {
+		return fmt.Sprintf("%s-%s.json", prefix, email)
+	} else if plan == "team" {
+		return fmt.Sprintf("%s-%s-%s-%s.json", prefix, hashAccountID, email, plan)
+	}
+	return fmt.Sprintf("%s-%s-%s.json", prefix, email, plan)
+}
+
+func normalizePlanTypeForFilename(planType string) string {
+	planType = strings.TrimSpace(planType)
+	if planType == "" {
+		return ""
+	}
+
+	parts := strings.FieldsFunc(planType, func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	})
+	if len(parts) == 0 {
+		return ""
+	}
+
+	for i, part := range parts {
+		parts[i] = strings.ToLower(strings.TrimSpace(part))
+	}
+	return strings.Join(parts, "-")
+}

internal/auth/codex/openai_auth.go

@@ -19,11 +19,12 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+// OAuth configuration constants for OpenAI Codex
 const (
-	openaiAuthURL  = "https://auth.openai.com/oauth/authorize"
-	openaiTokenURL = "https://auth.openai.com/oauth/token"
-	openaiClientID = "app_EMoamEEZ73f0CkXaXp7hrann"
-	redirectURI    = "http://localhost:1455/auth/callback"
+	AuthURL     = "https://auth.openai.com/oauth/authorize"
+	TokenURL    = "https://auth.openai.com/oauth/token"
+	ClientID    = "app_EMoamEEZ73f0CkXaXp7hrann"
+	RedirectURI = "http://localhost:1455/auth/callback"
 )
 
 // CodexAuth handles the OpenAI OAuth2 authentication flow.
@@ -50,9 +51,9 @@ func (o *CodexAuth) GenerateAuthURL(state string, pkceCodes *PKCECodes) (string,
 	}
 
 	params := url.Values{
-		"client_id":                  {openaiClientID},
+		"client_id":                  {ClientID},
 		"response_type":              {"code"},
-		"redirect_uri":               {redirectURI},
+		"redirect_uri":               {RedirectURI},
 		"scope":                      {"openid email profile offline_access"},
 		"state":                      {state},
 		"code_challenge":             {pkceCodes.CodeChallenge},
@@ -62,7 +63,7 @@ func (o *CodexAuth) GenerateAuthURL(state string, pkceCodes *PKCECodes) (string,
 		"codex_cli_simplified_flow":  {"true"},
 	}
 
-	authURL := fmt.Sprintf("%s?%s", openaiAuthURL, params.Encode())
+	authURL := fmt.Sprintf("%s?%s", AuthURL, params.Encode())
 	return authURL, nil
 }
 
@@ -77,13 +78,13 @@ func (o *CodexAuth) ExchangeCodeForTokens(ctx context.Context, code string, pkce
 	// Prepare token exchange request
 	data := url.Values{
 		"grant_type":    {"authorization_code"},
-		"client_id":     {openaiClientID},
+		"client_id":     {ClientID},
 		"code":          {code},
-		"redirect_uri":  {redirectURI},
+		"redirect_uri":  {RedirectURI},
 		"code_verifier": {pkceCodes.CodeVerifier},
 	}
 
-	req, err := http.NewRequestWithContext(ctx, "POST", openaiTokenURL, strings.NewReader(data.Encode()))
+	req, err := http.NewRequestWithContext(ctx, "POST", TokenURL, strings.NewReader(data.Encode()))
 	if err != nil {
 		return nil, fmt.Errorf("failed to create token request: %w", err)
 	}
@@ -163,13 +164,13 @@ func (o *CodexAuth) RefreshTokens(ctx context.Context, refreshToken string) (*Co
 	}
 
 	data := url.Values{
-		"client_id":     {openaiClientID},
+		"client_id":     {ClientID},
 		"grant_type":    {"refresh_token"},
 		"refresh_token": {refreshToken},
 		"scope":         {"openid profile email"},
 	}
 
-	req, err := http.NewRequestWithContext(ctx, "POST", openaiTokenURL, strings.NewReader(data.Encode()))
+	req, err := http.NewRequestWithContext(ctx, "POST", TokenURL, strings.NewReader(data.Encode()))
 	if err != nil {
 		return nil, fmt.Errorf("failed to create refresh request: %w", err)
 	}

internal/auth/gemini/gemini_auth.go

@@ -18,6 +18,7 @@ import (
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/codex"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/browser"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/misc"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	log "github.com/sirupsen/logrus"
 	"github.com/tidwall/gjson"
@@ -27,18 +28,19 @@ import (
 	"golang.org/x/oauth2/google"
 )
 
+// OAuth configuration constants for Gemini
 const (
-	geminiOauthClientID     = "681255809395-oo8ft2oprdrnp9e3aqf6av3hmdib135j.apps.googleusercontent.com"
-	geminiOauthClientSecret = "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"
+	ClientID            = "681255809395-oo8ft2oprdrnp9e3aqf6av3hmdib135j.apps.googleusercontent.com"
+	ClientSecret        = "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"
+	DefaultCallbackPort = 8085
 )
 
-var (
-	geminiOauthScopes = []string{
+// OAuth scopes for Gemini authentication
+var Scopes = []string{
 	"https://www.googleapis.com/auth/cloud-platform",
 	"https://www.googleapis.com/auth/userinfo.email",
 	"https://www.googleapis.com/auth/userinfo.profile",
 }
-)
 
 // GeminiAuth provides methods for handling the Gemini OAuth2 authentication flow.
 // It encapsulates the logic for obtaining, storing, and refreshing authentication tokens
@@ -46,6 +48,13 @@ var (
 type GeminiAuth struct {
 }
 
+// WebLoginOptions customizes the interactive OAuth flow.
+type WebLoginOptions struct {
+	NoBrowser    bool
+	CallbackPort int
+	Prompt       func(string) (string, error)
+}
+
 // NewGeminiAuth creates a new instance of GeminiAuth.
 func NewGeminiAuth() *GeminiAuth {
 	return &GeminiAuth{}
@@ -59,12 +68,18 @@ func NewGeminiAuth() *GeminiAuth {
 //   - ctx: The context for the HTTP client
 //   - ts: The Gemini token storage containing authentication tokens
 //   - cfg: The configuration containing proxy settings
-//   - noBrowser: Optional parameter to disable browser opening
+//   - opts: Optional parameters to customize browser and prompt behavior
 //
 // Returns:
 //   - *http.Client: An HTTP client configured with authentication
 //   - error: An error if the client configuration fails, nil otherwise
-func (g *GeminiAuth) GetAuthenticatedClient(ctx context.Context, ts *GeminiTokenStorage, cfg *config.Config, noBrowser ...bool) (*http.Client, error) {
+func (g *GeminiAuth) GetAuthenticatedClient(ctx context.Context, ts *GeminiTokenStorage, cfg *config.Config, opts *WebLoginOptions) (*http.Client, error) {
+	callbackPort := DefaultCallbackPort
+	if opts != nil && opts.CallbackPort > 0 {
+		callbackPort = opts.CallbackPort
+	}
+	callbackURL := fmt.Sprintf("http://localhost:%d/oauth2callback", callbackPort)
+
 	// Configure proxy settings for the HTTP client if a proxy URL is provided.
 	proxyURL, err := url.Parse(cfg.ProxyURL)
 	if err == nil {
@@ -97,10 +112,10 @@ func (g *GeminiAuth) GetAuthenticatedClient(ctx context.Context, ts *GeminiToken
 
 	// Configure the OAuth2 client.
 	conf := &oauth2.Config{
-		ClientID:     geminiOauthClientID,
-		ClientSecret: geminiOauthClientSecret,
-		RedirectURL:  "http://localhost:8085/oauth2callback", // This will be used by the local server.
-		Scopes:       geminiOauthScopes,
+		ClientID:     ClientID,
+		ClientSecret: ClientSecret,
+		RedirectURL:  callbackURL, // This will be used by the local server.
+		Scopes:       Scopes,
 		Endpoint:     google.Endpoint,
 	}
 
@@ -109,7 +124,7 @@ func (g *GeminiAuth) GetAuthenticatedClient(ctx context.Context, ts *GeminiToken
 	// If no token is found in storage, initiate the web-based OAuth flow.
 	if ts.Token == nil {
 		fmt.Printf("Could not load token from file, starting OAuth flow.\n")
-		token, err = g.getTokenFromWeb(ctx, conf, noBrowser...)
+		token, err = g.getTokenFromWeb(ctx, conf, opts)
 		if err != nil {
 			return nil, fmt.Errorf("failed to get token from web: %w", err)
 		}
@@ -183,9 +198,9 @@ func (g *GeminiAuth) createTokenStorage(ctx context.Context, config *oauth2.Conf
 	}
 
 	ifToken["token_uri"] = "https://oauth2.googleapis.com/token"
-	ifToken["client_id"] = geminiOauthClientID
-	ifToken["client_secret"] = geminiOauthClientSecret
-	ifToken["scopes"] = geminiOauthScopes
+	ifToken["client_id"] = ClientID
+	ifToken["client_secret"] = ClientSecret
+	ifToken["scopes"] = Scopes
 	ifToken["universe_domain"] = "googleapis.com"
 
 	ts := GeminiTokenStorage{
@@ -205,35 +220,50 @@ func (g *GeminiAuth) createTokenStorage(ctx context.Context, config *oauth2.Conf
 // Parameters:
 //   - ctx: The context for the HTTP client
 //   - config: The OAuth2 configuration
-//   - noBrowser: Optional parameter to disable browser opening
+//   - opts: Optional parameters to customize browser and prompt behavior
 //
 // Returns:
 //   - *oauth2.Token: The OAuth2 token obtained from the authorization flow
 //   - error: An error if the token acquisition fails, nil otherwise
-func (g *GeminiAuth) getTokenFromWeb(ctx context.Context, config *oauth2.Config, noBrowser ...bool) (*oauth2.Token, error) {
+func (g *GeminiAuth) getTokenFromWeb(ctx context.Context, config *oauth2.Config, opts *WebLoginOptions) (*oauth2.Token, error) {
+	callbackPort := DefaultCallbackPort
+	if opts != nil && opts.CallbackPort > 0 {
+		callbackPort = opts.CallbackPort
+	}
+	callbackURL := fmt.Sprintf("http://localhost:%d/oauth2callback", callbackPort)
+
 	// Use a channel to pass the authorization code from the HTTP handler to the main function.
-	codeChan := make(chan string)
-	errChan := make(chan error)
+	codeChan := make(chan string, 1)
+	errChan := make(chan error, 1)
 
 	// Create a new HTTP server with its own multiplexer.
 	mux := http.NewServeMux()
-	server := &http.Server{Addr: ":8085", Handler: mux}
-	config.RedirectURL = "http://localhost:8085/oauth2callback"
+	server := &http.Server{Addr: fmt.Sprintf(":%d", callbackPort), Handler: mux}
+	config.RedirectURL = callbackURL
 
 	mux.HandleFunc("/oauth2callback", func(w http.ResponseWriter, r *http.Request) {
 		if err := r.URL.Query().Get("error"); err != "" {
 			_, _ = fmt.Fprintf(w, "Authentication failed: %s", err)
-			errChan <- fmt.Errorf("authentication failed via callback: %s", err)
+			select {
+			case errChan <- fmt.Errorf("authentication failed via callback: %s", err):
+			default:
+			}
 			return
 		}
 		code := r.URL.Query().Get("code")
 		if code == "" {
 			_, _ = fmt.Fprint(w, "Authentication failed: code not found.")
-			errChan <- fmt.Errorf("code not found in callback")
+			select {
+			case errChan <- fmt.Errorf("code not found in callback"):
+			default:
+			}
 			return
 		}
 		_, _ = fmt.Fprint(w, "<html><body><h1>Authentication successful!</h1><p>You can close this window.</p></body></html>")
-		codeChan <- code
+		select {
+		case codeChan <- code:
+		default:
+		}
 	})
 
 	// Start the server in a goroutine.
@@ -250,19 +280,24 @@ func (g *GeminiAuth) getTokenFromWeb(ctx context.Context, config *oauth2.Config,
 	// Open the authorization URL in the user's browser.
 	authURL := config.AuthCodeURL("state-token", oauth2.AccessTypeOffline, oauth2.SetAuthURLParam("prompt", "consent"))
 
-	if len(noBrowser) == 1 && !noBrowser[0] {
+	noBrowser := false
+	if opts != nil {
+		noBrowser = opts.NoBrowser
+	}
+
+	if !noBrowser {
 		fmt.Println("Opening browser for authentication...")
 
 		// Check if browser is available
 		if !browser.IsAvailable() {
 			log.Warn("No browser available on this system")
-			util.PrintSSHTunnelInstructions(8085)
+			util.PrintSSHTunnelInstructions(callbackPort)
 			fmt.Printf("Please manually open this URL in your browser:\n\n%s\n", authURL)
 		} else {
 			if err := browser.OpenURL(authURL); err != nil {
 				authErr := codex.NewAuthenticationError(codex.ErrBrowserOpenFailed, err)
 				log.Warn(codex.GetUserFriendlyMessage(authErr))
-				util.PrintSSHTunnelInstructions(8085)
+				util.PrintSSHTunnelInstructions(callbackPort)
 				fmt.Printf("Please manually open this URL in your browser:\n\n%s\n", authURL)
 
 				// Log platform info for debugging
@@ -273,7 +308,7 @@ func (g *GeminiAuth) getTokenFromWeb(ctx context.Context, config *oauth2.Config,
 			}
 		}
 	} else {
-		util.PrintSSHTunnelInstructions(8085)
+		util.PrintSSHTunnelInstructions(callbackPort)
 		fmt.Printf("Please open this URL in your browser:\n\n%s\n", authURL)
 	}
 
@@ -281,14 +316,61 @@ func (g *GeminiAuth) getTokenFromWeb(ctx context.Context, config *oauth2.Config,
 
 	// Wait for the authorization code or an error.
 	var authCode string
+	timeoutTimer := time.NewTimer(5 * time.Minute)
+	defer timeoutTimer.Stop()
+
+	var manualPromptTimer *time.Timer
+	var manualPromptC <-chan time.Time
+	if opts != nil && opts.Prompt != nil {
+		manualPromptTimer = time.NewTimer(15 * time.Second)
+		manualPromptC = manualPromptTimer.C
+		defer manualPromptTimer.Stop()
+	}
+
+waitForCallback:
+	for {
 		select {
 		case code := <-codeChan:
 			authCode = code
+			break waitForCallback
 		case err := <-errChan:
 			return nil, err
-	case <-time.After(5 * time.Minute): // Timeout
+		case <-manualPromptC:
+			manualPromptC = nil
+			if manualPromptTimer != nil {
+				manualPromptTimer.Stop()
+			}
+			select {
+			case code := <-codeChan:
+				authCode = code
+				break waitForCallback
+			case err := <-errChan:
+				return nil, err
+			default:
+			}
+			input, err := opts.Prompt("Paste the Gemini callback URL (or press Enter to keep waiting): ")
+			if err != nil {
+				return nil, err
+			}
+			parsed, err := misc.ParseOAuthCallback(input)
+			if err != nil {
+				return nil, err
+			}
+			if parsed == nil {
+				continue
+			}
+			if parsed.Error != "" {
+				return nil, fmt.Errorf("authentication failed via callback: %s", parsed.Error)
+			}
+			if parsed.Code == "" {
+				return nil, fmt.Errorf("code not found in callback")
+			}
+			authCode = parsed.Code
+			break waitForCallback
+		case <-timeoutTimer.C:
 			return nil, fmt.Errorf("oauth flow timed out")
 		}
+	}
 
 	// Shutdown the server.
 	if err := server.Shutdown(ctx); err != nil {

internal/auth/iflow/cookie_helpers.go

@@ -1,7 +1,10 @@
 package iflow
 
 import (
+	"encoding/json"
 	"fmt"
+	"os"
+	"path/filepath"
 	"strings"
 )
 
@@ -36,3 +39,61 @@ func SanitizeIFlowFileName(raw string) string {
 	}
 	return strings.TrimSpace(result.String())
 }
+
+// ExtractBXAuth extracts the BXAuth value from a cookie string.
+func ExtractBXAuth(cookie string) string {
+	parts := strings.Split(cookie, ";")
+	for _, part := range parts {
+		part = strings.TrimSpace(part)
+		if strings.HasPrefix(part, "BXAuth=") {
+			return strings.TrimPrefix(part, "BXAuth=")
+		}
+	}
+	return ""
+}
+
+// CheckDuplicateBXAuth checks if the given BXAuth value already exists in any iflow auth file.
+// Returns the path of the existing file if found, empty string otherwise.
+func CheckDuplicateBXAuth(authDir, bxAuth string) (string, error) {
+	if bxAuth == "" {
+		return "", nil
+	}
+
+	entries, err := os.ReadDir(authDir)
+	if err != nil {
+		if os.IsNotExist(err) {
+			return "", nil
+		}
+		return "", fmt.Errorf("read auth dir failed: %w", err)
+	}
+
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		name := entry.Name()
+		if !strings.HasPrefix(name, "iflow-") || !strings.HasSuffix(name, ".json") {
+			continue
+		}
+
+		filePath := filepath.Join(authDir, name)
+		data, err := os.ReadFile(filePath)
+		if err != nil {
+			continue
+		}
+
+		var tokenData struct {
+			Cookie string `json:"cookie"`
+		}
+		if err := json.Unmarshal(data, &tokenData); err != nil {
+			continue
+		}
+
+		existingBXAuth := ExtractBXAuth(tokenData.Cookie)
+		if existingBXAuth != "" && existingBXAuth == bxAuth {
+			return filePath, nil
+		}
+	}
+
+	return "", nil
+}

internal/auth/iflow/iflow_auth.go

@@ -494,11 +494,18 @@ func (ia *IFlowAuth) CreateCookieTokenStorage(data *IFlowTokenData) *IFlowTokenS
 		return nil
 	}
 
+	// Only save the BXAuth field from the cookie
+	bxAuth := ExtractBXAuth(data.Cookie)
+	cookieToSave := ""
+	if bxAuth != "" {
+		cookieToSave = "BXAuth=" + bxAuth + ";"
+	}
+
 	return &IFlowTokenStorage{
 		APIKey:      data.APIKey,
 		Email:       data.Email,
 		Expire:      data.Expire,
-		Cookie:      data.Cookie,
+		Cookie:      cookieToSave,
 		LastRefresh: time.Now().Format(time.RFC3339),
 		Type:        "iflow",
 	}

internal/cache/signature_cache.go

@@ -0,0 +1,195 @@
+package cache
+
+import (
+	"crypto/sha256"
+	"encoding/hex"
+	"strings"
+	"sync"
+	"time"
+)
+
+// SignatureEntry holds a cached thinking signature with timestamp
+type SignatureEntry struct {
+	Signature string
+	Timestamp time.Time
+}
+
+const (
+	// SignatureCacheTTL is how long signatures are valid
+	SignatureCacheTTL = 3 * time.Hour
+
+	// SignatureTextHashLen is the length of the hash key (16 hex chars = 64-bit key space)
+	SignatureTextHashLen = 16
+
+	// MinValidSignatureLen is the minimum length for a signature to be considered valid
+	MinValidSignatureLen = 50
+
+	// CacheCleanupInterval controls how often stale entries are purged
+	CacheCleanupInterval = 10 * time.Minute
+)
+
+// signatureCache stores signatures by model group -> textHash -> SignatureEntry
+var signatureCache sync.Map
+
+// cacheCleanupOnce ensures the background cleanup goroutine starts only once
+var cacheCleanupOnce sync.Once
+
+// groupCache is the inner map type
+type groupCache struct {
+	mu      sync.RWMutex
+	entries map[string]SignatureEntry
+}
+
+// hashText creates a stable, Unicode-safe key from text content
+func hashText(text string) string {
+	h := sha256.Sum256([]byte(text))
+	return hex.EncodeToString(h[:])[:SignatureTextHashLen]
+}
+
+// getOrCreateGroupCache gets or creates a cache bucket for a model group
+func getOrCreateGroupCache(groupKey string) *groupCache {
+	// Start background cleanup on first access
+	cacheCleanupOnce.Do(startCacheCleanup)
+
+	if val, ok := signatureCache.Load(groupKey); ok {
+		return val.(*groupCache)
+	}
+	sc := &groupCache{entries: make(map[string]SignatureEntry)}
+	actual, _ := signatureCache.LoadOrStore(groupKey, sc)
+	return actual.(*groupCache)
+}
+
+// startCacheCleanup launches a background goroutine that periodically
+// removes caches where all entries have expired.
+func startCacheCleanup() {
+	go func() {
+		ticker := time.NewTicker(CacheCleanupInterval)
+		defer ticker.Stop()
+		for range ticker.C {
+			purgeExpiredCaches()
+		}
+	}()
+}
+
+// purgeExpiredCaches removes caches with no valid (non-expired) entries.
+func purgeExpiredCaches() {
+	now := time.Now()
+	signatureCache.Range(func(key, value any) bool {
+		sc := value.(*groupCache)
+		sc.mu.Lock()
+		// Remove expired entries
+		for k, entry := range sc.entries {
+			if now.Sub(entry.Timestamp) > SignatureCacheTTL {
+				delete(sc.entries, k)
+			}
+		}
+		isEmpty := len(sc.entries) == 0
+		sc.mu.Unlock()
+		// Remove cache bucket if empty
+		if isEmpty {
+			signatureCache.Delete(key)
+		}
+		return true
+	})
+}
+
+// CacheSignature stores a thinking signature for a given model group and text.
+// Used for Claude models that require signed thinking blocks in multi-turn conversations.
+func CacheSignature(modelName, text, signature string) {
+	if text == "" || signature == "" {
+		return
+	}
+	if len(signature) < MinValidSignatureLen {
+		return
+	}
+
+	groupKey := GetModelGroup(modelName)
+	textHash := hashText(text)
+	sc := getOrCreateGroupCache(groupKey)
+	sc.mu.Lock()
+	defer sc.mu.Unlock()
+
+	sc.entries[textHash] = SignatureEntry{
+		Signature: signature,
+		Timestamp: time.Now(),
+	}
+}
+
+// GetCachedSignature retrieves a cached signature for a given model group and text.
+// Returns empty string if not found or expired.
+func GetCachedSignature(modelName, text string) string {
+	groupKey := GetModelGroup(modelName)
+
+	if text == "" {
+		if groupKey == "gemini" {
+			return "skip_thought_signature_validator"
+		}
+		return ""
+	}
+	val, ok := signatureCache.Load(groupKey)
+	if !ok {
+		if groupKey == "gemini" {
+			return "skip_thought_signature_validator"
+		}
+		return ""
+	}
+	sc := val.(*groupCache)
+
+	textHash := hashText(text)
+
+	now := time.Now()
+
+	sc.mu.Lock()
+	entry, exists := sc.entries[textHash]
+	if !exists {
+		sc.mu.Unlock()
+		if groupKey == "gemini" {
+			return "skip_thought_signature_validator"
+		}
+		return ""
+	}
+	if now.Sub(entry.Timestamp) > SignatureCacheTTL {
+		delete(sc.entries, textHash)
+		sc.mu.Unlock()
+		if groupKey == "gemini" {
+			return "skip_thought_signature_validator"
+		}
+		return ""
+	}
+
+	// Refresh TTL on access (sliding expiration).
+	entry.Timestamp = now
+	sc.entries[textHash] = entry
+	sc.mu.Unlock()
+
+	return entry.Signature
+}
+
+// ClearSignatureCache clears signature cache for a specific model group or all groups.
+func ClearSignatureCache(modelName string) {
+	if modelName == "" {
+		signatureCache.Range(func(key, _ any) bool {
+			signatureCache.Delete(key)
+			return true
+		})
+		return
+	}
+	groupKey := GetModelGroup(modelName)
+	signatureCache.Delete(groupKey)
+}
+
+// HasValidSignature checks if a signature is valid (non-empty and long enough)
+func HasValidSignature(modelName, signature string) bool {
+	return (signature != "" && len(signature) >= MinValidSignatureLen) || (signature == "skip_thought_signature_validator" && GetModelGroup(modelName) == "gemini")
+}
+
+func GetModelGroup(modelName string) string {
+	if strings.Contains(modelName, "gpt") {
+		return "gpt"
+	} else if strings.Contains(modelName, "claude") {
+		return "claude"
+	} else if strings.Contains(modelName, "gemini") {
+		return "gemini"
+	}
+	return modelName
+}

internal/cache/signature_cache_test.go

@@ -0,0 +1,210 @@
+package cache
+
+import (
+	"testing"
+	"time"
+)
+
+const testModelName = "claude-sonnet-4-5"
+
+func TestCacheSignature_BasicStorageAndRetrieval(t *testing.T) {
+	ClearSignatureCache("")
+
+	text := "This is some thinking text content"
+	signature := "abc123validSignature1234567890123456789012345678901234567890"
+
+	// Store signature
+	CacheSignature(testModelName, text, signature)
+
+	// Retrieve signature
+	retrieved := GetCachedSignature(testModelName, text)
+	if retrieved != signature {
+		t.Errorf("Expected signature '%s', got '%s'", signature, retrieved)
+	}
+}
+
+func TestCacheSignature_DifferentModelGroups(t *testing.T) {
+	ClearSignatureCache("")
+
+	text := "Same text across models"
+	sig1 := "signature1_1234567890123456789012345678901234567890123456"
+	sig2 := "signature2_1234567890123456789012345678901234567890123456"
+
+	geminiModel := "gemini-3-pro-preview"
+	CacheSignature(testModelName, text, sig1)
+	CacheSignature(geminiModel, text, sig2)
+
+	if GetCachedSignature(testModelName, text) != sig1 {
+		t.Error("Claude signature mismatch")
+	}
+	if GetCachedSignature(geminiModel, text) != sig2 {
+		t.Error("Gemini signature mismatch")
+	}
+}
+
+func TestCacheSignature_NotFound(t *testing.T) {
+	ClearSignatureCache("")
+
+	// Non-existent session
+	if got := GetCachedSignature(testModelName, "some text"); got != "" {
+		t.Errorf("Expected empty string for nonexistent session, got '%s'", got)
+	}
+
+	// Existing session but different text
+	CacheSignature(testModelName, "text-a", "sigA12345678901234567890123456789012345678901234567890")
+	if got := GetCachedSignature(testModelName, "text-b"); got != "" {
+		t.Errorf("Expected empty string for different text, got '%s'", got)
+	}
+}
+
+func TestCacheSignature_EmptyInputs(t *testing.T) {
+	ClearSignatureCache("")
+
+	// All empty/invalid inputs should be no-ops
+	CacheSignature(testModelName, "", "sig12345678901234567890123456789012345678901234567890")
+	CacheSignature(testModelName, "text", "")
+	CacheSignature(testModelName, "text", "short") // Too short
+
+	if got := GetCachedSignature(testModelName, "text"); got != "" {
+		t.Errorf("Expected empty after invalid cache attempts, got '%s'", got)
+	}
+}
+
+func TestCacheSignature_ShortSignatureRejected(t *testing.T) {
+	ClearSignatureCache("")
+
+	text := "Some text"
+	shortSig := "abc123" // Less than 50 chars
+
+	CacheSignature(testModelName, text, shortSig)
+
+	if got := GetCachedSignature(testModelName, text); got != "" {
+		t.Errorf("Short signature should be rejected, got '%s'", got)
+	}
+}
+
+func TestClearSignatureCache_ModelGroup(t *testing.T) {
+	ClearSignatureCache("")
+
+	sig := "validSig1234567890123456789012345678901234567890123456"
+	CacheSignature(testModelName, "text", sig)
+	CacheSignature(testModelName, "text-2", sig)
+
+	ClearSignatureCache("session-1")
+
+	if got := GetCachedSignature(testModelName, "text"); got != sig {
+		t.Error("signature should remain when clearing unknown session")
+	}
+}
+
+func TestClearSignatureCache_AllSessions(t *testing.T) {
+	ClearSignatureCache("")
+
+	sig := "validSig1234567890123456789012345678901234567890123456"
+	CacheSignature(testModelName, "text", sig)
+	CacheSignature(testModelName, "text-2", sig)
+
+	ClearSignatureCache("")
+
+	if got := GetCachedSignature(testModelName, "text"); got != "" {
+		t.Error("text should be cleared")
+	}
+	if got := GetCachedSignature(testModelName, "text-2"); got != "" {
+		t.Error("text-2 should be cleared")
+	}
+}
+
+func TestHasValidSignature(t *testing.T) {
+	tests := []struct {
+		name      string
+		modelName string
+		signature string
+		expected  bool
+	}{
+		{"valid long signature", testModelName, "abc123validSignature1234567890123456789012345678901234567890", true},
+		{"exactly 50 chars", testModelName, "12345678901234567890123456789012345678901234567890", true},
+		{"49 chars - invalid", testModelName, "1234567890123456789012345678901234567890123456789", false},
+		{"empty string", testModelName, "", false},
+		{"short signature", testModelName, "abc", false},
+		{"gemini sentinel", "gemini-3-pro-preview", "skip_thought_signature_validator", true},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := HasValidSignature(tt.modelName, tt.signature)
+			if result != tt.expected {
+				t.Errorf("HasValidSignature(%q) = %v, expected %v", tt.signature, result, tt.expected)
+			}
+		})
+	}
+}
+
+func TestCacheSignature_TextHashCollisionResistance(t *testing.T) {
+	ClearSignatureCache("")
+
+	// Different texts should produce different hashes
+	text1 := "First thinking text"
+	text2 := "Second thinking text"
+	sig1 := "signature1_1234567890123456789012345678901234567890123456"
+	sig2 := "signature2_1234567890123456789012345678901234567890123456"
+
+	CacheSignature(testModelName, text1, sig1)
+	CacheSignature(testModelName, text2, sig2)
+
+	if GetCachedSignature(testModelName, text1) != sig1 {
+		t.Error("text1 signature mismatch")
+	}
+	if GetCachedSignature(testModelName, text2) != sig2 {
+		t.Error("text2 signature mismatch")
+	}
+}
+
+func TestCacheSignature_UnicodeText(t *testing.T) {
+	ClearSignatureCache("")
+
+	text := "한글 텍스트와 이모지 🎉 그리고 特殊文字"
+	sig := "unicodeSig123456789012345678901234567890123456789012345"
+
+	CacheSignature(testModelName, text, sig)
+
+	if got := GetCachedSignature(testModelName, text); got != sig {
+		t.Errorf("Unicode text signature retrieval failed, got '%s'", got)
+	}
+}
+
+func TestCacheSignature_Overwrite(t *testing.T) {
+	ClearSignatureCache("")
+
+	text := "Same text"
+	sig1 := "firstSignature12345678901234567890123456789012345678901"
+	sig2 := "secondSignature1234567890123456789012345678901234567890"
+
+	CacheSignature(testModelName, text, sig1)
+	CacheSignature(testModelName, text, sig2) // Overwrite
+
+	if got := GetCachedSignature(testModelName, text); got != sig2 {
+		t.Errorf("Expected overwritten signature '%s', got '%s'", sig2, got)
+	}
+}
+
+// Note: TTL expiration test is tricky to test without mocking time
+// We test the logic path exists but actual expiration would require time manipulation
+func TestCacheSignature_ExpirationLogic(t *testing.T) {
+	ClearSignatureCache("")
+
+	// This test verifies the expiration check exists
+	// In a real scenario, we'd mock time.Now()
+	text := "text"
+	sig := "validSig1234567890123456789012345678901234567890123456"
+
+	CacheSignature(testModelName, text, sig)
+
+	// Fresh entry should be retrievable
+	if got := GetCachedSignature(testModelName, text); got != sig {
+		t.Errorf("Fresh entry should be retrievable, got '%s'", got)
+	}
+
+	// We can't easily test actual expiration without time mocking
+	// but the logic is verified by the implementation
+	_ = time.Now() // Acknowledge we're not testing time passage
+}

internal/cmd/anthropic_login.go

@@ -24,12 +24,18 @@ func DoClaudeLogin(cfg *config.Config, options *LoginOptions) {
 		options = &LoginOptions{}
 	}
 
+	promptFn := options.Prompt
+	if promptFn == nil {
+		promptFn = defaultProjectPrompt()
+	}
+
 	manager := newAuthManager()
 
 	authOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
-		Prompt:    options.Prompt,
+		Prompt:       promptFn,
 	}
 
 	_, savedPath, err := manager.Login(context.Background(), "claude", cfg, authOpts)

internal/cmd/antigravity_login.go

@@ -15,11 +15,17 @@ func DoAntigravityLogin(cfg *config.Config, options *LoginOptions) {
 		options = &LoginOptions{}
 	}
 
+	promptFn := options.Prompt
+	if promptFn == nil {
+		promptFn = defaultProjectPrompt()
+	}
+
 	manager := newAuthManager()
 	authOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
-		Prompt:    options.Prompt,
+		Prompt:       promptFn,
 	}
 
 	record, savedPath, err := manager.Login(context.Background(), "antigravity", cfg, authOpts)

internal/cmd/iflow_cookie.go

@@ -5,7 +5,9 @@ import (
 	"context"
 	"fmt"
 	"os"
+	"path/filepath"
 	"strings"
+	"time"
 
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/auth/iflow"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
@@ -37,6 +39,16 @@ func DoIFlowCookieAuth(cfg *config.Config, options *LoginOptions) {
 		return
 	}
 
+	// Check for duplicate BXAuth before authentication
+	bxAuth := iflow.ExtractBXAuth(cookie)
+	if existingFile, err := iflow.CheckDuplicateBXAuth(cfg.AuthDir, bxAuth); err != nil {
+		fmt.Printf("Failed to check duplicate: %v\n", err)
+		return
+	} else if existingFile != "" {
+		fmt.Printf("Duplicate BXAuth found, authentication already exists: %s\n", filepath.Base(existingFile))
+		return
+	}
+
 	// Authenticate with cookie
 	auth := iflow.NewIFlowAuth(cfg)
 	ctx := context.Background()
@@ -82,5 +94,5 @@ func promptForCookie(promptFn func(string) (string, error)) (string, error) {
 // getAuthFilePath returns the auth file path for the given provider and email
 func getAuthFilePath(cfg *config.Config, provider, email string) string {
 	fileName := iflow.SanitizeIFlowFileName(email)
-	return fmt.Sprintf("%s/%s-%s.json", cfg.AuthDir, provider, fileName)
+	return fmt.Sprintf("%s/%s-%s-%d.json", cfg.AuthDir, provider, fileName, time.Now().Unix())
 }

internal/cmd/iflow_login.go

@@ -20,17 +20,12 @@ func DoIFlowLogin(cfg *config.Config, options *LoginOptions) {
 
 	promptFn := options.Prompt
 	if promptFn == nil {
-		promptFn = func(prompt string) (string, error) {
-			fmt.Println()
-			fmt.Println(prompt)
-			var value string
-			_, err := fmt.Scanln(&value)
-			return value, err
-		}
+		promptFn = defaultProjectPrompt()
 	}
 
 	authOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
 		Prompt:       promptFn,
 	}

internal/cmd/login.go

@@ -55,11 +55,23 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 
 	ctx := context.Background()
 
+	promptFn := options.Prompt
+	if promptFn == nil {
+		promptFn = defaultProjectPrompt()
+	}
+
+	trimmedProjectID := strings.TrimSpace(projectID)
+	callbackPrompt := promptFn
+	if trimmedProjectID == "" {
+		callbackPrompt = nil
+	}
+
 	loginOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
-		ProjectID: strings.TrimSpace(projectID),
+		ProjectID:    trimmedProjectID,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
-		Prompt:    options.Prompt,
+		Prompt:       callbackPrompt,
 	}
 
 	authenticator := sdkAuth.NewGeminiAuthenticator()
@@ -76,7 +88,11 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 	}
 
 	geminiAuth := gemini.NewGeminiAuth()
-	httpClient, errClient := geminiAuth.GetAuthenticatedClient(ctx, storage, cfg, options.NoBrowser)
+	httpClient, errClient := geminiAuth.GetAuthenticatedClient(ctx, storage, cfg, &gemini.WebLoginOptions{
+		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
+		Prompt:       callbackPrompt,
+	})
 	if errClient != nil {
 		log.Errorf("Gemini authentication failed: %v", errClient)
 		return
@@ -90,12 +106,7 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 		return
 	}
 
-	promptFn := options.Prompt
-	if promptFn == nil {
-		promptFn = defaultProjectPrompt()
-	}
-
-	selectedProjectID := promptForProjectSelection(projects, strings.TrimSpace(projectID), promptFn)
+	selectedProjectID := promptForProjectSelection(projects, trimmedProjectID, promptFn)
 	projectSelections, errSelection := resolveProjectSelections(selectedProjectID, projects)
 	if errSelection != nil {
 		log.Errorf("Invalid project selection: %v", errSelection)
@@ -107,6 +118,7 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 	}
 
 	activatedProjects := make([]string, 0, len(projectSelections))
+	seenProjects := make(map[string]bool)
 	for _, candidateID := range projectSelections {
 		log.Infof("Activating project %s", candidateID)
 		if errSetup := performGeminiCLISetup(ctx, httpClient, storage, candidateID); errSetup != nil {
@@ -123,6 +135,13 @@ func DoLogin(cfg *config.Config, projectID string, options *LoginOptions) {
 		if finalID == "" {
 			finalID = candidateID
 		}
+
+		// Skip duplicates
+		if seenProjects[finalID] {
+			log.Infof("Project %s already activated, skipping", finalID)
+			continue
+		}
+		seenProjects[finalID] = true
 		activatedProjects = append(activatedProjects, finalID)
 	}
 
@@ -250,7 +269,39 @@ func performGeminiCLISetup(ctx context.Context, httpClient *http.Client, storage
 			finalProjectID := projectID
 			if responseProjectID != "" {
 				if explicitProject && !strings.EqualFold(responseProjectID, projectID) {
+					// Check if this is a free user (gen-lang-client projects or free/legacy tier)
+					isFreeUser := strings.HasPrefix(projectID, "gen-lang-client-") ||
+						strings.EqualFold(tierID, "FREE") ||
+						strings.EqualFold(tierID, "LEGACY")
+
+					if isFreeUser {
+						// Interactive prompt for free users
+						fmt.Printf("\nGoogle returned a different project ID:\n")
+						fmt.Printf("  Requested (frontend): %s\n", projectID)
+						fmt.Printf("  Returned (backend):   %s\n\n", responseProjectID)
+						fmt.Printf("  Backend project IDs have access to preview models (gemini-3-*).\n")
+						fmt.Printf("  This is normal for free tier users.\n\n")
+						fmt.Printf("Which project ID would you like to use?\n")
+						fmt.Printf("  [1] Backend (recommended): %s\n", responseProjectID)
+						fmt.Printf("  [2] Frontend: %s\n\n", projectID)
+						fmt.Printf("Enter choice [1]: ")
+
+						reader := bufio.NewReader(os.Stdin)
+						choice, _ := reader.ReadString('\n')
+						choice = strings.TrimSpace(choice)
+
+						if choice == "2" {
+							log.Infof("Using frontend project ID: %s", projectID)
+							fmt.Println(". Warning: Frontend project IDs may not have access to preview models.")
+							finalProjectID = projectID
+						} else {
+							log.Infof("Using backend project ID: %s (recommended)", responseProjectID)
+							finalProjectID = responseProjectID
+						}
+					} else {
+						// Pro users: keep requested project ID (original behavior)
 						log.Warnf("Gemini onboarding returned project %s instead of requested %s; keeping requested project ID.", responseProjectID, projectID)
+					}
 				} else {
 					finalProjectID = responseProjectID
 				}

internal/cmd/openai_login.go

@@ -19,6 +19,9 @@ type LoginOptions struct {
 	// NoBrowser indicates whether to skip opening the browser automatically.
 	NoBrowser bool
 
+	// CallbackPort overrides the local OAuth callback port when set (>0).
+	CallbackPort int
+
 	// Prompt allows the caller to provide interactive input when needed.
 	Prompt func(prompt string) (string, error)
 }
@@ -35,12 +38,18 @@ func DoCodexLogin(cfg *config.Config, options *LoginOptions) {
 		options = &LoginOptions{}
 	}
 
+	promptFn := options.Prompt
+	if promptFn == nil {
+		promptFn = defaultProjectPrompt()
+	}
+
 	manager := newAuthManager()
 
 	authOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
-		Prompt:    options.Prompt,
+		Prompt:       promptFn,
 	}
 
 	_, savedPath, err := manager.Login(context.Background(), "codex", cfg, authOpts)

internal/cmd/qwen_login.go

@@ -37,6 +37,7 @@ func DoQwenLogin(cfg *config.Config, options *LoginOptions) {
 
 	authOpts := &sdkAuth.LoginOptions{
 		NoBrowser:    options.NoBrowser,
+		CallbackPort: options.CallbackPort,
 		Metadata:     map[string]string{},
 		Prompt:       promptFn,
 	}

internal/config/config.go

@@ -6,20 +6,23 @@ package config
 
 import (
 	"bytes"
+	"encoding/json"
 	"errors"
 	"fmt"
 	"os"
 	"strings"
 	"syscall"
 
-	"github.com/router-for-me/CLIProxyAPI/v6/sdk/config"
+	log "github.com/sirupsen/logrus"
 	"golang.org/x/crypto/bcrypt"
 	"gopkg.in/yaml.v3"
 )
 
+const DefaultPanelGitHubRepository = "https://github.com/router-for-me/Cli-Proxy-API-Management-Center"
+
 // Config represents the application's configuration, loaded from a YAML file.
 type Config struct {
-	config.SDKConfig `yaml:",inline"`
+	SDKConfig `yaml:",inline"`
 	// Host is the network host/interface on which the API server will bind.
 	// Default is empty ("") to bind all interfaces (IPv4 + IPv6). Use "127.0.0.1" or "localhost" for local-only access.
 	Host string `yaml:"host" json:"-"`
@@ -38,9 +41,20 @@ type Config struct {
 	// Debug enables or disables debug-level logging and other debug features.
 	Debug bool `yaml:"debug" json:"debug"`
 
+	// CommercialMode disables high-overhead HTTP middleware features to minimize per-request memory usage.
+	CommercialMode bool `yaml:"commercial-mode" json:"commercial-mode"`
+
 	// LoggingToFile controls whether application logs are written to rotating files or stdout.
 	LoggingToFile bool `yaml:"logging-to-file" json:"logging-to-file"`
 
+	// LogsMaxTotalSizeMB limits the total size (in MB) of log files under the logs directory.
+	// When exceeded, the oldest log files are deleted until within the limit. Set to 0 to disable.
+	LogsMaxTotalSizeMB int `yaml:"logs-max-total-size-mb" json:"logs-max-total-size-mb"`
+
+	// ErrorLogsMaxFiles limits the number of error log files retained when request logging is disabled.
+	// When exceeded, the oldest error log files are deleted. Default is 10. Set to 0 to disable cleanup.
+	ErrorLogsMaxFiles int `yaml:"error-logs-max-files" json:"error-logs-max-files"`
+
 	// UsageStatisticsEnabled toggles in-memory usage aggregation; when false, usage data is discarded.
 	UsageStatisticsEnabled bool `yaml:"usage-statistics-enabled" json:"usage-statistics-enabled"`
 
@@ -55,9 +69,17 @@ type Config struct {
 	// QuotaExceeded defines the behavior when a quota is exceeded.
 	QuotaExceeded QuotaExceeded `yaml:"quota-exceeded" json:"quota-exceeded"`
 
+	// Routing controls credential selection behavior.
+	Routing RoutingConfig `yaml:"routing" json:"routing"`
+
 	// WebsocketAuth enables or disables authentication for the WebSocket API.
 	WebsocketAuth bool `yaml:"ws-auth" json:"ws-auth"`
 
+	// CodexInstructionsEnabled controls whether official Codex instructions are injected.
+	// When false (default), CodexInstructionsForModel returns immediately without modification.
+	// When true, the original instruction injection logic is used.
+	CodexInstructionsEnabled bool `yaml:"codex-instructions-enabled" json:"codex-instructions-enabled"`
+
 	// GeminiKey defines Gemini API key configurations with optional routing overrides.
 	GeminiKey []GeminiKey `yaml:"gemini-api-key" json:"gemini-api-key"`
 
@@ -80,6 +102,14 @@ type Config struct {
 	// OAuthExcludedModels defines per-provider global model exclusions applied to OAuth/file-backed auth entries.
 	OAuthExcludedModels map[string][]string `yaml:"oauth-excluded-models,omitempty" json:"oauth-excluded-models,omitempty"`
 
+	// OAuthModelAlias defines global model name aliases for OAuth/file-backed auth channels.
+	// These aliases affect both model listing and model routing for supported channels:
+	// gemini-cli, vertex, aistudio, antigravity, claude, codex, qwen, iflow.
+	//
+	// NOTE: This does not apply to existing per-credential model alias features under:
+	// gemini-api-key, codex-api-key, claude-api-key, openai-compatibility, vertex-api-key, and ampcode.
+	OAuthModelAlias map[string][]OAuthModelAlias `yaml:"oauth-model-alias,omitempty" json:"oauth-model-alias,omitempty"`
+
 	// Payload defines default and override rules for provider payload parameters.
 	Payload PayloadConfig `yaml:"payload" json:"payload"`
 
@@ -104,6 +134,9 @@ type RemoteManagement struct {
 	SecretKey string `yaml:"secret-key"`
 	// DisableControlPanel skips serving and syncing the bundled management UI when true.
 	DisableControlPanel bool `yaml:"disable-control-panel"`
+	// PanelGitHubRepository overrides the GitHub repository used to fetch the management panel asset.
+	// Accepts either a repository URL (https://github.com/org/repo) or an API releases endpoint.
+	PanelGitHubRepository string `yaml:"panel-github-repository"`
 }
 
 // QuotaExceeded defines the behavior when API quota limits are exceeded.
@@ -116,6 +149,23 @@ type QuotaExceeded struct {
 	SwitchPreviewModel bool `yaml:"switch-preview-model" json:"switch-preview-model"`
 }
 
+// RoutingConfig configures how credentials are selected for requests.
+type RoutingConfig struct {
+	// Strategy selects the credential selection strategy.
+	// Supported values: "round-robin" (default), "fill-first".
+	Strategy string `yaml:"strategy,omitempty" json:"strategy,omitempty"`
+}
+
+// OAuthModelAlias defines a model ID alias for a specific channel.
+// It maps the upstream model name (Name) to the client-visible alias (Alias).
+// When Fork is true, the alias is added as an additional model in listings while
+// keeping the original model ID available.
+type OAuthModelAlias struct {
+	Name  string `yaml:"name" json:"name"`
+	Alias string `yaml:"alias" json:"alias"`
+	Fork  bool   `yaml:"fork,omitempty" json:"fork,omitempty"`
+}
+
 // AmpModelMapping defines a model name mapping for Amp CLI requests.
 // When Amp requests a model that isn't available locally, this mapping
 // allows routing to an alternative model that IS available.
@@ -126,6 +176,11 @@ type AmpModelMapping struct {
 	// To is the target model name to route to (e.g., "claude-sonnet-4").
 	// The target model must have available providers in the registry.
 	To string `yaml:"to" json:"to"`
+
+	// Regex indicates whether the 'from' field should be interpreted as a regular
+	// expression for matching model names. When true, this mapping is evaluated
+	// after exact matches and in the order provided. Defaults to false (exact match).
+	Regex bool `yaml:"regex,omitempty" json:"regex,omitempty"`
 }
 
 // AmpCode groups Amp CLI integration settings including upstream routing,
@@ -137,9 +192,14 @@ type AmpCode struct {
 	// UpstreamAPIKey optionally overrides the Authorization header when proxying Amp upstream calls.
 	UpstreamAPIKey string `yaml:"upstream-api-key" json:"upstream-api-key"`
 
+	// UpstreamAPIKeys maps client API keys (from top-level api-keys) to upstream API keys.
+	// When a client authenticates with a key that matches an entry, that upstream key is used.
+	// If no match is found, falls back to UpstreamAPIKey (default behavior).
+	UpstreamAPIKeys []AmpUpstreamAPIKeyEntry `yaml:"upstream-api-keys,omitempty" json:"upstream-api-keys,omitempty"`
+
 	// RestrictManagementToLocalhost restricts Amp management routes (/api/user, /api/threads, etc.)
 	// to only accept connections from localhost (127.0.0.1, ::1). When true, prevents drive-by
-	// browser attacks and remote access to management endpoints. Default: true (recommended).
+	// browser attacks and remote access to management endpoints. Default: false (API key auth is sufficient).
 	RestrictManagementToLocalhost bool `yaml:"restrict-management-to-localhost" json:"restrict-management-to-localhost"`
 
 	// ModelMappings defines model name mappings for Amp CLI requests.
@@ -152,12 +212,37 @@ type AmpCode struct {
 	ForceModelMappings bool `yaml:"force-model-mappings" json:"force-model-mappings"`
 }
 
+// AmpUpstreamAPIKeyEntry maps a set of client API keys to a specific upstream API key.
+// When a request is authenticated with one of the APIKeys, the corresponding UpstreamAPIKey
+// is used for the upstream Amp request.
+type AmpUpstreamAPIKeyEntry struct {
+	// UpstreamAPIKey is the API key to use when proxying to the Amp upstream.
+	UpstreamAPIKey string `yaml:"upstream-api-key" json:"upstream-api-key"`
+
+	// APIKeys are the client API keys (from top-level api-keys) that map to this upstream key.
+	APIKeys []string `yaml:"api-keys" json:"api-keys"`
+}
+
 // PayloadConfig defines default and override parameter rules applied to provider payloads.
 type PayloadConfig struct {
 	// Default defines rules that only set parameters when they are missing in the payload.
 	Default []PayloadRule `yaml:"default" json:"default"`
+	// DefaultRaw defines rules that set raw JSON values only when they are missing.
+	DefaultRaw []PayloadRule `yaml:"default-raw" json:"default-raw"`
 	// Override defines rules that always set parameters, overwriting any existing values.
 	Override []PayloadRule `yaml:"override" json:"override"`
+	// OverrideRaw defines rules that always set raw JSON values, overwriting any existing values.
+	OverrideRaw []PayloadRule `yaml:"override-raw" json:"override-raw"`
+	// Filter defines rules that remove parameters from the payload by JSON path.
+	Filter []PayloadFilterRule `yaml:"filter" json:"filter"`
+}
+
+// PayloadFilterRule describes a rule to remove specific JSON paths from matching model payloads.
+type PayloadFilterRule struct {
+	// Models lists model entries with name pattern and protocol constraint.
+	Models []PayloadModelRule `yaml:"models" json:"models"`
+	// Params lists JSON paths (gjson/sjson syntax) to remove from the payload.
+	Params []string `yaml:"params" json:"params"`
 }
 
 // PayloadRule describes a single rule targeting a list of models with parameter updates.
@@ -165,6 +250,7 @@ type PayloadRule struct {
 	// Models lists model entries with name pattern and protocol constraint.
 	Models []PayloadModelRule `yaml:"models" json:"models"`
 	// Params maps JSON paths (gjson/sjson syntax) to values written into the payload.
+	// For *-raw rules, values are treated as raw JSON fragments (strings are used as-is).
 	Params map[string]any `yaml:"params" json:"params"`
 }
 
@@ -176,12 +262,38 @@ type PayloadModelRule struct {
 	Protocol string `yaml:"protocol" json:"protocol"`
 }
 
+// CloakConfig configures request cloaking for non-Claude-Code clients.
+// Cloaking disguises API requests to appear as originating from the official Claude Code CLI.
+type CloakConfig struct {
+	// Mode controls cloaking behavior: "auto" (default), "always", or "never".
+	// - "auto": cloak only when client is not Claude Code (based on User-Agent)
+	// - "always": always apply cloaking regardless of client
+	// - "never": never apply cloaking
+	Mode string `yaml:"mode,omitempty" json:"mode,omitempty"`
+
+	// StrictMode controls how system prompts are handled when cloaking.
+	// - false (default): prepend Claude Code prompt to user system messages
+	// - true: strip all user system messages, keep only Claude Code prompt
+	StrictMode bool `yaml:"strict-mode,omitempty" json:"strict-mode,omitempty"`
+
+	// SensitiveWords is a list of words to obfuscate with zero-width characters.
+	// This can help bypass certain content filters.
+	SensitiveWords []string `yaml:"sensitive-words,omitempty" json:"sensitive-words,omitempty"`
+}
+
 // ClaudeKey represents the configuration for a Claude API key,
 // including the API key itself and an optional base URL for the API endpoint.
 type ClaudeKey struct {
 	// APIKey is the authentication key for accessing Claude API services.
 	APIKey string `yaml:"api-key" json:"api-key"`
 
+	// Priority controls selection preference when multiple credentials match.
+	// Higher values are preferred; defaults to 0.
+	Priority int `yaml:"priority,omitempty" json:"priority,omitempty"`
+
+	// Prefix optionally namespaces models for this credential (e.g., "teamA/claude-sonnet-4").
+	Prefix string `yaml:"prefix,omitempty" json:"prefix,omitempty"`
+
 	// BaseURL is the base URL for the Claude API endpoint.
 	// If empty, the default Claude API URL will be used.
 	BaseURL string `yaml:"base-url" json:"base-url"`
@@ -197,8 +309,14 @@ type ClaudeKey struct {
 
 	// ExcludedModels lists model IDs that should be excluded for this provider.
 	ExcludedModels []string `yaml:"excluded-models,omitempty" json:"excluded-models,omitempty"`
+
+	// Cloak configures request cloaking for non-Claude-Code clients.
+	Cloak *CloakConfig `yaml:"cloak,omitempty" json:"cloak,omitempty"`
 }
 
+func (k ClaudeKey) GetAPIKey() string  { return k.APIKey }
+func (k ClaudeKey) GetBaseURL() string { return k.BaseURL }
+
 // ClaudeModel describes a mapping between an alias and the actual upstream model name.
 type ClaudeModel struct {
 	// Name is the upstream model identifier used when issuing requests.
@@ -208,12 +326,22 @@ type ClaudeModel struct {
 	Alias string `yaml:"alias" json:"alias"`
 }
 
+func (m ClaudeModel) GetName() string  { return m.Name }
+func (m ClaudeModel) GetAlias() string { return m.Alias }
+
 // CodexKey represents the configuration for a Codex API key,
 // including the API key itself and an optional base URL for the API endpoint.
 type CodexKey struct {
 	// APIKey is the authentication key for accessing Codex API services.
 	APIKey string `yaml:"api-key" json:"api-key"`
 
+	// Priority controls selection preference when multiple credentials match.
+	// Higher values are preferred; defaults to 0.
+	Priority int `yaml:"priority,omitempty" json:"priority,omitempty"`
+
+	// Prefix optionally namespaces models for this credential (e.g., "teamA/gpt-5-codex").
+	Prefix string `yaml:"prefix,omitempty" json:"prefix,omitempty"`
+
 	// BaseURL is the base URL for the Codex API endpoint.
 	// If empty, the default Codex API URL will be used.
 	BaseURL string `yaml:"base-url" json:"base-url"`
@@ -221,6 +349,9 @@ type CodexKey struct {
 	// ProxyURL overrides the global proxy setting for this API key if provided.
 	ProxyURL string `yaml:"proxy-url" json:"proxy-url"`
 
+	// Models defines upstream model names and aliases for request routing.
+	Models []CodexModel `yaml:"models" json:"models"`
+
 	// Headers optionally adds extra HTTP headers for requests sent with this key.
 	Headers map[string]string `yaml:"headers,omitempty" json:"headers,omitempty"`
 
@@ -228,18 +359,43 @@ type CodexKey struct {
 	ExcludedModels []string `yaml:"excluded-models,omitempty" json:"excluded-models,omitempty"`
 }
 
+func (k CodexKey) GetAPIKey() string  { return k.APIKey }
+func (k CodexKey) GetBaseURL() string { return k.BaseURL }
+
+// CodexModel describes a mapping between an alias and the actual upstream model name.
+type CodexModel struct {
+	// Name is the upstream model identifier used when issuing requests.
+	Name string `yaml:"name" json:"name"`
+
+	// Alias is the client-facing model name that maps to Name.
+	Alias string `yaml:"alias" json:"alias"`
+}
+
+func (m CodexModel) GetName() string  { return m.Name }
+func (m CodexModel) GetAlias() string { return m.Alias }
+
 // GeminiKey represents the configuration for a Gemini API key,
 // including optional overrides for upstream base URL, proxy routing, and headers.
 type GeminiKey struct {
 	// APIKey is the authentication key for accessing Gemini API services.
 	APIKey string `yaml:"api-key" json:"api-key"`
 
+	// Priority controls selection preference when multiple credentials match.
+	// Higher values are preferred; defaults to 0.
+	Priority int `yaml:"priority,omitempty" json:"priority,omitempty"`
+
+	// Prefix optionally namespaces models for this credential (e.g., "teamA/gemini-3-pro-preview").
+	Prefix string `yaml:"prefix,omitempty" json:"prefix,omitempty"`
+
 	// BaseURL optionally overrides the Gemini API endpoint.
 	BaseURL string `yaml:"base-url,omitempty" json:"base-url,omitempty"`
 
 	// ProxyURL optionally overrides the global proxy for this API key.
 	ProxyURL string `yaml:"proxy-url,omitempty" json:"proxy-url,omitempty"`
 
+	// Models defines upstream model names and aliases for request routing.
+	Models []GeminiModel `yaml:"models,omitempty" json:"models,omitempty"`
+
 	// Headers optionally adds extra HTTP headers for requests sent with this key.
 	Headers map[string]string `yaml:"headers,omitempty" json:"headers,omitempty"`
 
@@ -247,12 +403,34 @@ type GeminiKey struct {
 	ExcludedModels []string `yaml:"excluded-models,omitempty" json:"excluded-models,omitempty"`
 }
 
+func (k GeminiKey) GetAPIKey() string  { return k.APIKey }
+func (k GeminiKey) GetBaseURL() string { return k.BaseURL }
+
+// GeminiModel describes a mapping between an alias and the actual upstream model name.
+type GeminiModel struct {
+	// Name is the upstream model identifier used when issuing requests.
+	Name string `yaml:"name" json:"name"`
+
+	// Alias is the client-facing model name that maps to Name.
+	Alias string `yaml:"alias" json:"alias"`
+}
+
+func (m GeminiModel) GetName() string  { return m.Name }
+func (m GeminiModel) GetAlias() string { return m.Alias }
+
 // OpenAICompatibility represents the configuration for OpenAI API compatibility
 // with external providers, allowing model aliases to be routed through OpenAI API format.
 type OpenAICompatibility struct {
 	// Name is the identifier for this OpenAI compatibility configuration.
 	Name string `yaml:"name" json:"name"`
 
+	// Priority controls selection preference when multiple providers or credentials match.
+	// Higher values are preferred; defaults to 0.
+	Priority int `yaml:"priority,omitempty" json:"priority,omitempty"`
+
+	// Prefix optionally namespaces model aliases for this provider (e.g., "teamA/kimi-k2").
+	Prefix string `yaml:"prefix,omitempty" json:"prefix,omitempty"`
+
 	// BaseURL is the base URL for the external OpenAI-compatible API endpoint.
 	BaseURL string `yaml:"base-url" json:"base-url"`
 
@@ -285,6 +463,9 @@ type OpenAICompatibilityModel struct {
 	Alias string `yaml:"alias" json:"alias"`
 }
 
+func (m OpenAICompatibilityModel) GetName() string  { return m.Name }
+func (m OpenAICompatibilityModel) GetAlias() string { return m.Alias }
+
 // LoadConfig reads a YAML configuration file from the given path,
 // unmarshals it into a Config struct, applies environment variable overrides,
 // and returns it.
@@ -303,6 +484,15 @@ func LoadConfig(configFile string) (*Config, error) {
 // If optional is true and the file is missing, it returns an empty Config.
 // If optional is true and the file is empty or invalid, it returns an empty Config.
 func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
+	// Perform oauth-model-alias migration before loading config.
+	// This migrates oauth-model-mappings to oauth-model-alias if needed.
+	if migrated, err := MigrateOAuthModelAlias(configFile); err != nil {
+		// Log warning but don't fail - config loading should still work
+		fmt.Printf("Warning: oauth-model-alias migration failed: %v\n", err)
+	} else if migrated {
+		fmt.Println("Migrated oauth-model-mappings to oauth-model-alias")
+	}
+
 	// Read the entire configuration file into memory.
 	data, err := os.ReadFile(configFile)
 	if err != nil {
@@ -325,9 +515,12 @@ func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
 	// Set defaults before unmarshal so that absent keys keep defaults.
 	cfg.Host = "" // Default empty: binds to all interfaces (IPv4 + IPv6)
 	cfg.LoggingToFile = false
+	cfg.LogsMaxTotalSizeMB = 0
+	cfg.ErrorLogsMaxFiles = 10
 	cfg.UsageStatisticsEnabled = false
 	cfg.DisableCooling = false
-	cfg.AmpCode.RestrictManagementToLocalhost = true // Default to secure: only localhost access
+	cfg.AmpCode.RestrictManagementToLocalhost = false // Default to false: API key auth is sufficient
+	cfg.RemoteManagement.PanelGitHubRepository = DefaultPanelGitHubRepository
 	if err = yaml.Unmarshal(data, &cfg); err != nil {
 		if optional {
 			// In cloud deploy mode, if YAML parsing fails, return empty config instead of error.
@@ -363,6 +556,19 @@ func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
 		_ = SaveConfigPreserveCommentsUpdateNestedScalar(configFile, []string{"remote-management", "secret-key"}, hashed)
 	}
 
+	cfg.RemoteManagement.PanelGitHubRepository = strings.TrimSpace(cfg.RemoteManagement.PanelGitHubRepository)
+	if cfg.RemoteManagement.PanelGitHubRepository == "" {
+		cfg.RemoteManagement.PanelGitHubRepository = DefaultPanelGitHubRepository
+	}
+
+	if cfg.LogsMaxTotalSizeMB < 0 {
+		cfg.LogsMaxTotalSizeMB = 0
+	}
+
+	if cfg.ErrorLogsMaxFiles < 0 {
+		cfg.ErrorLogsMaxFiles = 10
+	}
+
 	// Sync request authentication providers with inline API keys for backwards compatibility.
 	syncInlineAccessProvider(&cfg)
 
@@ -384,6 +590,12 @@ func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
 	// Normalize OAuth provider model exclusion map.
 	cfg.OAuthExcludedModels = NormalizeOAuthExcludedModels(cfg.OAuthExcludedModels)
 
+	// Normalize global OAuth model name aliases.
+	cfg.SanitizeOAuthModelAlias()
+
+	// Validate raw payload rules and drop invalid entries.
+	cfg.SanitizePayloadRules()
+
 	if cfg.legacyMigrationPending {
 		fmt.Println("Detected legacy configuration keys, attempting to persist the normalized config...")
 		if !optional && configFile != "" {
@@ -400,6 +612,99 @@ func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
 	return &cfg, nil
 }
 
+// SanitizePayloadRules validates raw JSON payload rule params and drops invalid rules.
+func (cfg *Config) SanitizePayloadRules() {
+	if cfg == nil {
+		return
+	}
+	cfg.Payload.DefaultRaw = sanitizePayloadRawRules(cfg.Payload.DefaultRaw, "default-raw")
+	cfg.Payload.OverrideRaw = sanitizePayloadRawRules(cfg.Payload.OverrideRaw, "override-raw")
+}
+
+func sanitizePayloadRawRules(rules []PayloadRule, section string) []PayloadRule {
+	if len(rules) == 0 {
+		return rules
+	}
+	out := make([]PayloadRule, 0, len(rules))
+	for i := range rules {
+		rule := rules[i]
+		if len(rule.Params) == 0 {
+			continue
+		}
+		invalid := false
+		for path, value := range rule.Params {
+			raw, ok := payloadRawString(value)
+			if !ok {
+				continue
+			}
+			trimmed := bytes.TrimSpace(raw)
+			if len(trimmed) == 0 || !json.Valid(trimmed) {
+				log.WithFields(log.Fields{
+					"section":    section,
+					"rule_index": i + 1,
+					"param":      path,
+				}).Warn("payload rule dropped: invalid raw JSON")
+				invalid = true
+				break
+			}
+		}
+		if invalid {
+			continue
+		}
+		out = append(out, rule)
+	}
+	return out
+}
+
+func payloadRawString(value any) ([]byte, bool) {
+	switch typed := value.(type) {
+	case string:
+		return []byte(typed), true
+	case []byte:
+		return typed, true
+	default:
+		return nil, false
+	}
+}
+
+// SanitizeOAuthModelAlias normalizes and deduplicates global OAuth model name aliases.
+// It trims whitespace, normalizes channel keys to lower-case, drops empty entries,
+// allows multiple aliases per upstream name, and ensures aliases are unique within each channel.
+func (cfg *Config) SanitizeOAuthModelAlias() {
+	if cfg == nil || len(cfg.OAuthModelAlias) == 0 {
+		return
+	}
+	out := make(map[string][]OAuthModelAlias, len(cfg.OAuthModelAlias))
+	for rawChannel, aliases := range cfg.OAuthModelAlias {
+		channel := strings.ToLower(strings.TrimSpace(rawChannel))
+		if channel == "" || len(aliases) == 0 {
+			continue
+		}
+		seenAlias := make(map[string]struct{}, len(aliases))
+		clean := make([]OAuthModelAlias, 0, len(aliases))
+		for _, entry := range aliases {
+			name := strings.TrimSpace(entry.Name)
+			alias := strings.TrimSpace(entry.Alias)
+			if name == "" || alias == "" {
+				continue
+			}
+			if strings.EqualFold(name, alias) {
+				continue
+			}
+			aliasKey := strings.ToLower(alias)
+			if _, ok := seenAlias[aliasKey]; ok {
+				continue
+			}
+			seenAlias[aliasKey] = struct{}{}
+			clean = append(clean, OAuthModelAlias{Name: name, Alias: alias, Fork: entry.Fork})
+		}
+		if len(clean) > 0 {
+			out[channel] = clean
+		}
+	}
+	cfg.OAuthModelAlias = out
+}
+
 // SanitizeOpenAICompatibility removes OpenAI-compatibility provider entries that are
 // not actionable, specifically those missing a BaseURL. It trims whitespace before
 // evaluation and preserves the relative order of remaining entries.
@@ -411,6 +716,7 @@ func (cfg *Config) SanitizeOpenAICompatibility() {
 	for i := range cfg.OpenAICompatibility {
 		e := cfg.OpenAICompatibility[i]
 		e.Name = strings.TrimSpace(e.Name)
+		e.Prefix = normalizeModelPrefix(e.Prefix)
 		e.BaseURL = strings.TrimSpace(e.BaseURL)
 		e.Headers = NormalizeHeaders(e.Headers)
 		if e.BaseURL == "" {
@@ -431,6 +737,7 @@ func (cfg *Config) SanitizeCodexKeys() {
 	out := make([]CodexKey, 0, len(cfg.CodexKey))
 	for i := range cfg.CodexKey {
 		e := cfg.CodexKey[i]
+		e.Prefix = normalizeModelPrefix(e.Prefix)
 		e.BaseURL = strings.TrimSpace(e.BaseURL)
 		e.Headers = NormalizeHeaders(e.Headers)
 		e.ExcludedModels = NormalizeExcludedModels(e.ExcludedModels)
@@ -449,6 +756,7 @@ func (cfg *Config) SanitizeClaudeKeys() {
 	}
 	for i := range cfg.ClaudeKey {
 		entry := &cfg.ClaudeKey[i]
+		entry.Prefix = normalizeModelPrefix(entry.Prefix)
 		entry.Headers = NormalizeHeaders(entry.Headers)
 		entry.ExcludedModels = NormalizeExcludedModels(entry.ExcludedModels)
 	}
@@ -468,6 +776,7 @@ func (cfg *Config) SanitizeGeminiKeys() {
 		if entry.APIKey == "" {
 			continue
 		}
+		entry.Prefix = normalizeModelPrefix(entry.Prefix)
 		entry.BaseURL = strings.TrimSpace(entry.BaseURL)
 		entry.ProxyURL = strings.TrimSpace(entry.ProxyURL)
 		entry.Headers = NormalizeHeaders(entry.Headers)
@@ -481,6 +790,18 @@ func (cfg *Config) SanitizeGeminiKeys() {
 	cfg.GeminiKey = out
 }
 
+func normalizeModelPrefix(prefix string) string {
+	trimmed := strings.TrimSpace(prefix)
+	trimmed = strings.Trim(trimmed, "/")
+	if trimmed == "" {
+		return ""
+	}
+	if strings.Contains(trimmed, "/") {
+		return ""
+	}
+	return trimmed
+}
+
 func syncInlineAccessProvider(cfg *Config) {
 	if cfg == nil {
 		return
@@ -621,6 +942,7 @@ func SaveConfigPreserveComments(configFile string, cfg *Config) error {
 	removeLegacyGenerativeLanguageKeys(original.Content[0])
 
 	pruneMappingToGeneratedKeys(original.Content[0], generated.Content[0], "oauth-excluded-models")
+	pruneMappingToGeneratedKeys(original.Content[0], generated.Content[0], "oauth-model-alias")
 
 	// Merge generated into original in-place, preserving comments/order of existing nodes.
 	mergeMappingPreserve(original.Content[0], generated.Content[0])
@@ -653,7 +975,7 @@ func sanitizeConfigForPersist(cfg *Config) *Config {
 	}
 	clone := *cfg
 	clone.SDKConfig = cfg.SDKConfig
-	clone.SDKConfig.Access = config.AccessConfig{}
+	clone.SDKConfig.Access = AccessConfig{}
 	return &clone
 }
 
@@ -752,8 +1074,8 @@ func getOrCreateMapValue(mapNode *yaml.Node, key string) *yaml.Node {
 }
 
 // mergeMappingPreserve merges keys from src into dst mapping node while preserving
-// key order and comments of existing keys in dst. Unknown keys from src are appended
-// to dst at the end, copying their node structure from src.
+// key order and comments of existing keys in dst. New keys are only added if their
+// value is non-zero to avoid polluting the config with defaults.
 func mergeMappingPreserve(dst, src *yaml.Node) {
 	if dst == nil || src == nil {
 		return
@@ -764,20 +1086,19 @@ func mergeMappingPreserve(dst, src *yaml.Node) {
 		copyNodeShallow(dst, src)
 		return
 	}
-	// Build a lookup of existing keys in dst
 	for i := 0; i+1 < len(src.Content); i += 2 {
 		sk := src.Content[i]
 		sv := src.Content[i+1]
 		idx := findMapKeyIndex(dst, sk.Value)
 		if idx >= 0 {
-			// Merge into existing value node
+			// Merge into existing value node (always update, even to zero values)
 			dv := dst.Content[idx+1]
 			mergeNodePreserve(dv, sv)
 		} else {
-			if shouldSkipEmptyCollectionOnPersist(sk.Value, sv) {
+			// New key: only add if value is non-zero to avoid polluting config with defaults
+			if isZeroValueNode(sv) {
 				continue
 			}
-			// Append new key/value pair by deep-copying from src
 			dst.Content = append(dst.Content, deepCopyNode(sk), deepCopyNode(sv))
 		}
 	}
@@ -860,33 +1181,50 @@ func findMapKeyIndex(mapNode *yaml.Node, key string) int {
 	return -1
 }
 
-func shouldSkipEmptyCollectionOnPersist(key string, node *yaml.Node) bool {
-	switch key {
-	case "generative-language-api-key",
-		"gemini-api-key",
-		"vertex-api-key",
-		"claude-api-key",
-		"codex-api-key",
-		"openai-compatibility":
-		return isEmptyCollectionNode(node)
-	default:
-		return false
-	}
-}
-
-func isEmptyCollectionNode(node *yaml.Node) bool {
+// isZeroValueNode returns true if the YAML node represents a zero/default value
+// that should not be written as a new key to preserve config cleanliness.
+// For mappings and sequences, recursively checks if all children are zero values.
+func isZeroValueNode(node *yaml.Node) bool {
 	if node == nil {
 		return true
 	}
 	switch node.Kind {
-	case yaml.SequenceNode:
-		return len(node.Content) == 0
 	case yaml.ScalarNode:
-		return node.Tag == "!!null"
-	default:
+		switch node.Tag {
+		case "!!bool":
+			return node.Value == "false"
+		case "!!int", "!!float":
+			return node.Value == "0" || node.Value == "0.0"
+		case "!!str":
+			return node.Value == ""
+		case "!!null":
+			return true
+		}
+	case yaml.SequenceNode:
+		if len(node.Content) == 0 {
+			return true
+		}
+		// Check if all elements are zero values
+		for _, child := range node.Content {
+			if !isZeroValueNode(child) {
 				return false
 			}
 		}
+		return true
+	case yaml.MappingNode:
+		if len(node.Content) == 0 {
+			return true
+		}
+		// Check if all values are zero values (values are at odd indices)
+		for i := 1; i < len(node.Content); i += 2 {
+			if !isZeroValueNode(node.Content[i]) {
+				return false
+			}
+		}
+		return true
+	}
+	return false
+}
 
 // deepCopyNode creates a deep copy of a yaml.Node graph.
 func deepCopyNode(n *yaml.Node) *yaml.Node {
@@ -1095,6 +1433,16 @@ func pruneMappingToGeneratedKeys(dstRoot, srcRoot *yaml.Node, key string) {
 	}
 	srcIdx := findMapKeyIndex(srcRoot, key)
 	if srcIdx < 0 {
+		// Keep an explicit empty mapping for oauth-model-alias when it was previously present.
+		//
+		// Rationale: LoadConfig runs MigrateOAuthModelAlias before unmarshalling. If the
+		// oauth-model-alias key is missing, migration will add the default antigravity aliases.
+		// When users delete the last channel from oauth-model-alias via the management API,
+		// we want that deletion to persist across hot reloads and restarts.
+		if key == "oauth-model-alias" {
+			dstRoot.Content[dstIdx+1] = &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+			return
+		}
 		removeMapKey(dstRoot, key)
 		return
 	}

internal/config/oauth_model_alias_migration.go

@@ -0,0 +1,275 @@
+package config
+
+import (
+	"os"
+	"strings"
+
+	"gopkg.in/yaml.v3"
+)
+
+// antigravityModelConversionTable maps old built-in aliases to actual model names
+// for the antigravity channel during migration.
+var antigravityModelConversionTable = map[string]string{
+	"gemini-2.5-computer-use-preview-10-2025": "rev19-uic3-1p",
+	"gemini-3-pro-image-preview":              "gemini-3-pro-image",
+	"gemini-3-pro-preview":                    "gemini-3-pro-high",
+	"gemini-3-flash-preview":                  "gemini-3-flash",
+	"gemini-claude-sonnet-4-5":                "claude-sonnet-4-5",
+	"gemini-claude-sonnet-4-5-thinking":       "claude-sonnet-4-5-thinking",
+	"gemini-claude-opus-4-5-thinking":         "claude-opus-4-5-thinking",
+}
+
+// defaultAntigravityAliases returns the default oauth-model-alias configuration
+// for the antigravity channel when neither field exists.
+func defaultAntigravityAliases() []OAuthModelAlias {
+	return []OAuthModelAlias{
+		{Name: "rev19-uic3-1p", Alias: "gemini-2.5-computer-use-preview-10-2025"},
+		{Name: "gemini-3-pro-image", Alias: "gemini-3-pro-image-preview"},
+		{Name: "gemini-3-pro-high", Alias: "gemini-3-pro-preview"},
+		{Name: "gemini-3-flash", Alias: "gemini-3-flash-preview"},
+		{Name: "claude-sonnet-4-5", Alias: "gemini-claude-sonnet-4-5"},
+		{Name: "claude-sonnet-4-5-thinking", Alias: "gemini-claude-sonnet-4-5-thinking"},
+		{Name: "claude-opus-4-5-thinking", Alias: "gemini-claude-opus-4-5-thinking"},
+	}
+}
+
+// MigrateOAuthModelAlias checks for and performs migration from oauth-model-mappings
+// to oauth-model-alias at startup. Returns true if migration was performed.
+//
+// Migration flow:
+// 1. Check if oauth-model-alias exists -> skip migration
+// 2. Check if oauth-model-mappings exists -> convert and migrate
+//   - For antigravity channel, convert old built-in aliases to actual model names
+//
+// 3. Neither exists -> add default antigravity config
+func MigrateOAuthModelAlias(configFile string) (bool, error) {
+	data, err := os.ReadFile(configFile)
+	if err != nil {
+		if os.IsNotExist(err) {
+			return false, nil
+		}
+		return false, err
+	}
+	if len(data) == 0 {
+		return false, nil
+	}
+
+	// Parse YAML into node tree to preserve structure
+	var root yaml.Node
+	if err := yaml.Unmarshal(data, &root); err != nil {
+		return false, nil
+	}
+	if root.Kind != yaml.DocumentNode || len(root.Content) == 0 {
+		return false, nil
+	}
+	rootMap := root.Content[0]
+	if rootMap == nil || rootMap.Kind != yaml.MappingNode {
+		return false, nil
+	}
+
+	// Check if oauth-model-alias already exists
+	if findMapKeyIndex(rootMap, "oauth-model-alias") >= 0 {
+		return false, nil
+	}
+
+	// Check if oauth-model-mappings exists
+	oldIdx := findMapKeyIndex(rootMap, "oauth-model-mappings")
+	if oldIdx >= 0 {
+		// Migrate from old field
+		return migrateFromOldField(configFile, &root, rootMap, oldIdx)
+	}
+
+	// Neither field exists - add default antigravity config
+	return addDefaultAntigravityConfig(configFile, &root, rootMap)
+}
+
+// migrateFromOldField converts oauth-model-mappings to oauth-model-alias
+func migrateFromOldField(configFile string, root *yaml.Node, rootMap *yaml.Node, oldIdx int) (bool, error) {
+	if oldIdx+1 >= len(rootMap.Content) {
+		return false, nil
+	}
+	oldValue := rootMap.Content[oldIdx+1]
+	if oldValue == nil || oldValue.Kind != yaml.MappingNode {
+		return false, nil
+	}
+
+	// Parse the old aliases
+	oldAliases := parseOldAliasNode(oldValue)
+	if len(oldAliases) == 0 {
+		// Remove the old field and write
+		removeMapKeyByIndex(rootMap, oldIdx)
+		return writeYAMLNode(configFile, root)
+	}
+
+	// Convert model names for antigravity channel
+	newAliases := make(map[string][]OAuthModelAlias, len(oldAliases))
+	for channel, entries := range oldAliases {
+		converted := make([]OAuthModelAlias, 0, len(entries))
+		for _, entry := range entries {
+			newEntry := OAuthModelAlias{
+				Name:  entry.Name,
+				Alias: entry.Alias,
+				Fork:  entry.Fork,
+			}
+			// Convert model names for antigravity channel
+			if strings.EqualFold(channel, "antigravity") {
+				if actual, ok := antigravityModelConversionTable[entry.Name]; ok {
+					newEntry.Name = actual
+				}
+			}
+			converted = append(converted, newEntry)
+		}
+		newAliases[channel] = converted
+	}
+
+	// For antigravity channel, supplement missing default aliases
+	if antigravityEntries, exists := newAliases["antigravity"]; exists {
+		// Build a set of already configured model names (upstream names)
+		configuredModels := make(map[string]bool, len(antigravityEntries))
+		for _, entry := range antigravityEntries {
+			configuredModels[entry.Name] = true
+		}
+
+		// Add missing default aliases
+		for _, defaultAlias := range defaultAntigravityAliases() {
+			if !configuredModels[defaultAlias.Name] {
+				antigravityEntries = append(antigravityEntries, defaultAlias)
+			}
+		}
+		newAliases["antigravity"] = antigravityEntries
+	}
+
+	// Build new node
+	newNode := buildOAuthModelAliasNode(newAliases)
+
+	// Replace old key with new key and value
+	rootMap.Content[oldIdx].Value = "oauth-model-alias"
+	rootMap.Content[oldIdx+1] = newNode
+
+	return writeYAMLNode(configFile, root)
+}
+
+// addDefaultAntigravityConfig adds the default antigravity configuration
+func addDefaultAntigravityConfig(configFile string, root *yaml.Node, rootMap *yaml.Node) (bool, error) {
+	defaults := map[string][]OAuthModelAlias{
+		"antigravity": defaultAntigravityAliases(),
+	}
+	newNode := buildOAuthModelAliasNode(defaults)
+
+	// Add new key-value pair
+	keyNode := &yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: "oauth-model-alias"}
+	rootMap.Content = append(rootMap.Content, keyNode, newNode)
+
+	return writeYAMLNode(configFile, root)
+}
+
+// parseOldAliasNode parses the old oauth-model-mappings node structure
+func parseOldAliasNode(node *yaml.Node) map[string][]OAuthModelAlias {
+	if node == nil || node.Kind != yaml.MappingNode {
+		return nil
+	}
+	result := make(map[string][]OAuthModelAlias)
+	for i := 0; i+1 < len(node.Content); i += 2 {
+		channelNode := node.Content[i]
+		entriesNode := node.Content[i+1]
+		if channelNode == nil || entriesNode == nil {
+			continue
+		}
+		channel := strings.ToLower(strings.TrimSpace(channelNode.Value))
+		if channel == "" || entriesNode.Kind != yaml.SequenceNode {
+			continue
+		}
+		entries := make([]OAuthModelAlias, 0, len(entriesNode.Content))
+		for _, entryNode := range entriesNode.Content {
+			if entryNode == nil || entryNode.Kind != yaml.MappingNode {
+				continue
+			}
+			entry := parseAliasEntry(entryNode)
+			if entry.Name != "" && entry.Alias != "" {
+				entries = append(entries, entry)
+			}
+		}
+		if len(entries) > 0 {
+			result[channel] = entries
+		}
+	}
+	return result
+}
+
+// parseAliasEntry parses a single alias entry node
+func parseAliasEntry(node *yaml.Node) OAuthModelAlias {
+	var entry OAuthModelAlias
+	for i := 0; i+1 < len(node.Content); i += 2 {
+		keyNode := node.Content[i]
+		valNode := node.Content[i+1]
+		if keyNode == nil || valNode == nil {
+			continue
+		}
+		switch strings.ToLower(strings.TrimSpace(keyNode.Value)) {
+		case "name":
+			entry.Name = strings.TrimSpace(valNode.Value)
+		case "alias":
+			entry.Alias = strings.TrimSpace(valNode.Value)
+		case "fork":
+			entry.Fork = strings.ToLower(strings.TrimSpace(valNode.Value)) == "true"
+		}
+	}
+	return entry
+}
+
+// buildOAuthModelAliasNode creates a YAML node for oauth-model-alias
+func buildOAuthModelAliasNode(aliases map[string][]OAuthModelAlias) *yaml.Node {
+	node := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+	for channel, entries := range aliases {
+		channelNode := &yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: channel}
+		entriesNode := &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
+		for _, entry := range entries {
+			entryNode := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+			entryNode.Content = append(entryNode.Content,
+				&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: "name"},
+				&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: entry.Name},
+				&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: "alias"},
+				&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: entry.Alias},
+			)
+			if entry.Fork {
+				entryNode.Content = append(entryNode.Content,
+					&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!str", Value: "fork"},
+					&yaml.Node{Kind: yaml.ScalarNode, Tag: "!!bool", Value: "true"},
+				)
+			}
+			entriesNode.Content = append(entriesNode.Content, entryNode)
+		}
+		node.Content = append(node.Content, channelNode, entriesNode)
+	}
+	return node
+}
+
+// removeMapKeyByIndex removes a key-value pair from a mapping node by index
+func removeMapKeyByIndex(mapNode *yaml.Node, keyIdx int) {
+	if mapNode == nil || mapNode.Kind != yaml.MappingNode {
+		return
+	}
+	if keyIdx < 0 || keyIdx+1 >= len(mapNode.Content) {
+		return
+	}
+	mapNode.Content = append(mapNode.Content[:keyIdx], mapNode.Content[keyIdx+2:]...)
+}
+
+// writeYAMLNode writes the YAML node tree back to file
+func writeYAMLNode(configFile string, root *yaml.Node) (bool, error) {
+	f, err := os.Create(configFile)
+	if err != nil {
+		return false, err
+	}
+	defer f.Close()
+
+	enc := yaml.NewEncoder(f)
+	enc.SetIndent(2)
+	if err := enc.Encode(root); err != nil {
+		return false, err
+	}
+	if err := enc.Close(); err != nil {
+		return false, err
+	}
+	return true, nil
+}

internal/config/oauth_model_alias_migration_test.go

@@ -0,0 +1,242 @@
+package config
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"gopkg.in/yaml.v3"
+)
+
+func TestMigrateOAuthModelAlias_SkipsIfNewFieldExists(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	content := `oauth-model-alias:
+  gemini-cli:
+    - name: "gemini-2.5-pro"
+      alias: "g2.5p"
+`
+	if err := os.WriteFile(configFile, []byte(content), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if migrated {
+		t.Fatal("expected no migration when oauth-model-alias already exists")
+	}
+
+	// Verify file unchanged
+	data, _ := os.ReadFile(configFile)
+	if !strings.Contains(string(data), "oauth-model-alias:") {
+		t.Fatal("file should still contain oauth-model-alias")
+	}
+}
+
+func TestMigrateOAuthModelAlias_MigratesOldField(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	content := `oauth-model-mappings:
+  gemini-cli:
+    - name: "gemini-2.5-pro"
+      alias: "g2.5p"
+      fork: true
+`
+	if err := os.WriteFile(configFile, []byte(content), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if !migrated {
+		t.Fatal("expected migration to occur")
+	}
+
+	// Verify new field exists and old field removed
+	data, _ := os.ReadFile(configFile)
+	if strings.Contains(string(data), "oauth-model-mappings:") {
+		t.Fatal("old field should be removed")
+	}
+	if !strings.Contains(string(data), "oauth-model-alias:") {
+		t.Fatal("new field should exist")
+	}
+
+	// Parse and verify structure
+	var root yaml.Node
+	if err := yaml.Unmarshal(data, &root); err != nil {
+		t.Fatal(err)
+	}
+}
+
+func TestMigrateOAuthModelAlias_ConvertsAntigravityModels(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	// Use old model names that should be converted
+	content := `oauth-model-mappings:
+  antigravity:
+    - name: "gemini-2.5-computer-use-preview-10-2025"
+      alias: "computer-use"
+    - name: "gemini-3-pro-preview"
+      alias: "g3p"
+`
+	if err := os.WriteFile(configFile, []byte(content), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if !migrated {
+		t.Fatal("expected migration to occur")
+	}
+
+	// Verify model names were converted
+	data, _ := os.ReadFile(configFile)
+	content = string(data)
+	if !strings.Contains(content, "rev19-uic3-1p") {
+		t.Fatal("expected gemini-2.5-computer-use-preview-10-2025 to be converted to rev19-uic3-1p")
+	}
+	if !strings.Contains(content, "gemini-3-pro-high") {
+		t.Fatal("expected gemini-3-pro-preview to be converted to gemini-3-pro-high")
+	}
+
+	// Verify missing default aliases were supplemented
+	if !strings.Contains(content, "gemini-3-pro-image") {
+		t.Fatal("expected missing default alias gemini-3-pro-image to be added")
+	}
+	if !strings.Contains(content, "gemini-3-flash") {
+		t.Fatal("expected missing default alias gemini-3-flash to be added")
+	}
+	if !strings.Contains(content, "claude-sonnet-4-5") {
+		t.Fatal("expected missing default alias claude-sonnet-4-5 to be added")
+	}
+	if !strings.Contains(content, "claude-sonnet-4-5-thinking") {
+		t.Fatal("expected missing default alias claude-sonnet-4-5-thinking to be added")
+	}
+	if !strings.Contains(content, "claude-opus-4-5-thinking") {
+		t.Fatal("expected missing default alias claude-opus-4-5-thinking to be added")
+	}
+}
+
+func TestMigrateOAuthModelAlias_AddsDefaultIfNeitherExists(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	content := `debug: true
+port: 8080
+`
+	if err := os.WriteFile(configFile, []byte(content), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if !migrated {
+		t.Fatal("expected migration to add default config")
+	}
+
+	// Verify default antigravity config was added
+	data, _ := os.ReadFile(configFile)
+	content = string(data)
+	if !strings.Contains(content, "oauth-model-alias:") {
+		t.Fatal("expected oauth-model-alias to be added")
+	}
+	if !strings.Contains(content, "antigravity:") {
+		t.Fatal("expected antigravity channel to be added")
+	}
+	if !strings.Contains(content, "rev19-uic3-1p") {
+		t.Fatal("expected default antigravity aliases to include rev19-uic3-1p")
+	}
+}
+
+func TestMigrateOAuthModelAlias_PreservesOtherConfig(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	content := `debug: true
+port: 8080
+oauth-model-mappings:
+  gemini-cli:
+    - name: "test"
+      alias: "t"
+api-keys:
+  - "key1"
+  - "key2"
+`
+	if err := os.WriteFile(configFile, []byte(content), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if !migrated {
+		t.Fatal("expected migration to occur")
+	}
+
+	// Verify other config preserved
+	data, _ := os.ReadFile(configFile)
+	content = string(data)
+	if !strings.Contains(content, "debug: true") {
+		t.Fatal("expected debug field to be preserved")
+	}
+	if !strings.Contains(content, "port: 8080") {
+		t.Fatal("expected port field to be preserved")
+	}
+	if !strings.Contains(content, "api-keys:") {
+		t.Fatal("expected api-keys field to be preserved")
+	}
+}
+
+func TestMigrateOAuthModelAlias_NonexistentFile(t *testing.T) {
+	t.Parallel()
+
+	migrated, err := MigrateOAuthModelAlias("/nonexistent/path/config.yaml")
+	if err != nil {
+		t.Fatalf("unexpected error for nonexistent file: %v", err)
+	}
+	if migrated {
+		t.Fatal("expected no migration for nonexistent file")
+	}
+}
+
+func TestMigrateOAuthModelAlias_EmptyFile(t *testing.T) {
+	t.Parallel()
+
+	dir := t.TempDir()
+	configFile := filepath.Join(dir, "config.yaml")
+
+	if err := os.WriteFile(configFile, []byte(""), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	migrated, err := MigrateOAuthModelAlias(configFile)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if migrated {
+		t.Fatal("expected no migration for empty file")
+	}
+}

internal/config/oauth_model_alias_test.go

@@ -0,0 +1,56 @@
+package config
+
+import "testing"
+
+func TestSanitizeOAuthModelAlias_PreservesForkFlag(t *testing.T) {
+	cfg := &Config{
+		OAuthModelAlias: map[string][]OAuthModelAlias{
+			" CoDeX ": {
+				{Name: " gpt-5 ", Alias: " g5 ", Fork: true},
+				{Name: "gpt-6", Alias: "g6"},
+			},
+		},
+	}
+
+	cfg.SanitizeOAuthModelAlias()
+
+	aliases := cfg.OAuthModelAlias["codex"]
+	if len(aliases) != 2 {
+		t.Fatalf("expected 2 sanitized aliases, got %d", len(aliases))
+	}
+	if aliases[0].Name != "gpt-5" || aliases[0].Alias != "g5" || !aliases[0].Fork {
+		t.Fatalf("expected first alias to be gpt-5->g5 fork=true, got name=%q alias=%q fork=%v", aliases[0].Name, aliases[0].Alias, aliases[0].Fork)
+	}
+	if aliases[1].Name != "gpt-6" || aliases[1].Alias != "g6" || aliases[1].Fork {
+		t.Fatalf("expected second alias to be gpt-6->g6 fork=false, got name=%q alias=%q fork=%v", aliases[1].Name, aliases[1].Alias, aliases[1].Fork)
+	}
+}
+
+func TestSanitizeOAuthModelAlias_AllowsMultipleAliasesForSameName(t *testing.T) {
+	cfg := &Config{
+		OAuthModelAlias: map[string][]OAuthModelAlias{
+			"antigravity": {
+				{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5-20251101", Fork: true},
+				{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5-20251101-thinking", Fork: true},
+				{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5", Fork: true},
+			},
+		},
+	}
+
+	cfg.SanitizeOAuthModelAlias()
+
+	aliases := cfg.OAuthModelAlias["antigravity"]
+	expected := []OAuthModelAlias{
+		{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5-20251101", Fork: true},
+		{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5-20251101-thinking", Fork: true},
+		{Name: "gemini-claude-opus-4-5-thinking", Alias: "claude-opus-4-5", Fork: true},
+	}
+	if len(aliases) != len(expected) {
+		t.Fatalf("expected %d sanitized aliases, got %d", len(expected), len(aliases))
+	}
+	for i, exp := range expected {
+		if aliases[i].Name != exp.Name || aliases[i].Alias != exp.Alias || aliases[i].Fork != exp.Fork {
+			t.Fatalf("expected alias %d to be name=%q alias=%q fork=%v, got name=%q alias=%q fork=%v", i, exp.Name, exp.Alias, exp.Fork, aliases[i].Name, aliases[i].Alias, aliases[i].Fork)
+		}
+	}
+}

internal/config/sdk_config.go

@@ -0,0 +1,106 @@
+// Package config provides configuration management for the CLI Proxy API server.
+// It handles loading and parsing YAML configuration files, and provides structured
+// access to application settings including server port, authentication directory,
+// debug settings, proxy configuration, and API keys.
+package config
+
+// SDKConfig represents the application's configuration, loaded from a YAML file.
+type SDKConfig struct {
+	// ProxyURL is the URL of an optional proxy server to use for outbound requests.
+	ProxyURL string `yaml:"proxy-url" json:"proxy-url"`
+
+	// ForceModelPrefix requires explicit model prefixes (e.g., "teamA/gemini-3-pro-preview")
+	// to target prefixed credentials. When false, unprefixed model requests may use prefixed
+	// credentials as well.
+	ForceModelPrefix bool `yaml:"force-model-prefix" json:"force-model-prefix"`
+
+	// RequestLog enables or disables detailed request logging functionality.
+	RequestLog bool `yaml:"request-log" json:"request-log"`
+
+	// APIKeys is a list of keys for authenticating clients to this proxy server.
+	APIKeys []string `yaml:"api-keys" json:"api-keys"`
+
+	// Access holds request authentication provider configuration.
+	Access AccessConfig `yaml:"auth,omitempty" json:"auth,omitempty"`
+
+	// Streaming configures server-side streaming behavior (keep-alives and safe bootstrap retries).
+	Streaming StreamingConfig `yaml:"streaming" json:"streaming"`
+
+	// NonStreamKeepAliveInterval controls how often blank lines are emitted for non-streaming responses.
+	// <= 0 disables keep-alives. Value is in seconds.
+	NonStreamKeepAliveInterval int `yaml:"nonstream-keepalive-interval,omitempty" json:"nonstream-keepalive-interval,omitempty"`
+}
+
+// StreamingConfig holds server streaming behavior configuration.
+type StreamingConfig struct {
+	// KeepAliveSeconds controls how often the server emits SSE heartbeats (": keep-alive\n\n").
+	// <= 0 disables keep-alives. Default is 0.
+	KeepAliveSeconds int `yaml:"keepalive-seconds,omitempty" json:"keepalive-seconds,omitempty"`
+
+	// BootstrapRetries controls how many times the server may retry a streaming request before any bytes are sent,
+	// to allow auth rotation / transient recovery.
+	// <= 0 disables bootstrap retries. Default is 0.
+	BootstrapRetries int `yaml:"bootstrap-retries,omitempty" json:"bootstrap-retries,omitempty"`
+}
+
+// AccessConfig groups request authentication providers.
+type AccessConfig struct {
+	// Providers lists configured authentication providers.
+	Providers []AccessProvider `yaml:"providers,omitempty" json:"providers,omitempty"`
+}
+
+// AccessProvider describes a request authentication provider entry.
+type AccessProvider struct {
+	// Name is the instance identifier for the provider.
+	Name string `yaml:"name" json:"name"`
+
+	// Type selects the provider implementation registered via the SDK.
+	Type string `yaml:"type" json:"type"`
+
+	// SDK optionally names a third-party SDK module providing this provider.
+	SDK string `yaml:"sdk,omitempty" json:"sdk,omitempty"`
+
+	// APIKeys lists inline keys for providers that require them.
+	APIKeys []string `yaml:"api-keys,omitempty" json:"api-keys,omitempty"`
+
+	// Config passes provider-specific options to the implementation.
+	Config map[string]any `yaml:"config,omitempty" json:"config,omitempty"`
+}
+
+const (
+	// AccessProviderTypeConfigAPIKey is the built-in provider validating inline API keys.
+	AccessProviderTypeConfigAPIKey = "config-api-key"
+
+	// DefaultAccessProviderName is applied when no provider name is supplied.
+	DefaultAccessProviderName = "config-inline"
+)
+
+// ConfigAPIKeyProvider returns the first inline API key provider if present.
+func (c *SDKConfig) ConfigAPIKeyProvider() *AccessProvider {
+	if c == nil {
+		return nil
+	}
+	for i := range c.Access.Providers {
+		if c.Access.Providers[i].Type == AccessProviderTypeConfigAPIKey {
+			if c.Access.Providers[i].Name == "" {
+				c.Access.Providers[i].Name = DefaultAccessProviderName
+			}
+			return &c.Access.Providers[i]
+		}
+	}
+	return nil
+}
+
+// MakeInlineAPIKeyProvider constructs an inline API key provider configuration.
+// It returns nil when no keys are supplied.
+func MakeInlineAPIKeyProvider(keys []string) *AccessProvider {
+	if len(keys) == 0 {
+		return nil
+	}
+	provider := &AccessProvider{
+		Name:    DefaultAccessProviderName,
+		Type:    AccessProviderTypeConfigAPIKey,
+		APIKeys: append([]string(nil), keys...),
+	}
+	return provider
+}

internal/config/vertex_compat.go

@@ -13,6 +13,13 @@ type VertexCompatKey struct {
 	// Maps to the x-goog-api-key header.
 	APIKey string `yaml:"api-key" json:"api-key"`
 
+	// Priority controls selection preference when multiple credentials match.
+	// Higher values are preferred; defaults to 0.
+	Priority int `yaml:"priority,omitempty" json:"priority,omitempty"`
+
+	// Prefix optionally namespaces model aliases for this credential (e.g., "teamA/vertex-pro").
+	Prefix string `yaml:"prefix,omitempty" json:"prefix,omitempty"`
+
 	// BaseURL is the base URL for the Vertex-compatible API endpoint.
 	// The executor will append "/v1/publishers/google/models/{model}:action" to this.
 	// Example: "https://zenmux.ai/api" becomes "https://zenmux.ai/api/v1/publishers/google/models/..."
@@ -29,6 +36,9 @@ type VertexCompatKey struct {
 	Models []VertexCompatModel `yaml:"models,omitempty" json:"models,omitempty"`
 }
 
+func (k VertexCompatKey) GetAPIKey() string  { return k.APIKey }
+func (k VertexCompatKey) GetBaseURL() string { return k.BaseURL }
+
 // VertexCompatModel represents a model configuration for Vertex compatibility,
 // including the actual model name and its alias for API routing.
 type VertexCompatModel struct {
@@ -39,6 +49,9 @@ type VertexCompatModel struct {
 	Alias string `yaml:"alias" json:"alias"`
 }
 
+func (m VertexCompatModel) GetName() string  { return m.Name }
+func (m VertexCompatModel) GetAlias() string { return m.Alias }
+
 // SanitizeVertexCompatKeys deduplicates and normalizes Vertex-compatible API key credentials.
 func (cfg *Config) SanitizeVertexCompatKeys() {
 	if cfg == nil {
@@ -53,6 +66,7 @@ func (cfg *Config) SanitizeVertexCompatKeys() {
 		if entry.APIKey == "" {
 			continue
 		}
+		entry.Prefix = normalizeModelPrefix(entry.Prefix)
 		entry.BaseURL = strings.TrimSpace(entry.BaseURL)
 		if entry.BaseURL == "" {
 			// BaseURL is required for Vertex API key entries

internal/logging/gin_logger.go

@@ -4,9 +4,11 @@
 package logging
 
 import (
+	"errors"
 	"fmt"
 	"net/http"
 	"runtime/debug"
+	"strings"
 	"time"
 
 	"github.com/gin-gonic/gin"
@@ -14,11 +16,24 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+// aiAPIPrefixes defines path prefixes for AI API requests that should have request ID tracking.
+var aiAPIPrefixes = []string{
+	"/v1/chat/completions",
+	"/v1/completions",
+	"/v1/messages",
+	"/v1/responses",
+	"/v1beta/models/",
+	"/api/provider/",
+}
+
 const skipGinLogKey = "__gin_skip_request_logging__"
 
 // GinLogrusLogger returns a Gin middleware handler that logs HTTP requests and responses
 // using logrus. It captures request details including method, path, status code, latency,
-// client IP, and any error messages, formatting them in a Gin-style log format.
+// client IP, and any error messages. Request ID is only added for AI API requests.
+//
+// Output format (AI API): [2025-12-23 20:14:10] [info ] | a1b2c3d4 | 200 |       23.559s | ...
+// Output format (others): [2025-12-23 20:14:10] [info ] | -------- | 200 |       23.559s | ...
 //
 // Returns:
 //   - gin.HandlerFunc: A middleware handler for request logging
@@ -28,6 +43,15 @@ func GinLogrusLogger() gin.HandlerFunc {
 		path := c.Request.URL.Path
 		raw := util.MaskSensitiveQuery(c.Request.URL.RawQuery)
 
+		// Only generate request ID for AI API paths
+		var requestID string
+		if isAIAPIPath(path) {
+			requestID = GenerateRequestID()
+			SetGinRequestID(c, requestID)
+			ctx := WithRequestID(c.Request.Context(), requestID)
+			c.Request = c.Request.WithContext(ctx)
+		}
+
 		c.Next()
 
 		if shouldSkipGinRequestLogging(c) {
@@ -49,23 +73,38 @@ func GinLogrusLogger() gin.HandlerFunc {
 		clientIP := c.ClientIP()
 		method := c.Request.Method
 		errorMessage := c.Errors.ByType(gin.ErrorTypePrivate).String()
-		timestamp := time.Now().Format("2006/01/02 - 15:04:05")
-		logLine := fmt.Sprintf("[GIN] %s | %3d | %13v | %15s | %-7s \"%s\"", timestamp, statusCode, latency, clientIP, method, path)
+
+		if requestID == "" {
+			requestID = "--------"
+		}
+		logLine := fmt.Sprintf("%3d | %13v | %15s | %-7s \"%s\"", statusCode, latency, clientIP, method, path)
 		if errorMessage != "" {
 			logLine = logLine + " | " + errorMessage
 		}
 
+		entry := log.WithField("request_id", requestID)
+
 		switch {
 		case statusCode >= http.StatusInternalServerError:
-			log.Error(logLine)
+			entry.Error(logLine)
 		case statusCode >= http.StatusBadRequest:
-			log.Warn(logLine)
+			entry.Warn(logLine)
 		default:
-			log.Info(logLine)
+			entry.Info(logLine)
 		}
 	}
 }
 
+// isAIAPIPath checks if the given path is an AI API endpoint that should have request ID tracking.
+func isAIAPIPath(path string) bool {
+	for _, prefix := range aiAPIPrefixes {
+		if strings.HasPrefix(path, prefix) {
+			return true
+		}
+	}
+	return false
+}
+
 // GinLogrusRecovery returns a Gin middleware handler that recovers from panics and logs
 // them using logrus. When a panic occurs, it captures the panic value, stack trace,
 // and request path, then returns a 500 Internal Server Error response to the client.
@@ -74,6 +113,11 @@ func GinLogrusLogger() gin.HandlerFunc {
 //   - gin.HandlerFunc: A middleware handler for panic recovery
 func GinLogrusRecovery() gin.HandlerFunc {
 	return gin.CustomRecovery(func(c *gin.Context, recovered interface{}) {
+		if err, ok := recovered.(error); ok && errors.Is(err, http.ErrAbortHandler) {
+			// Let net/http handle ErrAbortHandler so the connection is aborted without noisy stack logs.
+			panic(http.ErrAbortHandler)
+		}
+
 		log.WithFields(log.Fields{
 			"panic": recovered,
 			"stack": string(debug.Stack()),

internal/logging/gin_logger_test.go

@@ -0,0 +1,60 @@
+package logging
+
+import (
+	"errors"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/gin-gonic/gin"
+)
+
+func TestGinLogrusRecoveryRepanicsErrAbortHandler(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	engine := gin.New()
+	engine.Use(GinLogrusRecovery())
+	engine.GET("/abort", func(c *gin.Context) {
+		panic(http.ErrAbortHandler)
+	})
+
+	req := httptest.NewRequest(http.MethodGet, "/abort", nil)
+	recorder := httptest.NewRecorder()
+
+	defer func() {
+		recovered := recover()
+		if recovered == nil {
+			t.Fatalf("expected panic, got nil")
+		}
+		err, ok := recovered.(error)
+		if !ok {
+			t.Fatalf("expected error panic, got %T", recovered)
+		}
+		if !errors.Is(err, http.ErrAbortHandler) {
+			t.Fatalf("expected ErrAbortHandler, got %v", err)
+		}
+		if err != http.ErrAbortHandler {
+			t.Fatalf("expected exact ErrAbortHandler sentinel, got %v", err)
+		}
+	}()
+
+	engine.ServeHTTP(recorder, req)
+}
+
+func TestGinLogrusRecoveryHandlesRegularPanic(t *testing.T) {
+	gin.SetMode(gin.TestMode)
+
+	engine := gin.New()
+	engine.Use(GinLogrusRecovery())
+	engine.GET("/panic", func(c *gin.Context) {
+		panic("boom")
+	})
+
+	req := httptest.NewRequest(http.MethodGet, "/panic", nil)
+	recorder := httptest.NewRecorder()
+
+	engine.ServeHTTP(recorder, req)
+	if recorder.Code != http.StatusInternalServerError {
+		t.Fatalf("expected 500, got %d", recorder.Code)
+	}
+}

internal/logging/global_logger.go

@@ -10,6 +10,7 @@ import (
 	"sync"
 
 	"github.com/gin-gonic/gin"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 	log "github.com/sirupsen/logrus"
 	"gopkg.in/natefinch/lumberjack.v2"
@@ -24,9 +25,13 @@ var (
 )
 
 // LogFormatter defines a custom log format for logrus.
-// This formatter adds timestamp, level, and source location to each log entry.
+// This formatter adds timestamp, level, request ID, and source location to each log entry.
+// Format: [2025-12-23 20:14:04] [debug] [manager.go:524] | a1b2c3d4 | Use API key sk-9...0RHO for model gpt-5.2
 type LogFormatter struct{}
 
+// logFieldOrder defines the display order for common log fields.
+var logFieldOrder = []string{"provider", "model", "mode", "budget", "level", "original_mode", "original_value", "min", "max", "clamped_to", "error"}
+
 // Format renders a single log entry with custom formatting.
 func (m *LogFormatter) Format(entry *log.Entry) ([]byte, error) {
 	var buffer *bytes.Buffer
@@ -39,11 +44,36 @@ func (m *LogFormatter) Format(entry *log.Entry) ([]byte, error) {
 	timestamp := entry.Time.Format("2006-01-02 15:04:05")
 	message := strings.TrimRight(entry.Message, "\r\n")
 
+	reqID := "--------"
+	if id, ok := entry.Data["request_id"].(string); ok && id != "" {
+		reqID = id
+	}
+
+	level := entry.Level.String()
+	if level == "warning" {
+		level = "warn"
+	}
+	levelStr := fmt.Sprintf("%-5s", level)
+
+	// Build fields string (only print fields in logFieldOrder)
+	var fieldsStr string
+	if len(entry.Data) > 0 {
+		var fields []string
+		for _, k := range logFieldOrder {
+			if v, ok := entry.Data[k]; ok {
+				fields = append(fields, fmt.Sprintf("%s=%v", k, v))
+			}
+		}
+		if len(fields) > 0 {
+			fieldsStr = " " + strings.Join(fields, " ")
+		}
+	}
+
 	var formatted string
 	if entry.Caller != nil {
-		formatted = fmt.Sprintf("[%s] [%s] [%s:%d] %s\n", timestamp, entry.Level, filepath.Base(entry.Caller.File), entry.Caller.Line, message)
+		formatted = fmt.Sprintf("[%s] [%s] [%s] [%s:%d] %s%s\n", timestamp, reqID, levelStr, filepath.Base(entry.Caller.File), entry.Caller.Line, message, fieldsStr)
 	} else {
-		formatted = fmt.Sprintf("[%s] [%s] %s\n", timestamp, entry.Level, message)
+		formatted = fmt.Sprintf("[%s] [%s] [%s] %s%s\n", timestamp, reqID, levelStr, message, fieldsStr)
 	}
 	buffer.WriteString(formatted)
 
@@ -71,40 +101,81 @@ func SetupBaseLogger() {
 	})
 }
 
+// isDirWritable checks if the specified directory exists and is writable by attempting to create and remove a test file.
+func isDirWritable(dir string) bool {
+	info, err := os.Stat(dir)
+	if err != nil || !info.IsDir() {
+		return false
+	}
+
+	testFile := filepath.Join(dir, ".perm_test")
+	f, err := os.Create(testFile)
+	if err != nil {
+		return false
+	}
+
+	defer func() {
+		_ = f.Close()
+		_ = os.Remove(testFile)
+	}()
+	return true
+}
+
+// ResolveLogDirectory determines the directory used for application logs.
+func ResolveLogDirectory(cfg *config.Config) string {
+	logDir := "logs"
+	if base := util.WritablePath(); base != "" {
+		return filepath.Join(base, "logs")
+	}
+	if cfg == nil {
+		return logDir
+	}
+	if !isDirWritable(logDir) {
+		authDir := strings.TrimSpace(cfg.AuthDir)
+		if authDir != "" {
+			logDir = filepath.Join(authDir, "logs")
+		}
+	}
+	return logDir
+}
+
 // ConfigureLogOutput switches the global log destination between rotating files and stdout.
-func ConfigureLogOutput(loggingToFile bool) error {
+// When logsMaxTotalSizeMB > 0, a background cleaner removes the oldest log files in the logs directory
+// until the total size is within the limit.
+func ConfigureLogOutput(cfg *config.Config) error {
 	SetupBaseLogger()
 
 	writerMu.Lock()
 	defer writerMu.Unlock()
 
-	if loggingToFile {
-		logDir := "logs"
-		if base := util.WritablePath(); base != "" {
-			logDir = filepath.Join(base, "logs")
-		}
+	logDir := ResolveLogDirectory(cfg)
+
+	protectedPath := ""
+	if cfg.LoggingToFile {
 		if err := os.MkdirAll(logDir, 0o755); err != nil {
 			return fmt.Errorf("logging: failed to create log directory: %w", err)
 		}
 		if logWriter != nil {
 			_ = logWriter.Close()
 		}
+		protectedPath = filepath.Join(logDir, "main.log")
 		logWriter = &lumberjack.Logger{
-			Filename:   filepath.Join(logDir, "main.log"),
+			Filename:   protectedPath,
 			MaxSize:    10,
 			MaxBackups: 0,
 			MaxAge:     0,
 			Compress:   false,
 		}
 		log.SetOutput(logWriter)
-		return nil
-	}
-
+	} else {
 		if logWriter != nil {
 			_ = logWriter.Close()
 			logWriter = nil
 		}
 		log.SetOutput(os.Stdout)
+	}
+
+	configureLogDirCleanerLocked(logDir, cfg.LogsMaxTotalSizeMB, protectedPath)
 	return nil
 }
 
@@ -112,6 +183,8 @@ func closeLogOutputs() {
 	writerMu.Lock()
 	defer writerMu.Unlock()
 
+	stopLogDirCleanerLocked()
+
 	if logWriter != nil {
 		_ = logWriter.Close()
 		logWriter = nil

internal/logging/log_dir_cleaner.go

@@ -0,0 +1,166 @@
+package logging
+
+import (
+	"context"
+	"os"
+	"path/filepath"
+	"sort"
+	"strings"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+)
+
+const logDirCleanerInterval = time.Minute
+
+var logDirCleanerCancel context.CancelFunc
+
+func configureLogDirCleanerLocked(logDir string, maxTotalSizeMB int, protectedPath string) {
+	stopLogDirCleanerLocked()
+
+	if maxTotalSizeMB <= 0 {
+		return
+	}
+
+	maxBytes := int64(maxTotalSizeMB) * 1024 * 1024
+	if maxBytes <= 0 {
+		return
+	}
+
+	dir := strings.TrimSpace(logDir)
+	if dir == "" {
+		return
+	}
+
+	ctx, cancel := context.WithCancel(context.Background())
+	logDirCleanerCancel = cancel
+	go runLogDirCleaner(ctx, filepath.Clean(dir), maxBytes, strings.TrimSpace(protectedPath))
+}
+
+func stopLogDirCleanerLocked() {
+	if logDirCleanerCancel == nil {
+		return
+	}
+	logDirCleanerCancel()
+	logDirCleanerCancel = nil
+}
+
+func runLogDirCleaner(ctx context.Context, logDir string, maxBytes int64, protectedPath string) {
+	ticker := time.NewTicker(logDirCleanerInterval)
+	defer ticker.Stop()
+
+	cleanOnce := func() {
+		deleted, errClean := enforceLogDirSizeLimit(logDir, maxBytes, protectedPath)
+		if errClean != nil {
+			log.WithError(errClean).Warn("logging: failed to enforce log directory size limit")
+			return
+		}
+		if deleted > 0 {
+			log.Debugf("logging: removed %d old log file(s) to enforce log directory size limit", deleted)
+		}
+	}
+
+	cleanOnce()
+	for {
+		select {
+		case <-ctx.Done():
+			return
+		case <-ticker.C:
+			cleanOnce()
+		}
+	}
+}
+
+func enforceLogDirSizeLimit(logDir string, maxBytes int64, protectedPath string) (int, error) {
+	if maxBytes <= 0 {
+		return 0, nil
+	}
+
+	dir := strings.TrimSpace(logDir)
+	if dir == "" {
+		return 0, nil
+	}
+	dir = filepath.Clean(dir)
+
+	entries, errRead := os.ReadDir(dir)
+	if errRead != nil {
+		if os.IsNotExist(errRead) {
+			return 0, nil
+		}
+		return 0, errRead
+	}
+
+	protected := strings.TrimSpace(protectedPath)
+	if protected != "" {
+		protected = filepath.Clean(protected)
+	}
+
+	type logFile struct {
+		path    string
+		size    int64
+		modTime time.Time
+	}
+
+	var (
+		files []logFile
+		total int64
+	)
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		name := entry.Name()
+		if !isLogFileName(name) {
+			continue
+		}
+		info, errInfo := entry.Info()
+		if errInfo != nil {
+			continue
+		}
+		if !info.Mode().IsRegular() {
+			continue
+		}
+		path := filepath.Join(dir, name)
+		files = append(files, logFile{
+			path:    path,
+			size:    info.Size(),
+			modTime: info.ModTime(),
+		})
+		total += info.Size()
+	}
+
+	if total <= maxBytes {
+		return 0, nil
+	}
+
+	sort.Slice(files, func(i, j int) bool {
+		return files[i].modTime.Before(files[j].modTime)
+	})
+
+	deleted := 0
+	for _, file := range files {
+		if total <= maxBytes {
+			break
+		}
+		if protected != "" && filepath.Clean(file.path) == protected {
+			continue
+		}
+		if errRemove := os.Remove(file.path); errRemove != nil {
+			log.WithError(errRemove).Warnf("logging: failed to remove old log file: %s", filepath.Base(file.path))
+			continue
+		}
+		total -= file.size
+		deleted++
+	}
+
+	return deleted, nil
+}
+
+func isLogFileName(name string) bool {
+	trimmed := strings.TrimSpace(name)
+	if trimmed == "" {
+		return false
+	}
+	lower := strings.ToLower(trimmed)
+	return strings.HasSuffix(lower, ".log") || strings.HasSuffix(lower, ".log.gz")
+}

internal/logging/log_dir_cleaner_test.go

@@ -0,0 +1,70 @@
+package logging
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+	"time"
+)
+
+func TestEnforceLogDirSizeLimitDeletesOldest(t *testing.T) {
+	dir := t.TempDir()
+
+	writeLogFile(t, filepath.Join(dir, "old.log"), 60, time.Unix(1, 0))
+	writeLogFile(t, filepath.Join(dir, "mid.log"), 60, time.Unix(2, 0))
+	protected := filepath.Join(dir, "main.log")
+	writeLogFile(t, protected, 60, time.Unix(3, 0))
+
+	deleted, err := enforceLogDirSizeLimit(dir, 120, protected)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if deleted != 1 {
+		t.Fatalf("expected 1 deleted file, got %d", deleted)
+	}
+
+	if _, err := os.Stat(filepath.Join(dir, "old.log")); !os.IsNotExist(err) {
+		t.Fatalf("expected old.log to be removed, stat error: %v", err)
+	}
+	if _, err := os.Stat(filepath.Join(dir, "mid.log")); err != nil {
+		t.Fatalf("expected mid.log to remain, stat error: %v", err)
+	}
+	if _, err := os.Stat(protected); err != nil {
+		t.Fatalf("expected protected main.log to remain, stat error: %v", err)
+	}
+}
+
+func TestEnforceLogDirSizeLimitSkipsProtected(t *testing.T) {
+	dir := t.TempDir()
+
+	protected := filepath.Join(dir, "main.log")
+	writeLogFile(t, protected, 200, time.Unix(1, 0))
+	writeLogFile(t, filepath.Join(dir, "other.log"), 50, time.Unix(2, 0))
+
+	deleted, err := enforceLogDirSizeLimit(dir, 100, protected)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+	if deleted != 1 {
+		t.Fatalf("expected 1 deleted file, got %d", deleted)
+	}
+
+	if _, err := os.Stat(protected); err != nil {
+		t.Fatalf("expected protected main.log to remain, stat error: %v", err)
+	}
+	if _, err := os.Stat(filepath.Join(dir, "other.log")); !os.IsNotExist(err) {
+		t.Fatalf("expected other.log to be removed, stat error: %v", err)
+	}
+}
+
+func writeLogFile(t *testing.T, path string, size int, modTime time.Time) {
+	t.Helper()
+
+	data := make([]byte, size)
+	if err := os.WriteFile(path, data, 0o644); err != nil {
+		t.Fatalf("write file: %v", err)
+	}
+	if err := os.Chtimes(path, modTime, modTime); err != nil {
+		t.Fatalf("set times: %v", err)
+	}
+}

internal/logging/request_logger.go

@@ -14,6 +14,7 @@ import (
 	"regexp"
 	"sort"
 	"strings"
+	"sync/atomic"
 	"time"
 
 	"github.com/andybalholm/brotli"
@@ -25,6 +26,8 @@ import (
 	"github.com/router-for-me/CLIProxyAPI/v6/internal/util"
 )
 
+var requestLogID atomic.Uint64
+
 // RequestLogger defines the interface for logging HTTP requests and responses.
 // It provides methods for logging both regular and streaming HTTP request/response cycles.
 type RequestLogger interface {
@@ -40,10 +43,13 @@ type RequestLogger interface {
 	//   - response: The raw response data
 	//   - apiRequest: The API request data
 	//   - apiResponse: The API response data
+	//   - requestID: Optional request ID for log file naming
+	//   - requestTimestamp: When the request was received
+	//   - apiResponseTimestamp: When the API response was received
 	//
 	// Returns:
 	//   - error: An error if logging fails, nil otherwise
-	LogRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage) error
+	LogRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, requestID string, requestTimestamp, apiResponseTimestamp time.Time) error
 
 	// LogStreamingRequest initiates logging for a streaming request and returns a writer for chunks.
 	//
@@ -52,11 +58,12 @@ type RequestLogger interface {
 	//   - method: The HTTP method
 	//   - headers: The request headers
 	//   - body: The request body
+	//   - requestID: Optional request ID for log file naming
 	//
 	// Returns:
 	//   - StreamingLogWriter: A writer for streaming response chunks
 	//   - error: An error if logging initialization fails, nil otherwise
-	LogStreamingRequest(url, method string, headers map[string][]string, body []byte) (StreamingLogWriter, error)
+	LogStreamingRequest(url, method string, headers map[string][]string, body []byte, requestID string) (StreamingLogWriter, error)
 
 	// IsEnabled returns whether request logging is currently enabled.
 	//
@@ -104,6 +111,12 @@ type StreamingLogWriter interface {
 	//   - error: An error if writing fails, nil otherwise
 	WriteAPIResponse(apiResponse []byte) error
 
+	// SetFirstChunkTimestamp sets the TTFB timestamp captured when first chunk was received.
+	//
+	// Parameters:
+	//   - timestamp: The time when first response chunk was received
+	SetFirstChunkTimestamp(timestamp time.Time)
+
 	// Close finalizes the log file and cleans up resources.
 	//
 	// Returns:
@@ -119,6 +132,9 @@ type FileRequestLogger struct {
 
 	// logsDir is the directory where log files are stored.
 	logsDir string
+
+	// errorLogsMaxFiles limits the number of error log files retained.
+	errorLogsMaxFiles int
 }
 
 // NewFileRequestLogger creates a new file-based request logger.
@@ -128,10 +144,11 @@ type FileRequestLogger struct {
 //   - logsDir: The directory where log files should be stored (can be relative)
 //   - configDir: The directory of the configuration file; when logsDir is
 //     relative, it will be resolved relative to this directory
+//   - errorLogsMaxFiles: Maximum number of error log files to retain (0 = no cleanup)
 //
 // Returns:
 //   - *FileRequestLogger: A new file-based request logger instance
-func NewFileRequestLogger(enabled bool, logsDir string, configDir string) *FileRequestLogger {
+func NewFileRequestLogger(enabled bool, logsDir string, configDir string, errorLogsMaxFiles int) *FileRequestLogger {
 	// Resolve logsDir relative to the configuration file directory when it's not absolute.
 	if !filepath.IsAbs(logsDir) {
 		// If configDir is provided, resolve logsDir relative to it.
@@ -142,6 +159,7 @@ func NewFileRequestLogger(enabled bool, logsDir string, configDir string) *FileR
 	return &FileRequestLogger{
 		enabled:           enabled,
 		logsDir:           logsDir,
+		errorLogsMaxFiles: errorLogsMaxFiles,
 	}
 }
 
@@ -162,6 +180,11 @@ func (l *FileRequestLogger) SetEnabled(enabled bool) {
 	l.enabled = enabled
 }
 
+// SetErrorLogsMaxFiles updates the maximum number of error log files to retain.
+func (l *FileRequestLogger) SetErrorLogsMaxFiles(maxFiles int) {
+	l.errorLogsMaxFiles = maxFiles
+}
+
 // LogRequest logs a complete non-streaming request/response cycle to a file.
 //
 // Parameters:
@@ -174,20 +197,23 @@ func (l *FileRequestLogger) SetEnabled(enabled bool) {
 //   - response: The raw response data
 //   - apiRequest: The API request data
 //   - apiResponse: The API response data
+//   - requestID: Optional request ID for log file naming
+//   - requestTimestamp: When the request was received
+//   - apiResponseTimestamp: When the API response was received
 //
 // Returns:
 //   - error: An error if logging fails, nil otherwise
-func (l *FileRequestLogger) LogRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage) error {
-	return l.logRequest(url, method, requestHeaders, body, statusCode, responseHeaders, response, apiRequest, apiResponse, apiResponseErrors, false)
+func (l *FileRequestLogger) LogRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, requestID string, requestTimestamp, apiResponseTimestamp time.Time) error {
+	return l.logRequest(url, method, requestHeaders, body, statusCode, responseHeaders, response, apiRequest, apiResponse, apiResponseErrors, false, requestID, requestTimestamp, apiResponseTimestamp)
 }
 
 // LogRequestWithOptions logs a request with optional forced logging behavior.
 // The force flag allows writing error logs even when regular request logging is disabled.
-func (l *FileRequestLogger) LogRequestWithOptions(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, force bool) error {
-	return l.logRequest(url, method, requestHeaders, body, statusCode, responseHeaders, response, apiRequest, apiResponse, apiResponseErrors, force)
+func (l *FileRequestLogger) LogRequestWithOptions(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, force bool, requestID string, requestTimestamp, apiResponseTimestamp time.Time) error {
+	return l.logRequest(url, method, requestHeaders, body, statusCode, responseHeaders, response, apiRequest, apiResponse, apiResponseErrors, force, requestID, requestTimestamp, apiResponseTimestamp)
 }
 
-func (l *FileRequestLogger) logRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, force bool) error {
+func (l *FileRequestLogger) logRequest(url, method string, requestHeaders map[string][]string, body []byte, statusCode int, responseHeaders map[string][]string, response, apiRequest, apiResponse []byte, apiResponseErrors []*interfaces.ErrorMessage, force bool, requestID string, requestTimestamp, apiResponseTimestamp time.Time) error {
 	if !l.enabled && !force {
 		return nil
 	}
@@ -197,26 +223,61 @@ func (l *FileRequestLogger) logRequest(url, method string, requestHeaders map[st
 		return fmt.Errorf("failed to create logs directory: %w", errEnsure)
 	}
 
-	// Generate filename
-	filename := l.generateFilename(url)
+	// Generate filename with request ID
+	filename := l.generateFilename(url, requestID)
 	if force && !l.enabled {
-		filename = l.generateErrorFilename(url)
+		filename = l.generateErrorFilename(url, requestID)
 	}
 	filePath := filepath.Join(l.logsDir, filename)
 
-	// Decompress response if needed
-	decompressedResponse, err := l.decompressResponse(responseHeaders, response)
-	if err != nil {
-		// If decompression fails, log the error but continue with original response
-		decompressedResponse = append(response, []byte(fmt.Sprintf("\n[DECOMPRESSION ERROR: %v]", err))...)
+	requestBodyPath, errTemp := l.writeRequestBodyTempFile(body)
+	if errTemp != nil {
+		log.WithError(errTemp).Warn("failed to create request body temp file, falling back to direct write")
+	}
+	if requestBodyPath != "" {
+		defer func() {
+			if errRemove := os.Remove(requestBodyPath); errRemove != nil {
+				log.WithError(errRemove).Warn("failed to remove request body temp file")
+			}
+		}()
 	}
 
-	// Create log content
-	content := l.formatLogContent(url, method, requestHeaders, body, apiRequest, apiResponse, decompressedResponse, statusCode, responseHeaders, apiResponseErrors)
+	responseToWrite, decompressErr := l.decompressResponse(responseHeaders, response)
+	if decompressErr != nil {
+		// If decompression fails, continue with original response and annotate the log output.
+		responseToWrite = response
+	}
 
-	// Write to file
-	if err = os.WriteFile(filePath, []byte(content), 0644); err != nil {
-		return fmt.Errorf("failed to write log file: %w", err)
+	logFile, errOpen := os.OpenFile(filePath, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, 0644)
+	if errOpen != nil {
+		return fmt.Errorf("failed to create log file: %w", errOpen)
+	}
+
+	writeErr := l.writeNonStreamingLog(
+		logFile,
+		url,
+		method,
+		requestHeaders,
+		body,
+		requestBodyPath,
+		apiRequest,
+		apiResponse,
+		apiResponseErrors,
+		statusCode,
+		responseHeaders,
+		responseToWrite,
+		decompressErr,
+		requestTimestamp,
+		apiResponseTimestamp,
+	)
+	if errClose := logFile.Close(); errClose != nil {
+		log.WithError(errClose).Warn("failed to close request log file")
+		if writeErr == nil {
+			return errClose
+		}
+	}
+	if writeErr != nil {
+		return fmt.Errorf("failed to write log file: %w", writeErr)
 	}
 
 	if force && !l.enabled {
@@ -235,11 +296,12 @@ func (l *FileRequestLogger) logRequest(url, method string, requestHeaders map[st
 //   - method: The HTTP method
 //   - headers: The request headers
 //   - body: The request body
+//   - requestID: Optional request ID for log file naming
 //
 // Returns:
 //   - StreamingLogWriter: A writer for streaming response chunks
 //   - error: An error if logging initialization fails, nil otherwise
-func (l *FileRequestLogger) LogStreamingRequest(url, method string, headers map[string][]string, body []byte) (StreamingLogWriter, error) {
+func (l *FileRequestLogger) LogStreamingRequest(url, method string, headers map[string][]string, body []byte, requestID string) (StreamingLogWriter, error) {
 	if !l.enabled {
 		return &NoOpStreamingLogWriter{}, nil
 	}
@@ -249,30 +311,42 @@ func (l *FileRequestLogger) LogStreamingRequest(url, method string, headers map[
 		return nil, fmt.Errorf("failed to create logs directory: %w", err)
 	}
 
-	// Generate filename
-	filename := l.generateFilename(url)
+	// Generate filename with request ID
+	filename := l.generateFilename(url, requestID)
 	filePath := filepath.Join(l.logsDir, filename)
 
-	// Create and open file
-	file, err := os.Create(filePath)
-	if err != nil {
-		return nil, fmt.Errorf("failed to create log file: %w", err)
+	requestHeaders := make(map[string][]string, len(headers))
+	for key, values := range headers {
+		headerValues := make([]string, len(values))
+		copy(headerValues, values)
+		requestHeaders[key] = headerValues
 	}
 
-	// Write initial request information
-	requestInfo := l.formatRequestInfo(url, method, headers, body)
-	if _, err = file.WriteString(requestInfo); err != nil {
-		_ = file.Close()
-		return nil, fmt.Errorf("failed to write request info: %w", err)
+	requestBodyPath, errTemp := l.writeRequestBodyTempFile(body)
+	if errTemp != nil {
+		return nil, fmt.Errorf("failed to create request body temp file: %w", errTemp)
 	}
 
+	responseBodyFile, errCreate := os.CreateTemp(l.logsDir, "response-body-*.tmp")
+	if errCreate != nil {
+		_ = os.Remove(requestBodyPath)
+		return nil, fmt.Errorf("failed to create response body temp file: %w", errCreate)
+	}
+	responseBodyPath := responseBodyFile.Name()
+
 	// Create streaming writer
 	writer := &FileStreamingLogWriter{
-		file:           file,
+		logFilePath:      filePath,
+		url:              url,
+		method:           method,
+		timestamp:        time.Now(),
+		requestHeaders:   requestHeaders,
+		requestBodyPath:  requestBodyPath,
+		responseBodyPath: responseBodyPath,
+		responseBodyFile: responseBodyFile,
 		chunkChan:        make(chan []byte, 100), // Buffered channel for async writes
 		closeChan:        make(chan struct{}),
 		errorChan:        make(chan error, 1),
-		bufferedChunks: &bytes.Buffer{},
 	}
 
 	// Start async writer goroutine
@@ -282,8 +356,8 @@ func (l *FileRequestLogger) LogStreamingRequest(url, method string, headers map[
 }
 
 // generateErrorFilename creates a filename with an error prefix to differentiate forced error logs.
-func (l *FileRequestLogger) generateErrorFilename(url string) string {
-	return fmt.Sprintf("error-%s", l.generateFilename(url))
+func (l *FileRequestLogger) generateErrorFilename(url string, requestID ...string) string {
+	return fmt.Sprintf("error-%s", l.generateFilename(url, requestID...))
 }
 
 // ensureLogsDir creates the logs directory if it doesn't exist.
@@ -298,13 +372,15 @@ func (l *FileRequestLogger) ensureLogsDir() error {
 }
 
 // generateFilename creates a sanitized filename from the URL path and current timestamp.
+// Format: v1-responses-2025-12-23T195811-a1b2c3d4.log
 //
 // Parameters:
 //   - url: The request URL
+//   - requestID: Optional request ID to include in filename
 //
 // Returns:
 //   - string: A sanitized filename for the log file
-func (l *FileRequestLogger) generateFilename(url string) string {
+func (l *FileRequestLogger) generateFilename(url string, requestID ...string) string {
 	// Extract path from URL
 	path := url
 	if strings.Contains(url, "?") {
@@ -320,10 +396,18 @@ func (l *FileRequestLogger) generateFilename(url string) string {
 	sanitized := l.sanitizeForFilename(path)
 
 	// Add timestamp
-	timestamp := time.Now().Format("2006-01-02T150405-.000000000")
-	timestamp = strings.Replace(timestamp, ".", "", -1)
+	timestamp := time.Now().Format("2006-01-02T150405")
 
-	return fmt.Sprintf("%s-%s.log", sanitized, timestamp)
+	// Use request ID if provided, otherwise use sequential ID
+	var idPart string
+	if len(requestID) > 0 && requestID[0] != "" {
+		idPart = requestID[0]
+	} else {
+		id := requestLogID.Add(1)
+		idPart = fmt.Sprintf("%d", id)
+	}
+
+	return fmt.Sprintf("%s-%s-%s.log", sanitized, timestamp, idPart)
 }
 
 // sanitizeForFilename replaces characters that are not safe for filenames.
@@ -359,8 +443,12 @@ func (l *FileRequestLogger) sanitizeForFilename(path string) string {
 	return sanitized
 }
 
-// cleanupOldErrorLogs keeps only the newest 10 forced error log files.
+// cleanupOldErrorLogs keeps only the newest errorLogsMaxFiles forced error log files.
 func (l *FileRequestLogger) cleanupOldErrorLogs() error {
+	if l.errorLogsMaxFiles <= 0 {
+		return nil
+	}
+
 	entries, errRead := os.ReadDir(l.logsDir)
 	if errRead != nil {
 		return errRead
@@ -388,7 +476,7 @@ func (l *FileRequestLogger) cleanupOldErrorLogs() error {
 		files = append(files, logFile{name: name, modTime: info.ModTime()})
 	}
 
-	if len(files) <= 10 {
+	if len(files) <= l.errorLogsMaxFiles {
 		return nil
 	}
 
@@ -396,7 +484,7 @@ func (l *FileRequestLogger) cleanupOldErrorLogs() error {
 		return files[i].modTime.After(files[j].modTime)
 	})
 
-	for _, file := range files[10:] {
+	for _, file := range files[l.errorLogsMaxFiles:] {
 		if errRemove := os.Remove(filepath.Join(l.logsDir, file.name)); errRemove != nil {
 			log.WithError(errRemove).Warnf("failed to remove old error log: %s", file.name)
 		}
@@ -405,6 +493,230 @@ func (l *FileRequestLogger) cleanupOldErrorLogs() error {
 	return nil
 }
 
+func (l *FileRequestLogger) writeRequestBodyTempFile(body []byte) (string, error) {
+	tmpFile, errCreate := os.CreateTemp(l.logsDir, "request-body-*.tmp")
+	if errCreate != nil {
+		return "", errCreate
+	}
+	tmpPath := tmpFile.Name()
+
+	if _, errCopy := io.Copy(tmpFile, bytes.NewReader(body)); errCopy != nil {
+		_ = tmpFile.Close()
+		_ = os.Remove(tmpPath)
+		return "", errCopy
+	}
+	if errClose := tmpFile.Close(); errClose != nil {
+		_ = os.Remove(tmpPath)
+		return "", errClose
+	}
+	return tmpPath, nil
+}
+
+func (l *FileRequestLogger) writeNonStreamingLog(
+	w io.Writer,
+	url, method string,
+	requestHeaders map[string][]string,
+	requestBody []byte,
+	requestBodyPath string,
+	apiRequest []byte,
+	apiResponse []byte,
+	apiResponseErrors []*interfaces.ErrorMessage,
+	statusCode int,
+	responseHeaders map[string][]string,
+	response []byte,
+	decompressErr error,
+	requestTimestamp time.Time,
+	apiResponseTimestamp time.Time,
+) error {
+	if requestTimestamp.IsZero() {
+		requestTimestamp = time.Now()
+	}
+	if errWrite := writeRequestInfoWithBody(w, url, method, requestHeaders, requestBody, requestBodyPath, requestTimestamp); errWrite != nil {
+		return errWrite
+	}
+	if errWrite := writeAPISection(w, "=== API REQUEST ===\n", "=== API REQUEST", apiRequest, time.Time{}); errWrite != nil {
+		return errWrite
+	}
+	if errWrite := writeAPIErrorResponses(w, apiResponseErrors); errWrite != nil {
+		return errWrite
+	}
+	if errWrite := writeAPISection(w, "=== API RESPONSE ===\n", "=== API RESPONSE", apiResponse, apiResponseTimestamp); errWrite != nil {
+		return errWrite
+	}
+	return writeResponseSection(w, statusCode, true, responseHeaders, bytes.NewReader(response), decompressErr, true)
+}
+
+func writeRequestInfoWithBody(
+	w io.Writer,
+	url, method string,
+	headers map[string][]string,
+	body []byte,
+	bodyPath string,
+	timestamp time.Time,
+) error {
+	if _, errWrite := io.WriteString(w, "=== REQUEST INFO ===\n"); errWrite != nil {
+		return errWrite
+	}
+	if _, errWrite := io.WriteString(w, fmt.Sprintf("Version: %s\n", buildinfo.Version)); errWrite != nil {
+		return errWrite
+	}
+	if _, errWrite := io.WriteString(w, fmt.Sprintf("URL: %s\n", url)); errWrite != nil {
+		return errWrite
+	}
+	if _, errWrite := io.WriteString(w, fmt.Sprintf("Method: %s\n", method)); errWrite != nil {
+		return errWrite
+	}
+	if _, errWrite := io.WriteString(w, fmt.Sprintf("Timestamp: %s\n", timestamp.Format(time.RFC3339Nano))); errWrite != nil {
+		return errWrite
+	}
+	if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+		return errWrite
+	}
+
+	if _, errWrite := io.WriteString(w, "=== HEADERS ===\n"); errWrite != nil {
+		return errWrite
+	}
+	for key, values := range headers {
+		for _, value := range values {
+			masked := util.MaskSensitiveHeaderValue(key, value)
+			if _, errWrite := io.WriteString(w, fmt.Sprintf("%s: %s\n", key, masked)); errWrite != nil {
+				return errWrite
+			}
+		}
+	}
+	if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+		return errWrite
+	}
+
+	if _, errWrite := io.WriteString(w, "=== REQUEST BODY ===\n"); errWrite != nil {
+		return errWrite
+	}
+
+	if bodyPath != "" {
+		bodyFile, errOpen := os.Open(bodyPath)
+		if errOpen != nil {
+			return errOpen
+		}
+		if _, errCopy := io.Copy(w, bodyFile); errCopy != nil {
+			_ = bodyFile.Close()
+			return errCopy
+		}
+		if errClose := bodyFile.Close(); errClose != nil {
+			log.WithError(errClose).Warn("failed to close request body temp file")
+		}
+	} else if _, errWrite := w.Write(body); errWrite != nil {
+		return errWrite
+	}
+
+	if _, errWrite := io.WriteString(w, "\n\n"); errWrite != nil {
+		return errWrite
+	}
+	return nil
+}
+
+func writeAPISection(w io.Writer, sectionHeader string, sectionPrefix string, payload []byte, timestamp time.Time) error {
+	if len(payload) == 0 {
+		return nil
+	}
+
+	if bytes.HasPrefix(payload, []byte(sectionPrefix)) {
+		if _, errWrite := w.Write(payload); errWrite != nil {
+			return errWrite
+		}
+		if !bytes.HasSuffix(payload, []byte("\n")) {
+			if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+				return errWrite
+			}
+		}
+	} else {
+		if _, errWrite := io.WriteString(w, sectionHeader); errWrite != nil {
+			return errWrite
+		}
+		if !timestamp.IsZero() {
+			if _, errWrite := io.WriteString(w, fmt.Sprintf("Timestamp: %s\n", timestamp.Format(time.RFC3339Nano))); errWrite != nil {
+				return errWrite
+			}
+		}
+		if _, errWrite := w.Write(payload); errWrite != nil {
+			return errWrite
+		}
+		if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+			return errWrite
+		}
+	}
+
+	if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+		return errWrite
+	}
+	return nil
+}
+
+func writeAPIErrorResponses(w io.Writer, apiResponseErrors []*interfaces.ErrorMessage) error {
+	for i := 0; i < len(apiResponseErrors); i++ {
+		if apiResponseErrors[i] == nil {
+			continue
+		}
+		if _, errWrite := io.WriteString(w, "=== API ERROR RESPONSE ===\n"); errWrite != nil {
+			return errWrite
+		}
+		if _, errWrite := io.WriteString(w, fmt.Sprintf("HTTP Status: %d\n", apiResponseErrors[i].StatusCode)); errWrite != nil {
+			return errWrite
+		}
+		if apiResponseErrors[i].Error != nil {
+			if _, errWrite := io.WriteString(w, apiResponseErrors[i].Error.Error()); errWrite != nil {
+				return errWrite
+			}
+		}
+		if _, errWrite := io.WriteString(w, "\n\n"); errWrite != nil {
+			return errWrite
+		}
+	}
+	return nil
+}
+
+func writeResponseSection(w io.Writer, statusCode int, statusWritten bool, responseHeaders map[string][]string, responseReader io.Reader, decompressErr error, trailingNewline bool) error {
+	if _, errWrite := io.WriteString(w, "=== RESPONSE ===\n"); errWrite != nil {
+		return errWrite
+	}
+	if statusWritten {
+		if _, errWrite := io.WriteString(w, fmt.Sprintf("Status: %d\n", statusCode)); errWrite != nil {
+			return errWrite
+		}
+	}
+
+	if responseHeaders != nil {
+		for key, values := range responseHeaders {
+			for _, value := range values {
+				if _, errWrite := io.WriteString(w, fmt.Sprintf("%s: %s\n", key, value)); errWrite != nil {
+					return errWrite
+				}
+			}
+		}
+	}
+
+	if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+		return errWrite
+	}
+
+	if responseReader != nil {
+		if _, errCopy := io.Copy(w, responseReader); errCopy != nil {
+			return errCopy
+		}
+	}
+	if decompressErr != nil {
+		if _, errWrite := io.WriteString(w, fmt.Sprintf("\n[DECOMPRESSION ERROR: %v]", decompressErr)); errWrite != nil {
+			return errWrite
+		}
+	}
+
+	if trailingNewline {
+		if _, errWrite := io.WriteString(w, "\n"); errWrite != nil {
+			return errWrite
+		}
+	}
+	return nil
+}
+
 // formatLogContent creates the complete log content for non-streaming requests.
 //
 // Parameters:
@@ -648,13 +960,34 @@ func (l *FileRequestLogger) formatRequestInfo(url, method string, headers map[st
 }
 
 // FileStreamingLogWriter implements StreamingLogWriter for file-based streaming logs.
-// It handles asynchronous writing of streaming response chunks to a file.
-// All data is buffered and written in the correct order when Close is called.
+// It spools streaming response chunks to a temporary file to avoid retaining large responses in memory.
+// The final log file is assembled when Close is called.
 type FileStreamingLogWriter struct {
-	// file is the file where log data is written.
-	file *os.File
+	// logFilePath is the final log file path.
+	logFilePath string
 
-	// chunkChan is a channel for receiving response chunks to buffer.
+	// url is the request URL (masked upstream in middleware).
+	url string
+
+	// method is the HTTP method.
+	method string
+
+	// timestamp is captured when the streaming log is initialized.
+	timestamp time.Time
+
+	// requestHeaders stores the request headers.
+	requestHeaders map[string][]string
+
+	// requestBodyPath is a temporary file path holding the request body.
+	requestBodyPath string
+
+	// responseBodyPath is a temporary file path holding the streaming response body.
+	responseBodyPath string
+
+	// responseBodyFile is the temp file where chunks are appended by the async writer.
+	responseBodyFile *os.File
+
+	// chunkChan is a channel for receiving response chunks to spool.
 	chunkChan chan []byte
 
 	// closeChan is a channel for signaling when the writer is closed.
@@ -663,9 +996,6 @@ type FileStreamingLogWriter struct {
 	// errorChan is a channel for reporting errors during writing.
 	errorChan chan error
 
-	// bufferedChunks stores the response chunks in order.
-	bufferedChunks *bytes.Buffer
-
 	// responseStatus stores the HTTP status code.
 	responseStatus int
 
@@ -680,6 +1010,9 @@ type FileStreamingLogWriter struct {
 
 	// apiResponse stores the upstream API response data.
 	apiResponse []byte
+
+	// apiResponseTimestamp captures when the API response was received.
+	apiResponseTimestamp time.Time
 }
 
 // WriteChunkAsync writes a response chunk asynchronously (non-blocking).
@@ -759,6 +1092,12 @@ func (w *FileStreamingLogWriter) WriteAPIResponse(apiResponse []byte) error {
 	return nil
 }
 
+func (w *FileStreamingLogWriter) SetFirstChunkTimestamp(timestamp time.Time) {
+	if !timestamp.IsZero() {
+		w.apiResponseTimestamp = timestamp
+	}
+}
+
 // Close finalizes the log file and cleans up resources.
 // It writes all buffered data to the file in the correct order:
 // API REQUEST -> API RESPONSE -> RESPONSE (status, headers, body chunks)
@@ -770,85 +1109,115 @@ func (w *FileStreamingLogWriter) Close() error {
 		close(w.chunkChan)
 	}
 
-	// Wait for async writer to finish buffering chunks
+	// Wait for async writer to finish spooling chunks
 	if w.closeChan != nil {
 		<-w.closeChan
 		w.chunkChan = nil
 	}
 
-	if w.file == nil {
+	select {
+	case errWrite := <-w.errorChan:
+		w.cleanupTempFiles()
+		return errWrite
+	default:
+	}
+
+	if w.logFilePath == "" {
+		w.cleanupTempFiles()
 		return nil
 	}
 
-	// Write all content in the correct order
-	var content strings.Builder
-
-	// 1. Write API REQUEST section
-	if len(w.apiRequest) > 0 {
-		if bytes.HasPrefix(w.apiRequest, []byte("=== API REQUEST")) {
-			content.Write(w.apiRequest)
-			if !bytes.HasSuffix(w.apiRequest, []byte("\n")) {
-				content.WriteString("\n")
-			}
-		} else {
-			content.WriteString("=== API REQUEST ===\n")
-			content.Write(w.apiRequest)
-			content.WriteString("\n")
-		}
-		content.WriteString("\n")
+	logFile, errOpen := os.OpenFile(w.logFilePath, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, 0644)
+	if errOpen != nil {
+		w.cleanupTempFiles()
+		return fmt.Errorf("failed to create log file: %w", errOpen)
 	}
 
-	// 2. Write API RESPONSE section
-	if len(w.apiResponse) > 0 {
-		if bytes.HasPrefix(w.apiResponse, []byte("=== API RESPONSE")) {
-			content.Write(w.apiResponse)
-			if !bytes.HasSuffix(w.apiResponse, []byte("\n")) {
-				content.WriteString("\n")
+	writeErr := w.writeFinalLog(logFile)
+	if errClose := logFile.Close(); errClose != nil {
+		log.WithError(errClose).Warn("failed to close request log file")
+		if writeErr == nil {
+			writeErr = errClose
 		}
-		} else {
-			content.WriteString("=== API RESPONSE ===\n")
-			content.Write(w.apiResponse)
-			content.WriteString("\n")
-		}
-		content.WriteString("\n")
 	}
 
-	// 3. Write RESPONSE section (status, headers, buffered chunks)
-	content.WriteString("=== RESPONSE ===\n")
-	if w.statusWritten {
-		content.WriteString(fmt.Sprintf("Status: %d\n", w.responseStatus))
-	}
-
-	for key, values := range w.responseHeaders {
-		for _, value := range values {
-			content.WriteString(fmt.Sprintf("%s: %s\n", key, value))
-		}
-	}
-	content.WriteString("\n")
-
-	// Write buffered response body chunks
-	if w.bufferedChunks != nil && w.bufferedChunks.Len() > 0 {
-		content.Write(w.bufferedChunks.Bytes())
-	}
-
-	// Write the complete content to file
-	if _, err := w.file.WriteString(content.String()); err != nil {
-		_ = w.file.Close()
-		return err
-	}
-
-	return w.file.Close()
+	w.cleanupTempFiles()
+	return writeErr
 }
 
 // asyncWriter runs in a goroutine to buffer chunks from the channel.
-// It continuously reads chunks from the channel and buffers them for later writing.
+// It continuously reads chunks from the channel and appends them to a temp file for later assembly.
 func (w *FileStreamingLogWriter) asyncWriter() {
 	defer close(w.closeChan)
 
 	for chunk := range w.chunkChan {
-		if w.bufferedChunks != nil {
-			w.bufferedChunks.Write(chunk)
+		if w.responseBodyFile == nil {
+			continue
 		}
+		if _, errWrite := w.responseBodyFile.Write(chunk); errWrite != nil {
+			select {
+			case w.errorChan <- errWrite:
+			default:
+			}
+			if errClose := w.responseBodyFile.Close(); errClose != nil {
+				select {
+				case w.errorChan <- errClose:
+				default:
+				}
+			}
+			w.responseBodyFile = nil
+		}
+	}
+
+	if w.responseBodyFile == nil {
+		return
+	}
+	if errClose := w.responseBodyFile.Close(); errClose != nil {
+		select {
+		case w.errorChan <- errClose:
+		default:
+		}
+	}
+	w.responseBodyFile = nil
+}
+
+func (w *FileStreamingLogWriter) writeFinalLog(logFile *os.File) error {
+	if errWrite := writeRequestInfoWithBody(logFile, w.url, w.method, w.requestHeaders, nil, w.requestBodyPath, w.timestamp); errWrite != nil {
+		return errWrite
+	}
+	if errWrite := writeAPISection(logFile, "=== API REQUEST ===\n", "=== API REQUEST", w.apiRequest, time.Time{}); errWrite != nil {
+		return errWrite
+	}
+	if errWrite := writeAPISection(logFile, "=== API RESPONSE ===\n", "=== API RESPONSE", w.apiResponse, w.apiResponseTimestamp); errWrite != nil {
+		return errWrite
+	}
+
+	responseBodyFile, errOpen := os.Open(w.responseBodyPath)
+	if errOpen != nil {
+		return errOpen
+	}
+	defer func() {
+		if errClose := responseBodyFile.Close(); errClose != nil {
+			log.WithError(errClose).Warn("failed to close response body temp file")
+		}
+	}()
+
+	return writeResponseSection(logFile, w.responseStatus, w.statusWritten, w.responseHeaders, responseBodyFile, nil, false)
+}
+
+func (w *FileStreamingLogWriter) cleanupTempFiles() {
+	if w.requestBodyPath != "" {
+		if errRemove := os.Remove(w.requestBodyPath); errRemove != nil {
+			log.WithError(errRemove).Warn("failed to remove request body temp file")
+		}
+		w.requestBodyPath = ""
+	}
+
+	if w.responseBodyPath != "" {
+		if errRemove := os.Remove(w.responseBodyPath); errRemove != nil {
+			log.WithError(errRemove).Warn("failed to remove response body temp file")
+		}
+		w.responseBodyPath = ""
 	}
 }
 
@@ -896,6 +1265,8 @@ func (w *NoOpStreamingLogWriter) WriteAPIResponse(_ []byte) error {
 	return nil
 }
 
+func (w *NoOpStreamingLogWriter) SetFirstChunkTimestamp(_ time.Time) {}
+
 // Close is a no-op implementation that does nothing and always returns nil.
 //
 // Returns:

internal/logging/requestid.go

@@ -0,0 +1,61 @@
+package logging
+
+import (
+	"context"
+	"crypto/rand"
+	"encoding/hex"
+
+	"github.com/gin-gonic/gin"
+)
+
+// requestIDKey is the context key for storing/retrieving request IDs.
+type requestIDKey struct{}
+
+// ginRequestIDKey is the Gin context key for request IDs.
+const ginRequestIDKey = "__request_id__"
+
+// GenerateRequestID creates a new 8-character hex request ID.
+func GenerateRequestID() string {
+	b := make([]byte, 4)
+	if _, err := rand.Read(b); err != nil {
+		return "00000000"
+	}
+	return hex.EncodeToString(b)
+}
+
+// WithRequestID returns a new context with the request ID attached.
+func WithRequestID(ctx context.Context, requestID string) context.Context {
+	return context.WithValue(ctx, requestIDKey{}, requestID)
+}
+
+// GetRequestID retrieves the request ID from the context.
+// Returns empty string if not found.
+func GetRequestID(ctx context.Context) string {
+	if ctx == nil {
+		return ""
+	}
+	if id, ok := ctx.Value(requestIDKey{}).(string); ok {
+		return id
+	}
+	return ""
+}
+
+// SetGinRequestID stores the request ID in the Gin context.
+func SetGinRequestID(c *gin.Context, requestID string) {
+	if c != nil {
+		c.Set(ginRequestIDKey, requestID)
+	}
+}
+
+// GetGinRequestID retrieves the request ID from the Gin context.
+func GetGinRequestID(c *gin.Context) string {
+	if c == nil {
+		return ""
+	}
+	if id, exists := c.Get(ginRequestIDKey); exists {
+		if s, ok := id.(string); ok {
+			return s
+		}
+	}
+	return ""
+}

internal/managementasset/updater.go

@@ -9,6 +9,7 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"net/url"
 	"os"
 	"path/filepath"
 	"strings"
@@ -23,7 +24,8 @@ import (
 )
 
 const (
-	managementReleaseURL = "https://api.github.com/repos/router-for-me/Cli-Proxy-API-Management-Center/releases/latest"
+	defaultManagementReleaseURL  = "https://api.github.com/repos/router-for-me/Cli-Proxy-API-Management-Center/releases/latest"
+	defaultManagementFallbackURL = "https://cpamc.router-for.me/"
 	managementAssetName          = "management.html"
 	httpUserAgent                = "CLIProxyAPI-management-updater"
 	updateCheckInterval          = 3 * time.Hour
@@ -97,7 +99,7 @@ func runAutoUpdater(ctx context.Context) {
 
 		configPath, _ := schedulerConfigPath.Load().(string)
 		staticDir := StaticDir(configPath)
-		EnsureLatestManagementHTML(ctx, staticDir, cfg.ProxyURL)
+		EnsureLatestManagementHTML(ctx, staticDir, cfg.ProxyURL, cfg.RemoteManagement.PanelGitHubRepository)
 	}
 
 	runOnce()
@@ -181,7 +183,7 @@ func FilePath(configFilePath string) string {
 // EnsureLatestManagementHTML checks the latest management.html asset and updates the local copy when needed.
 // The function is designed to run in a background goroutine and will never panic.
 // It enforces a 3-hour rate limit to avoid frequent checks on config/auth file changes.
-func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL string) {
+func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL string, panelRepository string) {
 	if ctx == nil {
 		ctx = context.Background()
 	}
@@ -197,6 +199,16 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 		return
 	}
 
+	localPath := filepath.Join(staticDir, managementAssetName)
+	localFileMissing := false
+	if _, errStat := os.Stat(localPath); errStat != nil {
+		if errors.Is(errStat, os.ErrNotExist) {
+			localFileMissing = true
+		} else {
+			log.WithError(errStat).Debug("failed to stat local management asset")
+		}
+	}
+
 	// Rate limiting: check only once every 3 hours
 	lastUpdateCheckMu.Lock()
 	now := time.Now()
@@ -209,14 +221,14 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 	lastUpdateCheckTime = now
 	lastUpdateCheckMu.Unlock()
 
-	if err := os.MkdirAll(staticDir, 0o755); err != nil {
-		log.WithError(err).Warn("failed to prepare static directory for management asset")
+	if errMkdirAll := os.MkdirAll(staticDir, 0o755); errMkdirAll != nil {
+		log.WithError(errMkdirAll).Warn("failed to prepare static directory for management asset")
 		return
 	}
 
+	releaseURL := resolveReleaseURL(panelRepository)
 	client := newHTTPClient(proxyURL)
 
-	localPath := filepath.Join(staticDir, managementAssetName)
 	localHash, err := fileSHA256(localPath)
 	if err != nil {
 		if !errors.Is(err, os.ErrNotExist) {
@@ -225,8 +237,15 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 		localHash = ""
 	}
 
-	asset, remoteHash, err := fetchLatestAsset(ctx, client)
+	asset, remoteHash, err := fetchLatestAsset(ctx, client, releaseURL)
 	if err != nil {
+		if localFileMissing {
+			log.WithError(err).Warn("failed to fetch latest management release information, trying fallback page")
+			if ensureFallbackManagementHTML(ctx, client, localPath) {
+				return
+			}
+			return
+		}
 		log.WithError(err).Warn("failed to fetch latest management release information")
 		return
 	}
@@ -238,6 +257,13 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 
 	data, downloadedHash, err := downloadAsset(ctx, client, asset.BrowserDownloadURL)
 	if err != nil {
+		if localFileMissing {
+			log.WithError(err).Warn("failed to download management asset, trying fallback page")
+			if ensureFallbackManagementHTML(ctx, client, localPath) {
+				return
+			}
+			return
+		}
 		log.WithError(err).Warn("failed to download management asset")
 		return
 	}
@@ -254,8 +280,60 @@ func EnsureLatestManagementHTML(ctx context.Context, staticDir string, proxyURL
 	log.Infof("management asset updated successfully (hash=%s)", downloadedHash)
 }
 
-func fetchLatestAsset(ctx context.Context, client *http.Client) (*releaseAsset, string, error) {
-	req, err := http.NewRequestWithContext(ctx, http.MethodGet, managementReleaseURL, nil)
+func ensureFallbackManagementHTML(ctx context.Context, client *http.Client, localPath string) bool {
+	data, downloadedHash, err := downloadAsset(ctx, client, defaultManagementFallbackURL)
+	if err != nil {
+		log.WithError(err).Warn("failed to download fallback management control panel page")
+		return false
+	}
+
+	if err = atomicWriteFile(localPath, data); err != nil {
+		log.WithError(err).Warn("failed to persist fallback management control panel page")
+		return false
+	}
+
+	log.Infof("management asset updated from fallback page successfully (hash=%s)", downloadedHash)
+	return true
+}
+
+func resolveReleaseURL(repo string) string {
+	repo = strings.TrimSpace(repo)
+	if repo == "" {
+		return defaultManagementReleaseURL
+	}
+
+	parsed, err := url.Parse(repo)
+	if err != nil || parsed.Host == "" {
+		return defaultManagementReleaseURL
+	}
+
+	host := strings.ToLower(parsed.Host)
+	parsed.Path = strings.TrimSuffix(parsed.Path, "/")
+
+	if host == "api.github.com" {
+		if !strings.HasSuffix(strings.ToLower(parsed.Path), "/releases/latest") {
+			parsed.Path = parsed.Path + "/releases/latest"
+		}
+		return parsed.String()
+	}
+
+	if host == "github.com" {
+		parts := strings.Split(strings.Trim(parsed.Path, "/"), "/")
+		if len(parts) >= 2 && parts[0] != "" && parts[1] != "" {
+			repoName := strings.TrimSuffix(parts[1], ".git")
+			return fmt.Sprintf("https://api.github.com/repos/%s/%s/releases/latest", parts[0], repoName)
+		}
+	}
+
+	return defaultManagementReleaseURL
+}
+
+func fetchLatestAsset(ctx context.Context, client *http.Client, releaseURL string) (*releaseAsset, string, error) {
+	if strings.TrimSpace(releaseURL) == "" {
+		releaseURL = defaultManagementReleaseURL
+	}
+
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, releaseURL, nil)
 	if err != nil {
 		return nil, "", fmt.Errorf("create release request: %w", err)
 	}

internal/misc/codex_instructions.go

@@ -7,12 +7,93 @@ import (
 	"embed"
 	_ "embed"
 	"strings"
+	"sync/atomic"
+
+	"github.com/tidwall/gjson"
+	"github.com/tidwall/sjson"
 )
 
+// codexInstructionsEnabled controls whether CodexInstructionsForModel returns official instructions.
+// When false (default), CodexInstructionsForModel returns (true, "") immediately.
+// Set via SetCodexInstructionsEnabled from config.
+var codexInstructionsEnabled atomic.Bool
+
+// SetCodexInstructionsEnabled sets whether codex instructions processing is enabled.
+func SetCodexInstructionsEnabled(enabled bool) {
+	codexInstructionsEnabled.Store(enabled)
+}
+
+// GetCodexInstructionsEnabled returns whether codex instructions processing is enabled.
+func GetCodexInstructionsEnabled() bool {
+	return codexInstructionsEnabled.Load()
+}
+
 //go:embed codex_instructions
 var codexInstructionsDir embed.FS
 
-func CodexInstructionsForModel(modelName, systemInstructions string) (bool, string) {
+//go:embed opencode_codex_instructions.txt
+var opencodeCodexInstructions string
+
+const (
+	codexUserAgentKey  = "__cpa_user_agent"
+	userAgentOpenAISDK = "opencode/"
+)
+
+func InjectCodexUserAgent(raw []byte, userAgent string) []byte {
+	if len(raw) == 0 {
+		return raw
+	}
+	trimmed := strings.TrimSpace(userAgent)
+	if trimmed == "" {
+		return raw
+	}
+	updated, err := sjson.SetBytes(raw, codexUserAgentKey, trimmed)
+	if err != nil {
+		return raw
+	}
+	return updated
+}
+
+func ExtractCodexUserAgent(raw []byte) string {
+	if len(raw) == 0 {
+		return ""
+	}
+	return strings.TrimSpace(gjson.GetBytes(raw, codexUserAgentKey).String())
+}
+
+func StripCodexUserAgent(raw []byte) []byte {
+	if len(raw) == 0 {
+		return raw
+	}
+	if !gjson.GetBytes(raw, codexUserAgentKey).Exists() {
+		return raw
+	}
+	updated, err := sjson.DeleteBytes(raw, codexUserAgentKey)
+	if err != nil {
+		return raw
+	}
+	return updated
+}
+
+func codexInstructionsForOpenCode(systemInstructions string) (bool, string) {
+	if opencodeCodexInstructions == "" {
+		return false, ""
+	}
+	if strings.HasPrefix(systemInstructions, opencodeCodexInstructions) {
+		return true, ""
+	}
+	return false, opencodeCodexInstructions
+}
+
+func useOpenCodeInstructions(userAgent string) bool {
+	return strings.Contains(strings.ToLower(userAgent), userAgentOpenAISDK)
+}
+
+func IsOpenCodeUserAgent(userAgent string) bool {
+	return useOpenCodeInstructions(userAgent)
+}
+
+func codexInstructionsForCodex(modelName, systemInstructions string) (bool, string) {
 	entries, _ := codexInstructionsDir.ReadDir("codex_instructions")
 
 	lastPrompt := ""
@@ -20,6 +101,7 @@ func CodexInstructionsForModel(modelName, systemInstructions string) (bool, stri
 	lastCodexMaxPrompt := ""
 	last51Prompt := ""
 	last52Prompt := ""
+	last52CodexPrompt := ""
 	// lastReviewPrompt := ""
 	for _, entry := range entries {
 		content, _ := codexInstructionsDir.ReadFile("codex_instructions/" + entry.Name())
@@ -36,12 +118,16 @@ func CodexInstructionsForModel(modelName, systemInstructions string) (bool, stri
 			last51Prompt = string(content)
 		} else if strings.HasPrefix(entry.Name(), "gpt_5_2_prompt.md") {
 			last52Prompt = string(content)
+		} else if strings.HasPrefix(entry.Name(), "gpt-5.2-codex_prompt.md") {
+			last52CodexPrompt = string(content)
 		} else if strings.HasPrefix(entry.Name(), "review_prompt.md") {
 			// lastReviewPrompt = string(content)
 		}
 	}
 	if strings.Contains(modelName, "codex-max") {
 		return false, lastCodexMaxPrompt
+	} else if strings.Contains(modelName, "5.2-codex") {
+		return false, last52CodexPrompt
 	} else if strings.Contains(modelName, "codex") {
 		return false, lastCodexPrompt
 	} else if strings.Contains(modelName, "5.1") {
@@ -52,3 +138,13 @@ func CodexInstructionsForModel(modelName, systemInstructions string) (bool, stri
 		return false, lastPrompt
 	}
 }
+
+func CodexInstructionsForModel(modelName, systemInstructions, userAgent string) (bool, string) {
+	if !GetCodexInstructionsEnabled() {
+		return true, ""
+	}
+	if IsOpenCodeUserAgent(userAgent) {
+		return codexInstructionsForOpenCode(systemInstructions)
+	}
+	return codexInstructionsForCodex(modelName, systemInstructions)
+}

internal/misc/codex_instructions/gpt-5.2-codex_prompt.md-001-f084e5264b1b0ae9eb8c63c950c0953f40966fed

@@ -0,0 +1,117 @@
+You are Codex, based on GPT-5. You are running as a coding agent in the Codex CLI on a user's computer.
+
+## General
+
+- When searching for text or files, prefer using `rg` or `rg --files` respectively because `rg` is much faster than alternatives like `grep`. (If the `rg` command is not found, then use alternatives.)
+
+## Editing constraints
+
+- Default to ASCII when editing or creating files. Only introduce non-ASCII or other Unicode characters when there is a clear justification and the file already uses them.
+- Add succinct code comments that explain what is going on if code is not self-explanatory. You should not add comments like "Assigns the value to the variable", but a brief comment might be useful ahead of a complex code block that the user would otherwise have to spend time parsing out. Usage of these comments should be rare.
+- Try to use apply_patch for single file edits, but it is fine to explore other options to make the edit if it does not work well. Do not use apply_patch for changes that are auto-generated (i.e. generating package.json or running a lint or format command like gofmt) or when scripting is more efficient (such as search and replacing a string across a codebase).
+- You may be in a dirty git worktree.
+    * NEVER revert existing changes you did not make unless explicitly requested, since these changes were made by the user.
+    * If asked to make a commit or code edits and there are unrelated changes to your work or changes that you didn't make in those files, don't revert those changes.
+    * If the changes are in files you've touched recently, you should read carefully and understand how you can work with the changes rather than reverting them.
+    * If the changes are in unrelated files, just ignore them and don't revert them.
+- Do not amend a commit unless explicitly requested to do so.
+- While you are working, you might notice unexpected changes that you didn't make. If this happens, STOP IMMEDIATELY and ask the user how they would like to proceed.
+- **NEVER** use destructive commands like `git reset --hard` or `git checkout --` unless specifically requested or approved by the user.
+
+## Plan tool
+
+When using the planning tool:
+- Skip using the planning tool for straightforward tasks (roughly the easiest 25%).
+- Do not make single-step plans.
+- When you made a plan, update it after having performed one of the sub-tasks that you shared on the plan.
+
+## Codex CLI harness, sandboxing, and approvals
+
+The Codex CLI harness supports several different configurations for sandboxing and escalation approvals that the user can choose from.
+
+Filesystem sandboxing defines which files can be read or written. The options for `sandbox_mode` are:
+- **read-only**: The sandbox only permits reading files.
+- **workspace-write**: The sandbox permits reading files, and editing files in `cwd` and `writable_roots`. Editing files in other directories requires approval.
+- **danger-full-access**: No filesystem sandboxing - all commands are permitted.
+
+Network sandboxing defines whether network can be accessed without approval. Options for `network_access` are:
+- **restricted**: Requires approval
+- **enabled**: No approval needed
+
+Approvals are your mechanism to get user consent to run shell commands without the sandbox. Possible configuration options for `approval_policy` are
+- **untrusted**: The harness will escalate most commands for user approval, apart from a limited allowlist of safe "read" commands.
+- **on-failure**: The harness will allow all commands to run in the sandbox (if enabled), and failures will be escalated to the user for approval to run again without the sandbox.
+- **on-request**: Commands will be run in the sandbox by default, and you can specify in your tool call if you want to escalate a command to run without sandboxing. (Note that this mode is not always available. If it is, you'll see parameters for it in the `shell` command description.)
+- **never**: This is a non-interactive mode where you may NEVER ask the user for approval to run commands. Instead, you must always persist and work around constraints to solve the task for the user. You MUST do your utmost best to finish the task and validate your work before yielding. If this mode is paired with `danger-full-access`, take advantage of it to deliver the best outcome for the user. Further, in this mode, your default testing philosophy is overridden: Even if you don't see local patterns for testing, you may add tests and scripts to validate your work. Just remove them before yielding.
+
+When you are running with `approval_policy == on-request`, and sandboxing enabled, here are scenarios where you'll need to request approval:
+- You need to run a command that writes to a directory that requires it (e.g. running tests that write to /var)
+- You need to run a GUI app (e.g., open/xdg-open/osascript) to open browsers or files.
+- You are running sandboxed and need to run a command that requires network access (e.g. installing packages)
+- If you run a command that is important to solving the user's query, but it fails because of sandboxing, rerun the command with approval. ALWAYS proceed to use the `sandbox_permissions` and `justification` parameters - do not message the user before requesting approval for the command.
+- You are about to take a potentially destructive action such as an `rm` or `git reset` that the user did not explicitly ask for
+- (for all of these, you should weigh alternative paths that do not require approval)
+
+When `sandbox_mode` is set to read-only, you'll need to request approval for any command that isn't a read.
+
+You will be told what filesystem sandboxing, network sandboxing, and approval mode are active in a developer or user message. If you are not told about this, assume that you are running with workspace-write, network sandboxing enabled, and approval on-failure.
+
+Although they introduce friction to the user because your work is paused until the user responds, you should leverage them when necessary to accomplish important work. If the completing the task requires escalated permissions, Do not let these settings or the sandbox deter you from attempting to accomplish the user's task unless it is set to "never", in which case never ask for approvals.
+
+When requesting approval to execute a command that will require escalated privileges:
+  - Provide the `sandbox_permissions` parameter with the value `"require_escalated"`
+  - Include a short, 1 sentence explanation for why you need escalated permissions in the justification parameter
+
+## Special user requests
+
+- If the user makes a simple request (such as asking for the time) which you can fulfill by running a terminal command (such as `date`), you should do so.
+- If the user asks for a "review", default to a code review mindset: prioritise identifying bugs, risks, behavioural regressions, and missing tests. Findings must be the primary focus of the response - keep summaries or overviews brief and only after enumerating the issues. Present findings first (ordered by severity with file/line references), follow with open questions or assumptions, and offer a change-summary only as a secondary detail. If no findings are discovered, state that explicitly and mention any residual risks or testing gaps.
+
+## Frontend tasks
+When doing frontend design tasks, avoid collapsing into "AI slop" or safe, average-looking layouts.
+Aim for interfaces that feel intentional, bold, and a bit surprising.
+- Typography: Use expressive, purposeful fonts and avoid default stacks (Inter, Roboto, Arial, system).
+- Color & Look: Choose a clear visual direction; define CSS variables; avoid purple-on-white defaults. No purple bias or dark mode bias.
+- Motion: Use a few meaningful animations (page-load, staggered reveals) instead of generic micro-motions.
+- Background: Don't rely on flat, single-color backgrounds; use gradients, shapes, or subtle patterns to build atmosphere.
+- Overall: Avoid boilerplate layouts and interchangeable UI patterns. Vary themes, type families, and visual languages across outputs.
+- Ensure the page loads properly on both desktop and mobile
+
+Exception: If working within an existing website or design system, preserve the established patterns, structure, and visual language.
+
+## Presenting your work and final message
+
+You are producing plain text that will later be styled by the CLI. Follow these rules exactly. Formatting should make results easy to scan, but not feel mechanical. Use judgment to decide how much structure adds value.
+
+- Default: be very concise; friendly coding teammate tone.
+- Ask only when needed; suggest ideas; mirror the user's style.
+- For substantial work, summarize clearly; follow final‑answer formatting.
+- Skip heavy formatting for simple confirmations.
+- Don't dump large files you've written; reference paths only.
+- No "save/copy this file" - User is on the same machine.
+- Offer logical next steps (tests, commits, build) briefly; add verify steps if you couldn't do something.
+- For code changes:
+  * Lead with a quick explanation of the change, and then give more details on the context covering where and why a change was made. Do not start this explanation with "summary", just jump right in.
+  * If there are natural next steps the user may want to take, suggest them at the end of your response. Do not make suggestions if there are no natural next steps.
+  * When suggesting multiple options, use numeric lists for the suggestions so the user can quickly respond with a single number.
+- The user does not command execution outputs. When asked to show the output of a command (e.g. `git show`), relay the important details in your answer or summarize the key lines so the user understands the result.
+
+### Final answer structure and style guidelines
+
+- Plain text; CLI handles styling. Use structure only when it helps scanability.
+- Headers: optional; short Title Case (1-3 words) wrapped in **…**; no blank line before the first bullet; add only if they truly help.
+- Bullets: use - ; merge related points; keep to one line when possible; 4–6 per list ordered by importance; keep phrasing consistent.
+- Monospace: backticks for commands/paths/env vars/code ids and inline examples; use for literal keyword bullets; never combine with **.
+- Code samples or multi-line snippets should be wrapped in fenced code blocks; include an info string as often as possible.
+- Structure: group related bullets; order sections general → specific → supporting; for subsections, start with a bolded keyword bullet, then items; match complexity to the task.
+- Tone: collaborative, concise, factual; present tense, active voice; self‑contained; no "above/below"; parallel wording.
+- Don'ts: no nested bullets/hierarchies; no ANSI codes; don't cram unrelated keywords; keep keyword lists short—wrap/reformat if long; avoid naming formatting styles in answers.
+- Adaptation: code explanations → precise, structured with code refs; simple tasks → lead with outcome; big changes → logical walkthrough + rationale + next actions; casual one-offs → plain sentences, no headers/bullets.
+- File References: When referencing files in your response follow the below rules:
+  * Use inline code to make file paths clickable.
+  * Each reference should have a stand alone path. Even if it's the same file.
+  * Accepted: absolute, workspace‑relative, a/ or b/ diff prefixes, or bare filename/suffix.
+  * Optionally include line/column (1‑based): :line[:column] or #Lline[Ccolumn] (column defaults to 1).
+  * Do not use URIs like file://, vscode://, or https://.
+  * Do not provide range of lines
+  * Examples: src/app.ts, src/app.ts:42, b/server/index.js#L10, C:\repo\project\main.rs:12:5

internal/misc/oauth.go

@@ -4,6 +4,8 @@ import (
 	"crypto/rand"
 	"encoding/hex"
 	"fmt"
+	"net/url"
+	"strings"
 )
 
 // GenerateRandomState generates a cryptographically secure random state parameter
@@ -19,3 +21,83 @@ func GenerateRandomState() (string, error) {
 	}
 	return hex.EncodeToString(bytes), nil
 }
+
+// OAuthCallback captures the parsed OAuth callback parameters.
+type OAuthCallback struct {
+	Code             string
+	State            string
+	Error            string
+	ErrorDescription string
+}
+
+// ParseOAuthCallback extracts OAuth parameters from a callback URL.
+// It returns nil when the input is empty.
+func ParseOAuthCallback(input string) (*OAuthCallback, error) {
+	trimmed := strings.TrimSpace(input)
+	if trimmed == "" {
+		return nil, nil
+	}
+
+	candidate := trimmed
+	if !strings.Contains(candidate, "://") {
+		if strings.HasPrefix(candidate, "?") {
+			candidate = "http://localhost" + candidate
+		} else if strings.ContainsAny(candidate, "/?#") || strings.Contains(candidate, ":") {
+			candidate = "http://" + candidate
+		} else if strings.Contains(candidate, "=") {
+			candidate = "http://localhost/?" + candidate
+		} else {
+			return nil, fmt.Errorf("invalid callback URL")
+		}
+	}
+
+	parsedURL, err := url.Parse(candidate)
+	if err != nil {
+		return nil, err
+	}
+
+	query := parsedURL.Query()
+	code := strings.TrimSpace(query.Get("code"))
+	state := strings.TrimSpace(query.Get("state"))
+	errCode := strings.TrimSpace(query.Get("error"))
+	errDesc := strings.TrimSpace(query.Get("error_description"))
+
+	if parsedURL.Fragment != "" {
+		if fragQuery, errFrag := url.ParseQuery(parsedURL.Fragment); errFrag == nil {
+			if code == "" {
+				code = strings.TrimSpace(fragQuery.Get("code"))
+			}
+			if state == "" {
+				state = strings.TrimSpace(fragQuery.Get("state"))
+			}
+			if errCode == "" {
+				errCode = strings.TrimSpace(fragQuery.Get("error"))
+			}
+			if errDesc == "" {
+				errDesc = strings.TrimSpace(fragQuery.Get("error_description"))
+			}
+		}
+	}
+
+	if code != "" && state == "" && strings.Contains(code, "#") {
+		parts := strings.SplitN(code, "#", 2)
+		code = parts[0]
+		state = parts[1]
+	}
+
+	if errCode == "" && errDesc != "" {
+		errCode = errDesc
+		errDesc = ""
+	}
+
+	if code == "" && errCode == "" {
+		return nil, fmt.Errorf("callback URL missing code")
+	}
+
+	return &OAuthCallback{
+		Code:             code,
+		State:            state,
+		Error:            errCode,
+		ErrorDescription: errDesc,
+	}, nil
+}

internal/misc/opencode_codex_instructions.txt

@@ -0,0 +1,79 @@
+You are OpenCode, the best coding agent on the planet.
+
+You are an interactive CLI tool that helps users with software engineering tasks. Use the instructions below and the tools available to you to assist the user.
+
+## Editing constraints
+- Default to ASCII when editing or creating files. Only introduce non-ASCII or other Unicode characters when there is a clear justification and the file already uses them.
+- Only add comments if they are necessary to make a non-obvious block easier to understand.
+- Try to use apply_patch for single file edits, but it is fine to explore other options to make the edit if it does not work well. Do not use apply_patch for changes that are auto-generated (i.e. generating package.json or running a lint or format command like gofmt) or when scripting is more efficient (such as search and replacing a string across a codebase).
+
+## Tool usage
+- Prefer specialized tools over shell for file operations:
+  - Use Read to view files, Edit to modify files, and Write only when needed.
+  - Use Glob to find files by name and Grep to search file contents.
+- Use Bash for terminal operations (git, bun, builds, tests, running scripts).
+- Run tool calls in parallel when neither call needs the other’s output; otherwise run sequentially.
+
+## Git and workspace hygiene
+- You may be in a dirty git worktree.
+    * NEVER revert existing changes you did not make unless explicitly requested, since these changes were made by the user.
+    * If asked to make a commit or code edits and there are unrelated changes to your work or changes that you didn't make in those files, don't revert those changes.
+    * If the changes are in files you've touched recently, you should read carefully and understand how you can work with the changes rather than reverting them.
+    * If the changes are in unrelated files, just ignore them and don't revert them.
+- Do not amend commits unless explicitly requested.
+- **NEVER** use destructive commands like `git reset --hard` or `git checkout --` unless specifically requested or approved by the user.
+
+## Frontend tasks
+When doing frontend design tasks, avoid collapsing into bland, generic layouts.
+Aim for interfaces that feel intentional and deliberate.
+- Typography: Use expressive, purposeful fonts and avoid default stacks (Inter, Roboto, Arial, system).
+- Color & Look: Choose a clear visual direction; define CSS variables; avoid purple-on-white defaults. No purple bias or dark mode bias.
+- Motion: Use a few meaningful animations (page-load, staggered reveals) instead of generic micro-motions.
+- Background: Don't rely on flat, single-color backgrounds; use gradients, shapes, or subtle patterns to build atmosphere.
+- Overall: Avoid boilerplate layouts and interchangeable UI patterns. Vary themes, type families, and visual languages across outputs.
+- Ensure the page loads properly on both desktop and mobile.
+
+Exception: If working within an existing website or design system, preserve the established patterns, structure, and visual language.
+
+## Presenting your work and final message
+
+You are producing plain text that will later be styled by the CLI. Follow these rules exactly. Formatting should make results easy to scan, but not feel mechanical. Use judgment to decide how much structure adds value.
+
+- Default: be very concise; friendly coding teammate tone.
+- Default: do the work without asking questions. Treat short tasks as sufficient direction; infer missing details by reading the codebase and following existing conventions.
+- Questions: only ask when you are truly blocked after checking relevant context AND you cannot safely pick a reasonable default. This usually means one of:
+  * The request is ambiguous in a way that materially changes the result and you cannot disambiguate by reading the repo.
+  * The action is destructive/irreversible, touches production, or changes billing/security posture.
+  * You need a secret/credential/value that cannot be inferred (API key, account id, etc.).
+- If you must ask: do all non-blocked work first, then ask exactly one targeted question, include your recommended default, and state what would change based on the answer.
+- Never ask permission questions like "Should I proceed?" or "Do you want me to run tests?"; proceed with the most reasonable option and mention what you did.
+- For substantial work, summarize clearly; follow final‑answer formatting.
+- Skip heavy formatting for simple confirmations.
+- Don't dump large files you've written; reference paths only.
+- No "save/copy this file" - User is on the same machine.
+- Offer logical next steps (tests, commits, build) briefly; add verify steps if you couldn't do something.
+- For code changes:
+  * Lead with a quick explanation of the change, and then give more details on the context covering where and why a change was made. Do not start this explanation with "summary", just jump right in.
+  * If there are natural next steps the user may want to take, suggest them at the end of your response. Do not make suggestions if there are no natural next steps.
+  * When suggesting multiple options, use numeric lists for the suggestions so the user can quickly respond with a single number.
+- The user does not command execution outputs. When asked to show the output of a command (e.g. `git show`), relay the important details in your answer or summarize the key lines so the user understands the result.
+
+## Final answer structure and style guidelines
+
+- Plain text; CLI handles styling. Use structure only when it helps scanability.
+- Headers: optional; short Title Case (1-3 words) wrapped in **…**; no blank line before the first bullet; add only if they truly help.
+- Bullets: use - ; merge related points; keep to one line when possible; 4–6 per list ordered by importance; keep phrasing consistent.
+- Monospace: backticks for commands/paths/env vars/code ids and inline examples; use for literal keyword bullets; never combine with **.
+- Code samples or multi-line snippets should be wrapped in fenced code blocks; include an info string as often as possible.
+- Structure: group related bullets; order sections general → specific → supporting; for subsections, start with a bolded keyword bullet, then items; match complexity to the task.
+- Tone: collaborative, concise, factual; present tense, active voice; self‑contained; no "above/below"; parallel wording.
+- Don'ts: no nested bullets/hierarchies; no ANSI codes; don't cram unrelated keywords; keep keyword lists short—wrap/reformat if long; avoid naming formatting styles in answers.
+- Adaptation: code explanations → precise, structured with code refs; simple tasks → lead with outcome; big changes → logical walkthrough + rationale + next actions; casual one-offs → plain sentences, no headers/bullets.
+- File References: When referencing files in your response follow the below rules:
+  * Use inline code to make file paths clickable.
+  * Each reference should have a stand alone path. Even if it's the same file.
+  * Accepted: absolute, workspace‑relative, a/ or b/ diff prefixes, or bare filename/suffix.
+  * Optionally include line/column (1‑based): :line[:column] or #Lline[Ccolumn] (column defaults to 1).
+  * Do not use URIs like file://, vscode://, or https://.
+  * Do not provide range of lines
+  * Examples: src/app.ts, src/app.ts:42, b/server/index.js#L10, C:\repo\project\main.rs:12:5

internal/registry/model_definitions.go

@@ -1,698 +1,105 @@
-// Package registry provides model definitions for various AI service providers.
-// This file contains static model definitions that can be used by clients
-// when registering their supported models.
+// Package registry provides model definitions and lookup helpers for various AI providers.
+// Static model metadata is stored in model_definitions_static_data.go.
 package registry
 
-// GetClaudeModels returns the standard Claude model definitions
-func GetClaudeModels() []*ModelInfo {
-	return []*ModelInfo{
+import (
+	"sort"
+	"strings"
+)
 
-		{
-			ID:                  "claude-haiku-4-5-20251001",
-			Object:              "model",
-			Created:             1759276800, // 2025-10-01
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4.5 Haiku",
-			ContextLength:       200000,
-			MaxCompletionTokens: 64000,
-			// Thinking: not supported for Haiku models
-		},
-		{
-			ID:                  "claude-sonnet-4-5-20250929",
-			Object:              "model",
-			Created:             1759104000, // 2025-09-29
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4.5 Sonnet",
-			ContextLength:       200000,
-			MaxCompletionTokens: 64000,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-opus-4-5-20251101",
-			Object:              "model",
-			Created:             1761955200, // 2025-11-01
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4.5 Opus",
-			Description:         "Premium model combining maximum intelligence with practical performance",
-			ContextLength:       200000,
-			MaxCompletionTokens: 64000,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-opus-4-1-20250805",
-			Object:              "model",
-			Created:             1722945600, // 2025-08-05
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4.1 Opus",
-			ContextLength:       200000,
-			MaxCompletionTokens: 32000,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-opus-4-20250514",
-			Object:              "model",
-			Created:             1715644800, // 2025-05-14
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4 Opus",
-			ContextLength:       200000,
-			MaxCompletionTokens: 32000,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-sonnet-4-20250514",
-			Object:              "model",
-			Created:             1715644800, // 2025-05-14
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 4 Sonnet",
-			ContextLength:       200000,
-			MaxCompletionTokens: 64000,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-3-7-sonnet-20250219",
-			Object:              "model",
-			Created:             1708300800, // 2025-02-19
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 3.7 Sonnet",
-			ContextLength:       128000,
-			MaxCompletionTokens: 8192,
-			Thinking:            &ThinkingSupport{Min: 1024, Max: 100000, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                  "claude-3-5-haiku-20241022",
-			Object:              "model",
-			Created:             1729555200, // 2024-10-22
-			OwnedBy:             "anthropic",
-			Type:                "claude",
-			DisplayName:         "Claude 3.5 Haiku",
-			ContextLength:       128000,
-			MaxCompletionTokens: 8192,
-			// Thinking: not supported for Haiku models
-		},
+// GetStaticModelDefinitionsByChannel returns static model definitions for a given channel/provider.
+// It returns nil when the channel is unknown.
+//
+// Supported channels:
+//   - claude
+//   - gemini
+//   - vertex
+//   - gemini-cli
+//   - aistudio
+//   - codex
+//   - qwen
+//   - iflow
+//   - antigravity (returns static overrides only)
+func GetStaticModelDefinitionsByChannel(channel string) []*ModelInfo {
+	key := strings.ToLower(strings.TrimSpace(channel))
+	switch key {
+	case "claude":
+		return GetClaudeModels()
+	case "gemini":
+		return GetGeminiModels()
+	case "vertex":
+		return GetGeminiVertexModels()
+	case "gemini-cli":
+		return GetGeminiCLIModels()
+	case "aistudio":
+		return GetAIStudioModels()
+	case "codex":
+		return GetOpenAIModels()
+	case "qwen":
+		return GetQwenModels()
+	case "iflow":
+		return GetIFlowModels()
+	case "antigravity":
+		cfg := GetAntigravityModelConfig()
+		if len(cfg) == 0 {
+			return nil
 		}
+		models := make([]*ModelInfo, 0, len(cfg))
+		for modelID, entry := range cfg {
+			if modelID == "" || entry == nil {
+				continue
 			}
-
-// GetGeminiModels returns the standard Gemini model definitions
-func GetGeminiModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-pro",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-pro",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Pro",
-			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-lite",
-			Object:                     "model",
-			Created:                    1753142400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-lite",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Lite",
-			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Preview",
-			Description:                "Gemini 3 Pro Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-image-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-image-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Image Preview",
-			Description:                "Gemini 3 Pro Image Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-	}
-}
-
-func GetGeminiVertexModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-pro",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-pro",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Pro",
-			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-lite",
-			Object:                     "model",
-			Created:                    1753142400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-lite",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Lite",
-			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Preview",
-			Description:                "Gemini 3 Pro Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-image-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-image-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Image Preview",
-			Description:                "Gemini 3 Pro Image Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-	}
-}
-
-// GetGeminiCLIModels returns the standard Gemini model definitions
-func GetGeminiCLIModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-pro",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-pro",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Pro",
-			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-lite",
-			Object:                     "model",
-			Created:                    1753142400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-lite",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Lite",
-			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Preview",
-			Description:                "Gemini 3 Pro Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-	}
-}
-
-// GetAIStudioModels returns the Gemini model definitions for AI Studio integrations
-func GetAIStudioModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                         "gemini-2.5-pro",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-pro",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Pro",
-			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash",
-			Version:                    "001",
-			DisplayName:                "Gemini 2.5 Flash",
-			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-lite",
-			Object:                     "model",
-			Created:                    1753142400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-lite",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Lite",
-			Description:                "Our smallest and most cost effective model, built for at scale usage.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-3-pro-preview",
-			Object:                     "model",
-			Created:                    1737158400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-3-pro-preview",
-			Version:                    "3.0",
-			DisplayName:                "Gemini 3 Pro Preview",
-			Description:                "Gemini 3 Pro Preview",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-pro-latest",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-pro-latest",
-			Version:                    "2.5",
-			DisplayName:                "Gemini Pro Latest",
-			Description:                "Latest release of Gemini Pro",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-flash-latest",
-			Object:                     "model",
-			Created:                    1750118400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-flash-latest",
-			Version:                    "2.5",
-			DisplayName:                "Gemini Flash Latest",
-			Description:                "Latest release of Gemini Flash",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-flash-lite-latest",
-			Object:                     "model",
-			Created:                    1753142400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-flash-lite-latest",
-			Version:                    "2.5",
-			DisplayName:                "Gemini Flash-Lite Latest",
-			Description:                "Latest release of Gemini Flash-Lite",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           65536,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			Thinking:                   &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
-		},
-		{
-			ID:                         "gemini-2.5-flash-image-preview",
-			Object:                     "model",
-			Created:                    1756166400,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-image-preview",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Image Preview",
-			Description:                "State-of-the-art image generation and editing model.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           8192,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			// image models don't support thinkingConfig; leave Thinking nil
-		},
-		{
-			ID:                         "gemini-2.5-flash-image",
-			Object:                     "model",
-			Created:                    1759363200,
-			OwnedBy:                    "google",
-			Type:                       "gemini",
-			Name:                       "models/gemini-2.5-flash-image",
-			Version:                    "2.5",
-			DisplayName:                "Gemini 2.5 Flash Image",
-			Description:                "State-of-the-art image generation and editing model.",
-			InputTokenLimit:            1048576,
-			OutputTokenLimit:           8192,
-			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
-			// image models don't support thinkingConfig; leave Thinking nil
-		},
-	}
-}
-
-// GetOpenAIModels returns the standard OpenAI model definitions
-func GetOpenAIModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                  "gpt-5",
-			Object:              "model",
-			Created:             1754524800,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5-2025-08-07",
-			DisplayName:         "GPT 5",
-			Description:         "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"minimal", "low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5-codex",
-			Object:              "model",
-			Created:             1757894400,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5-2025-09-15",
-			DisplayName:         "GPT 5 Codex",
-			Description:         "Stable version of GPT 5 Codex, The best model for coding and agentic tasks across domains.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5-codex-mini",
-			Object:              "model",
-			Created:             1762473600,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5-2025-11-07",
-			DisplayName:         "GPT 5 Codex Mini",
-			Description:         "Stable version of GPT 5 Codex Mini: cheaper, faster, but less capable version of GPT 5 Codex.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5.1",
-			Object:              "model",
-			Created:             1762905600,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5.1-2025-11-12",
-			DisplayName:         "GPT 5",
-			Description:         "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"none", "low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5.1-codex",
-			Object:              "model",
-			Created:             1762905600,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5.1-2025-11-12",
-			DisplayName:         "GPT 5.1 Codex",
-			Description:         "Stable version of GPT 5.1 Codex, The best model for coding and agentic tasks across domains.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5.1-codex-mini",
-			Object:              "model",
-			Created:             1762905600,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5.1-2025-11-12",
-			DisplayName:         "GPT 5.1 Codex Mini",
-			Description:         "Stable version of GPT 5.1 Codex Mini: cheaper, faster, but less capable version of GPT 5.1 Codex.",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
-		},
-		{
-			ID:                  "gpt-5.1-codex-max",
-			Object:              "model",
-			Created:             1763424000,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5.1-max",
-			DisplayName:         "GPT 5.1 Codex Max",
-			Description:         "Stable version of GPT 5.1 Codex Max",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
-		},
-		{
-			ID:                  "gpt-5.2",
-			Object:              "model",
-			Created:             1765440000,
-			OwnedBy:             "openai",
-			Type:                "openai",
-			Version:             "gpt-5.2",
-			DisplayName:         "GPT 5.2",
-			Description:         "Stable version of GPT 5.2",
-			ContextLength:       400000,
-			MaxCompletionTokens: 128000,
-			SupportedParameters: []string{"tools"},
-			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
-		},
-	}
-}
-
-// GetQwenModels returns the standard Qwen model definitions
-func GetQwenModels() []*ModelInfo {
-	return []*ModelInfo{
-		{
-			ID:                  "qwen3-coder-plus",
-			Object:              "model",
-			Created:             1753228800,
-			OwnedBy:             "qwen",
-			Type:                "qwen",
-			Version:             "3.0",
-			DisplayName:         "Qwen3 Coder Plus",
-			Description:         "Advanced code generation and understanding model",
-			ContextLength:       32768,
-			MaxCompletionTokens: 8192,
-			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
-		},
-		{
-			ID:                  "qwen3-coder-flash",
-			Object:              "model",
-			Created:             1753228800,
-			OwnedBy:             "qwen",
-			Type:                "qwen",
-			Version:             "3.0",
-			DisplayName:         "Qwen3 Coder Flash",
-			Description:         "Fast code generation model",
-			ContextLength:       8192,
-			MaxCompletionTokens: 2048,
-			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
-		},
-		{
-			ID:                  "vision-model",
-			Object:              "model",
-			Created:             1758672000,
-			OwnedBy:             "qwen",
-			Type:                "qwen",
-			Version:             "3.0",
-			DisplayName:         "Qwen3 Vision Model",
-			Description:         "Vision model model",
-			ContextLength:       32768,
-			MaxCompletionTokens: 2048,
-			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
-		},
-	}
-}
-
-// GetIFlowModels returns supported models for iFlow OAuth accounts.
-func GetIFlowModels() []*ModelInfo {
-	entries := []struct {
-		ID          string
-		DisplayName string
-		Description string
-		Created     int64
-		Thinking    *ThinkingSupport
-	}{
-		{ID: "tstars2.0", DisplayName: "TStars-2.0", Description: "iFlow TStars-2.0 multimodal assistant", Created: 1746489600},
-		{ID: "qwen3-coder-plus", DisplayName: "Qwen3-Coder-Plus", Description: "Qwen3 Coder Plus code generation", Created: 1753228800},
-		{ID: "qwen3-max", DisplayName: "Qwen3-Max", Description: "Qwen3 flagship model", Created: 1758672000},
-		{ID: "qwen3-vl-plus", DisplayName: "Qwen3-VL-Plus", Description: "Qwen3 multimodal vision-language", Created: 1758672000},
-		{ID: "qwen3-max-preview", DisplayName: "Qwen3-Max-Preview", Description: "Qwen3 Max preview build", Created: 1757030400},
-		{ID: "kimi-k2-0905", DisplayName: "Kimi-K2-Instruct-0905", Description: "Moonshot Kimi K2 instruct 0905", Created: 1757030400},
-		{ID: "glm-4.6", DisplayName: "GLM-4.6", Description: "Zhipu GLM 4.6 general model", Created: 1759190400},
-		{ID: "kimi-k2", DisplayName: "Kimi-K2", Description: "Moonshot Kimi K2 general model", Created: 1752192000},
-		{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 thinking model", Created: 1762387200, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
-		{ID: "deepseek-v3.2-chat", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2", Created: 1764576000},
-		{ID: "deepseek-v3.2", DisplayName: "DeepSeek-V3.2-Exp", Description: "DeepSeek V3.2 experimental", Created: 1759104000},
-		{ID: "deepseek-v3.1", DisplayName: "DeepSeek-V3.1-Terminus", Description: "DeepSeek V3.1 Terminus", Created: 1756339200},
-		{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
-		{ID: "deepseek-v3", DisplayName: "DeepSeek-V3-671B", Description: "DeepSeek V3 671B", Created: 1734307200},
-		{ID: "qwen3-32b", DisplayName: "Qwen3-32B", Description: "Qwen3 32B", Created: 1747094400},
-		{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
-		{ID: "qwen3-235b-a22b-instruct", DisplayName: "Qwen3-235B-A22B-Instruct", Description: "Qwen3 235B A22B Instruct", Created: 1753401600},
-		{ID: "qwen3-235b", DisplayName: "Qwen3-235B-A22B", Description: "Qwen3 235B A22B", Created: 1753401600},
-		{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000, Thinking: &ThinkingSupport{Levels: []string{"low", "medium", "high"}}},
-	}
-	models := make([]*ModelInfo, 0, len(entries))
-	for _, entry := range entries {
 			models = append(models, &ModelInfo{
-			ID:          entry.ID,
+				ID:                  modelID,
 				Object:              "model",
-			Created:     entry.Created,
-			OwnedBy:     "iflow",
-			Type:        "iflow",
-			DisplayName: entry.DisplayName,
-			Description: entry.Description,
+				OwnedBy:             "antigravity",
+				Type:                "antigravity",
 				Thinking:            entry.Thinking,
+				MaxCompletionTokens: entry.MaxCompletionTokens,
 			})
 		}
+		sort.Slice(models, func(i, j int) bool {
+			return strings.ToLower(models[i].ID) < strings.ToLower(models[j].ID)
+		})
 		return models
+	default:
+		return nil
+	}
 }
 
-// AntigravityModelConfig captures static antigravity model overrides, including
-// Thinking budget limits and provider max completion tokens.
-type AntigravityModelConfig struct {
-	Thinking            *ThinkingSupport
-	MaxCompletionTokens int
-	Name                string
+// LookupStaticModelInfo searches all static model definitions for a model by ID.
+// Returns nil if no matching model is found.
+func LookupStaticModelInfo(modelID string) *ModelInfo {
+	if modelID == "" {
+		return nil
 	}
 
-// GetAntigravityModelConfig returns static configuration for antigravity models.
-// Keys use the ALIASED model names (after modelName2Alias conversion) for direct lookup.
-func GetAntigravityModelConfig() map[string]*AntigravityModelConfig {
-	return map[string]*AntigravityModelConfig{
-		"gemini-2.5-flash":                        {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}, Name: "models/gemini-2.5-flash"},
-		"gemini-2.5-flash-lite":                   {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}, Name: "models/gemini-2.5-flash-lite"},
-		"gemini-2.5-computer-use-preview-10-2025": {Name: "models/gemini-2.5-computer-use-preview-10-2025"},
-		"gemini-3-pro-preview":                    {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}, Name: "models/gemini-3-pro-preview"},
-		"gemini-3-pro-image-preview":              {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}, Name: "models/gemini-3-pro-image-preview"},
-		"gemini-claude-sonnet-4-5-thinking":       {Thinking: &ThinkingSupport{Min: 1024, Max: 200000, ZeroAllowed: false, DynamicAllowed: true}, MaxCompletionTokens: 64000},
-		"gemini-claude-opus-4-5-thinking":         {Thinking: &ThinkingSupport{Min: 1024, Max: 200000, ZeroAllowed: false, DynamicAllowed: true}, MaxCompletionTokens: 64000},
+	allModels := [][]*ModelInfo{
+		GetClaudeModels(),
+		GetGeminiModels(),
+		GetGeminiVertexModels(),
+		GetGeminiCLIModels(),
+		GetAIStudioModels(),
+		GetOpenAIModels(),
+		GetQwenModels(),
+		GetIFlowModels(),
+	}
+	for _, models := range allModels {
+		for _, m := range models {
+			if m != nil && m.ID == modelID {
+				return m
 			}
 		}
+	}
+
+	// Check Antigravity static config
+	if cfg := GetAntigravityModelConfig()[modelID]; cfg != nil {
+		return &ModelInfo{
+			ID:                  modelID,
+			Thinking:            cfg.Thinking,
+			MaxCompletionTokens: cfg.MaxCompletionTokens,
+		}
+	}
+
+	return nil
+}

internal/registry/model_definitions_static_data.go

@@ -0,0 +1,846 @@
+// Package registry provides model definitions for various AI service providers.
+// This file stores the static model metadata catalog.
+package registry
+
+// GetClaudeModels returns the standard Claude model definitions
+func GetClaudeModels() []*ModelInfo {
+	return []*ModelInfo{
+
+		{
+			ID:                  "claude-haiku-4-5-20251001",
+			Object:              "model",
+			Created:             1759276800, // 2025-10-01
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4.5 Haiku",
+			ContextLength:       200000,
+			MaxCompletionTokens: 64000,
+			// Thinking: not supported for Haiku models
+		},
+		{
+			ID:                  "claude-sonnet-4-5-20250929",
+			Object:              "model",
+			Created:             1759104000, // 2025-09-29
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4.5 Sonnet",
+			ContextLength:       200000,
+			MaxCompletionTokens: 64000,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-opus-4-5-20251101",
+			Object:              "model",
+			Created:             1761955200, // 2025-11-01
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4.5 Opus",
+			Description:         "Premium model combining maximum intelligence with practical performance",
+			ContextLength:       200000,
+			MaxCompletionTokens: 64000,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-opus-4-1-20250805",
+			Object:              "model",
+			Created:             1722945600, // 2025-08-05
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4.1 Opus",
+			ContextLength:       200000,
+			MaxCompletionTokens: 32000,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-opus-4-20250514",
+			Object:              "model",
+			Created:             1715644800, // 2025-05-14
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4 Opus",
+			ContextLength:       200000,
+			MaxCompletionTokens: 32000,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-sonnet-4-20250514",
+			Object:              "model",
+			Created:             1715644800, // 2025-05-14
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 4 Sonnet",
+			ContextLength:       200000,
+			MaxCompletionTokens: 64000,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-3-7-sonnet-20250219",
+			Object:              "model",
+			Created:             1708300800, // 2025-02-19
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 3.7 Sonnet",
+			ContextLength:       128000,
+			MaxCompletionTokens: 8192,
+			Thinking:            &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: false, DynamicAllowed: false},
+		},
+		{
+			ID:                  "claude-3-5-haiku-20241022",
+			Object:              "model",
+			Created:             1729555200, // 2024-10-22
+			OwnedBy:             "anthropic",
+			Type:                "claude",
+			DisplayName:         "Claude 3.5 Haiku",
+			ContextLength:       128000,
+			MaxCompletionTokens: 8192,
+			// Thinking: not supported for Haiku models
+		},
+	}
+}
+
+// GetGeminiModels returns the standard Gemini model definitions
+func GetGeminiModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Gemini 3 Pro Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
+		},
+		{
+			ID:                         "gemini-3-flash-preview",
+			Object:                     "model",
+			Created:                    1765929600,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-flash-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Flash Preview",
+			Description:                "Gemini 3 Flash Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
+		},
+		{
+			ID:                         "gemini-3-pro-image-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-image-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Image Preview",
+			Description:                "Gemini 3 Pro Image Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
+		},
+	}
+}
+
+func GetGeminiVertexModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Gemini 3 Pro Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
+		},
+		{
+			ID:                         "gemini-3-flash-preview",
+			Object:                     "model",
+			Created:                    1765929600,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-flash-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Flash Preview",
+			Description:                "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
+		},
+		{
+			ID:                         "gemini-3-pro-image-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-image-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Image Preview",
+			Description:                "Gemini 3 Pro Image Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
+		},
+		// Imagen image generation models - use :predict action
+		{
+			ID:                         "imagen-4.0-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Generate",
+			Description:                "Imagen 4.0 image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-4.0-ultra-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-ultra-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Ultra Generate",
+			Description:                "Imagen 4.0 Ultra high-quality image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-3.0-generate-002",
+			Object:                     "model",
+			Created:                    1740000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-3.0-generate-002",
+			Version:                    "3.0",
+			DisplayName:                "Imagen 3.0 Generate",
+			Description:                "Imagen 3.0 image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-3.0-fast-generate-001",
+			Object:                     "model",
+			Created:                    1740000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-3.0-fast-generate-001",
+			Version:                    "3.0",
+			DisplayName:                "Imagen 3.0 Fast Generate",
+			Description:                "Imagen 3.0 fast image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+		{
+			ID:                         "imagen-4.0-fast-generate-001",
+			Object:                     "model",
+			Created:                    1750000000,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/imagen-4.0-fast-generate-001",
+			Version:                    "4.0",
+			DisplayName:                "Imagen 4.0 Fast Generate",
+			Description:                "Imagen 4.0 fast image generation model",
+			SupportedGenerationMethods: []string{"predict"},
+		},
+	}
+}
+
+// GetGeminiCLIModels returns the standard Gemini model definitions
+func GetGeminiCLIModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Our most intelligent model with SOTA reasoning and multimodal understanding, and powerful agentic and vibe coding capabilities",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}},
+		},
+		{
+			ID:                         "gemini-3-flash-preview",
+			Object:                     "model",
+			Created:                    1765929600,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-flash-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Flash Preview",
+			Description:                "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}},
+		},
+	}
+}
+
+// GetAIStudioModels returns the Gemini model definitions for AI Studio integrations
+func GetAIStudioModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                         "gemini-2.5-pro",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-pro",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Pro",
+			Description:                "Stable release (June 17th, 2025) of Gemini 2.5 Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash",
+			Version:                    "001",
+			DisplayName:                "Gemini 2.5 Flash",
+			Description:                "Stable version of Gemini 2.5 Flash, our mid-size multimodal model that supports up to 1 million tokens, released in June of 2025.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-2.5-flash-lite",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-lite",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Lite",
+			Description:                "Our smallest and most cost effective model, built for at scale usage.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-pro-preview",
+			Object:                     "model",
+			Created:                    1737158400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-pro-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Pro Preview",
+			Description:                "Gemini 3 Pro Preview",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-3-flash-preview",
+			Object:                     "model",
+			Created:                    1765929600,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-3-flash-preview",
+			Version:                    "3.0",
+			DisplayName:                "Gemini 3 Flash Preview",
+			Description:                "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-pro-latest",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-pro-latest",
+			Version:                    "2.5",
+			DisplayName:                "Gemini Pro Latest",
+			Description:                "Latest release of Gemini Pro",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-flash-latest",
+			Object:                     "model",
+			Created:                    1750118400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-flash-latest",
+			Version:                    "2.5",
+			DisplayName:                "Gemini Flash Latest",
+			Description:                "Latest release of Gemini Flash",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		{
+			ID:                         "gemini-flash-lite-latest",
+			Object:                     "model",
+			Created:                    1753142400,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-flash-lite-latest",
+			Version:                    "2.5",
+			DisplayName:                "Gemini Flash-Lite Latest",
+			Description:                "Latest release of Gemini Flash-Lite",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           65536,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			Thinking:                   &ThinkingSupport{Min: 512, Max: 24576, ZeroAllowed: true, DynamicAllowed: true},
+		},
+		// {
+		// 	ID:                         "gemini-2.5-flash-image-preview",
+		// 	Object:                     "model",
+		// 	Created:                    1756166400,
+		// 	OwnedBy:                    "google",
+		// 	Type:                       "gemini",
+		// 	Name:                       "models/gemini-2.5-flash-image-preview",
+		// 	Version:                    "2.5",
+		// 	DisplayName:                "Gemini 2.5 Flash Image Preview",
+		// 	Description:                "State-of-the-art image generation and editing model.",
+		// 	InputTokenLimit:            1048576,
+		// 	OutputTokenLimit:           8192,
+		// 	SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+		// 	// image models don't support thinkingConfig; leave Thinking nil
+		// },
+		{
+			ID:                         "gemini-2.5-flash-image",
+			Object:                     "model",
+			Created:                    1759363200,
+			OwnedBy:                    "google",
+			Type:                       "gemini",
+			Name:                       "models/gemini-2.5-flash-image",
+			Version:                    "2.5",
+			DisplayName:                "Gemini 2.5 Flash Image",
+			Description:                "State-of-the-art image generation and editing model.",
+			InputTokenLimit:            1048576,
+			OutputTokenLimit:           8192,
+			SupportedGenerationMethods: []string{"generateContent", "countTokens", "createCachedContent", "batchGenerateContent"},
+			// image models don't support thinkingConfig; leave Thinking nil
+		},
+	}
+}
+
+// GetOpenAIModels returns the standard OpenAI model definitions
+func GetOpenAIModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                  "gpt-5",
+			Object:              "model",
+			Created:             1754524800,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5-2025-08-07",
+			DisplayName:         "GPT 5",
+			Description:         "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"minimal", "low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5-codex",
+			Object:              "model",
+			Created:             1757894400,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5-2025-09-15",
+			DisplayName:         "GPT 5 Codex",
+			Description:         "Stable version of GPT 5 Codex, The best model for coding and agentic tasks across domains.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5-codex-mini",
+			Object:              "model",
+			Created:             1762473600,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5-2025-11-07",
+			DisplayName:         "GPT 5 Codex Mini",
+			Description:         "Stable version of GPT 5 Codex Mini: cheaper, faster, but less capable version of GPT 5 Codex.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5.1",
+			Object:              "model",
+			Created:             1762905600,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.1-2025-11-12",
+			DisplayName:         "GPT 5",
+			Description:         "Stable version of GPT 5, The best model for coding and agentic tasks across domains.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"none", "low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5.1-codex",
+			Object:              "model",
+			Created:             1762905600,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.1-2025-11-12",
+			DisplayName:         "GPT 5.1 Codex",
+			Description:         "Stable version of GPT 5.1 Codex, The best model for coding and agentic tasks across domains.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5.1-codex-mini",
+			Object:              "model",
+			Created:             1762905600,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.1-2025-11-12",
+			DisplayName:         "GPT 5.1 Codex Mini",
+			Description:         "Stable version of GPT 5.1 Codex Mini: cheaper, faster, but less capable version of GPT 5.1 Codex.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high"}},
+		},
+		{
+			ID:                  "gpt-5.1-codex-max",
+			Object:              "model",
+			Created:             1763424000,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.1-max",
+			DisplayName:         "GPT 5.1 Codex Max",
+			Description:         "Stable version of GPT 5.1 Codex Max",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
+		},
+		{
+			ID:                  "gpt-5.2",
+			Object:              "model",
+			Created:             1765440000,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.2",
+			DisplayName:         "GPT 5.2",
+			Description:         "Stable version of GPT 5.2",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"none", "low", "medium", "high", "xhigh"}},
+		},
+		{
+			ID:                  "gpt-5.2-codex",
+			Object:              "model",
+			Created:             1765440000,
+			OwnedBy:             "openai",
+			Type:                "openai",
+			Version:             "gpt-5.2",
+			DisplayName:         "GPT 5.2 Codex",
+			Description:         "Stable version of GPT 5.2 Codex, The best model for coding and agentic tasks across domains.",
+			ContextLength:       400000,
+			MaxCompletionTokens: 128000,
+			SupportedParameters: []string{"tools"},
+			Thinking:            &ThinkingSupport{Levels: []string{"low", "medium", "high", "xhigh"}},
+		},
+	}
+}
+
+// GetQwenModels returns the standard Qwen model definitions
+func GetQwenModels() []*ModelInfo {
+	return []*ModelInfo{
+		{
+			ID:                  "qwen3-coder-plus",
+			Object:              "model",
+			Created:             1753228800,
+			OwnedBy:             "qwen",
+			Type:                "qwen",
+			Version:             "3.0",
+			DisplayName:         "Qwen3 Coder Plus",
+			Description:         "Advanced code generation and understanding model",
+			ContextLength:       32768,
+			MaxCompletionTokens: 8192,
+			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
+		},
+		{
+			ID:                  "qwen3-coder-flash",
+			Object:              "model",
+			Created:             1753228800,
+			OwnedBy:             "qwen",
+			Type:                "qwen",
+			Version:             "3.0",
+			DisplayName:         "Qwen3 Coder Flash",
+			Description:         "Fast code generation model",
+			ContextLength:       8192,
+			MaxCompletionTokens: 2048,
+			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
+		},
+		{
+			ID:                  "vision-model",
+			Object:              "model",
+			Created:             1758672000,
+			OwnedBy:             "qwen",
+			Type:                "qwen",
+			Version:             "3.0",
+			DisplayName:         "Qwen3 Vision Model",
+			Description:         "Vision model model",
+			ContextLength:       32768,
+			MaxCompletionTokens: 2048,
+			SupportedParameters: []string{"temperature", "top_p", "max_tokens", "stream", "stop"},
+		},
+	}
+}
+
+// iFlowThinkingSupport is a shared ThinkingSupport configuration for iFlow models
+// that support thinking mode via chat_template_kwargs.enable_thinking (boolean toggle).
+// Uses level-based configuration so standard normalization flows apply before conversion.
+var iFlowThinkingSupport = &ThinkingSupport{
+	Levels: []string{"none", "auto", "minimal", "low", "medium", "high", "xhigh"},
+}
+
+// GetIFlowModels returns supported models for iFlow OAuth accounts.
+func GetIFlowModels() []*ModelInfo {
+	entries := []struct {
+		ID          string
+		DisplayName string
+		Description string
+		Created     int64
+		Thinking    *ThinkingSupport
+	}{
+		{ID: "tstars2.0", DisplayName: "TStars-2.0", Description: "iFlow TStars-2.0 multimodal assistant", Created: 1746489600},
+		{ID: "qwen3-coder-plus", DisplayName: "Qwen3-Coder-Plus", Description: "Qwen3 Coder Plus code generation", Created: 1753228800},
+		{ID: "qwen3-max", DisplayName: "Qwen3-Max", Description: "Qwen3 flagship model", Created: 1758672000},
+		{ID: "qwen3-vl-plus", DisplayName: "Qwen3-VL-Plus", Description: "Qwen3 multimodal vision-language", Created: 1758672000},
+		{ID: "qwen3-max-preview", DisplayName: "Qwen3-Max-Preview", Description: "Qwen3 Max preview build", Created: 1757030400, Thinking: iFlowThinkingSupport},
+		{ID: "kimi-k2-0905", DisplayName: "Kimi-K2-Instruct-0905", Description: "Moonshot Kimi K2 instruct 0905", Created: 1757030400},
+		{ID: "glm-4.6", DisplayName: "GLM-4.6", Description: "Zhipu GLM 4.6 general model", Created: 1759190400, Thinking: iFlowThinkingSupport},
+		{ID: "glm-4.7", DisplayName: "GLM-4.7", Description: "Zhipu GLM 4.7 general model", Created: 1766448000, Thinking: iFlowThinkingSupport},
+		{ID: "kimi-k2", DisplayName: "Kimi-K2", Description: "Moonshot Kimi K2 general model", Created: 1752192000},
+		{ID: "kimi-k2-thinking", DisplayName: "Kimi-K2-Thinking", Description: "Moonshot Kimi K2 thinking model", Created: 1762387200},
+		{ID: "deepseek-v3.2-chat", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Chat", Created: 1764576000},
+		{ID: "deepseek-v3.2-reasoner", DisplayName: "DeepSeek-V3.2", Description: "DeepSeek V3.2 Reasoner", Created: 1764576000},
+		{ID: "deepseek-v3.2", DisplayName: "DeepSeek-V3.2-Exp", Description: "DeepSeek V3.2 experimental", Created: 1759104000, Thinking: iFlowThinkingSupport},
+		{ID: "deepseek-v3.1", DisplayName: "DeepSeek-V3.1-Terminus", Description: "DeepSeek V3.1 Terminus", Created: 1756339200, Thinking: iFlowThinkingSupport},
+		{ID: "deepseek-r1", DisplayName: "DeepSeek-R1", Description: "DeepSeek reasoning model R1", Created: 1737331200},
+		{ID: "deepseek-v3", DisplayName: "DeepSeek-V3-671B", Description: "DeepSeek V3 671B", Created: 1734307200},
+		{ID: "qwen3-32b", DisplayName: "Qwen3-32B", Description: "Qwen3 32B", Created: 1747094400},
+		{ID: "qwen3-235b-a22b-thinking-2507", DisplayName: "Qwen3-235B-A22B-Thinking", Description: "Qwen3 235B A22B Thinking (2507)", Created: 1753401600},
+		{ID: "qwen3-235b-a22b-instruct", DisplayName: "Qwen3-235B-A22B-Instruct", Description: "Qwen3 235B A22B Instruct", Created: 1753401600},
+		{ID: "qwen3-235b", DisplayName: "Qwen3-235B-A22B", Description: "Qwen3 235B A22B", Created: 1753401600},
+		{ID: "minimax-m2", DisplayName: "MiniMax-M2", Description: "MiniMax M2", Created: 1758672000, Thinking: iFlowThinkingSupport},
+		{ID: "minimax-m2.1", DisplayName: "MiniMax-M2.1", Description: "MiniMax M2.1", Created: 1766448000, Thinking: iFlowThinkingSupport},
+		{ID: "iflow-rome-30ba3b", DisplayName: "iFlow-ROME", Description: "iFlow Rome 30BA3B model", Created: 1736899200},
+	}
+	models := make([]*ModelInfo, 0, len(entries))
+	for _, entry := range entries {
+		models = append(models, &ModelInfo{
+			ID:          entry.ID,
+			Object:      "model",
+			Created:     entry.Created,
+			OwnedBy:     "iflow",
+			Type:        "iflow",
+			DisplayName: entry.DisplayName,
+			Description: entry.Description,
+			Thinking:    entry.Thinking,
+		})
+	}
+	return models
+}
+
+// AntigravityModelConfig captures static antigravity model overrides, including
+// Thinking budget limits and provider max completion tokens.
+type AntigravityModelConfig struct {
+	Thinking            *ThinkingSupport
+	MaxCompletionTokens int
+}
+
+// GetAntigravityModelConfig returns static configuration for antigravity models.
+// Keys use upstream model names returned by the Antigravity models endpoint.
+func GetAntigravityModelConfig() map[string]*AntigravityModelConfig {
+	return map[string]*AntigravityModelConfig{
+		// "rev19-uic3-1p":              {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true}},
+		"gemini-2.5-flash":           {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
+		"gemini-2.5-flash-lite":      {Thinking: &ThinkingSupport{Min: 0, Max: 24576, ZeroAllowed: true, DynamicAllowed: true}},
+		"gemini-3-pro-high":          {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
+		"gemini-3-pro-image":         {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"low", "high"}}},
+		"gemini-3-flash":             {Thinking: &ThinkingSupport{Min: 128, Max: 32768, ZeroAllowed: false, DynamicAllowed: true, Levels: []string{"minimal", "low", "medium", "high"}}},
+		"claude-sonnet-4-5-thinking": {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
+		"claude-opus-4-5-thinking":   {Thinking: &ThinkingSupport{Min: 1024, Max: 128000, ZeroAllowed: true, DynamicAllowed: true}, MaxCompletionTokens: 64000},
+		"claude-sonnet-4-5":          {MaxCompletionTokens: 64000},
+		"gpt-oss-120b-medium":        {},
+		"tab_flash_lite_preview":     {},
+	}
+}

internal/registry/model_registry.go

@@ -4,6 +4,7 @@
 package registry
 
 import (
+	"context"
 	"fmt"
 	"sort"
 	"strings"
@@ -50,6 +51,11 @@ type ModelInfo struct {
 	// Thinking holds provider-specific reasoning/thinking budget capabilities.
 	// This is optional and currently used for Gemini thinking budget normalization.
 	Thinking *ThinkingSupport `json:"thinking,omitempty"`
+
+	// UserDefined indicates this model was defined through config file's models[]
+	// array (e.g., openai-compatibility.*.models[], *-api-key.models[]).
+	// UserDefined models have thinking configuration passed through without validation.
+	UserDefined bool `json:"-"`
 }
 
 // ThinkingSupport describes a model family's supported internal reasoning budget range.
@@ -72,6 +78,8 @@ type ThinkingSupport struct {
 type ModelRegistration struct {
 	// Info contains the model metadata
 	Info *ModelInfo
+	// InfoByProvider maps provider identifiers to specific ModelInfo to support differing capabilities.
+	InfoByProvider map[string]*ModelInfo
 	// Count is the number of active clients that can provide this model
 	Count int
 	// LastUpdated tracks when this registration was last modified
@@ -84,16 +92,28 @@ type ModelRegistration struct {
 	SuspendedClients map[string]string
 }
 
+// ModelRegistryHook provides optional callbacks for external integrations to track model list changes.
+// Hook implementations must be non-blocking and resilient; calls are executed asynchronously and panics are recovered.
+type ModelRegistryHook interface {
+	OnModelsRegistered(ctx context.Context, provider, clientID string, models []*ModelInfo)
+	OnModelsUnregistered(ctx context.Context, provider, clientID string)
+}
+
 // ModelRegistry manages the global registry of available models
 type ModelRegistry struct {
 	// models maps model ID to registration information
 	models map[string]*ModelRegistration
 	// clientModels maps client ID to the models it provides
 	clientModels map[string][]string
+	// clientModelInfos maps client ID to a map of model ID -> ModelInfo
+	// This preserves the original model info provided by each client
+	clientModelInfos map[string]map[string]*ModelInfo
 	// clientProviders maps client ID to its provider identifier
 	clientProviders map[string]string
 	// mutex ensures thread-safe access to the registry
 	mutex *sync.RWMutex
+	// hook is an optional callback sink for model registration changes
+	hook ModelRegistryHook
 }
 
 // Global model registry instance
@@ -106,6 +126,7 @@ func GetGlobalRegistry() *ModelRegistry {
 		globalRegistry = &ModelRegistry{
 			models:           make(map[string]*ModelRegistration),
 			clientModels:     make(map[string][]string),
+			clientModelInfos: make(map[string]map[string]*ModelInfo),
 			clientProviders:  make(map[string]string),
 			mutex:            &sync.RWMutex{},
 		}
@@ -113,6 +134,71 @@ func GetGlobalRegistry() *ModelRegistry {
 	return globalRegistry
 }
 
+// LookupModelInfo searches dynamic registry (provider-specific > global) then static definitions.
+func LookupModelInfo(modelID string, provider ...string) *ModelInfo {
+	modelID = strings.TrimSpace(modelID)
+	if modelID == "" {
+		return nil
+	}
+
+	p := ""
+	if len(provider) > 0 {
+		p = strings.ToLower(strings.TrimSpace(provider[0]))
+	}
+
+	if info := GetGlobalRegistry().GetModelInfo(modelID, p); info != nil {
+		return info
+	}
+	return LookupStaticModelInfo(modelID)
+}
+
+// SetHook sets an optional hook for observing model registration changes.
+func (r *ModelRegistry) SetHook(hook ModelRegistryHook) {
+	if r == nil {
+		return
+	}
+	r.mutex.Lock()
+	defer r.mutex.Unlock()
+	r.hook = hook
+}
+
+const defaultModelRegistryHookTimeout = 5 * time.Second
+
+func (r *ModelRegistry) triggerModelsRegistered(provider, clientID string, models []*ModelInfo) {
+	hook := r.hook
+	if hook == nil {
+		return
+	}
+	modelsCopy := cloneModelInfosUnique(models)
+	go func() {
+		defer func() {
+			if recovered := recover(); recovered != nil {
+				log.Errorf("model registry hook OnModelsRegistered panic: %v", recovered)
+			}
+		}()
+		ctx, cancel := context.WithTimeout(context.Background(), defaultModelRegistryHookTimeout)
+		defer cancel()
+		hook.OnModelsRegistered(ctx, provider, clientID, modelsCopy)
+	}()
+}
+
+func (r *ModelRegistry) triggerModelsUnregistered(provider, clientID string) {
+	hook := r.hook
+	if hook == nil {
+		return
+	}
+	go func() {
+		defer func() {
+			if recovered := recover(); recovered != nil {
+				log.Errorf("model registry hook OnModelsUnregistered panic: %v", recovered)
+			}
+		}()
+		ctx, cancel := context.WithTimeout(context.Background(), defaultModelRegistryHookTimeout)
+		defer cancel()
+		hook.OnModelsUnregistered(ctx, provider, clientID)
+	}()
+}
+
 // RegisterClient registers a client and its supported models
 // Parameters:
 //   - clientID: Unique identifier for the client
@@ -144,6 +230,7 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 		// No models supplied; unregister existing client state if present.
 		r.unregisterClientInternal(clientID)
 		delete(r.clientModels, clientID)
+		delete(r.clientModelInfos, clientID)
 		delete(r.clientProviders, clientID)
 		misc.LogCredentialSeparator()
 		return
@@ -152,7 +239,7 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 	now := time.Now()
 
 	oldModels, hadExisting := r.clientModels[clientID]
-	oldProvider, _ := r.clientProviders[clientID]
+	oldProvider := r.clientProviders[clientID]
 	providerChanged := oldProvider != provider
 	if !hadExisting {
 		// Pure addition path.
@@ -161,11 +248,18 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 			r.addModelRegistration(modelID, provider, model, now)
 		}
 		r.clientModels[clientID] = append([]string(nil), rawModelIDs...)
+		// Store client's own model infos
+		clientInfos := make(map[string]*ModelInfo, len(newModels))
+		for id, m := range newModels {
+			clientInfos[id] = cloneModelInfo(m)
+		}
+		r.clientModelInfos[clientID] = clientInfos
 		if provider != "" {
 			r.clientProviders[clientID] = provider
 		} else {
 			delete(r.clientProviders, clientID)
 		}
+		r.triggerModelsRegistered(provider, clientID, models)
 		log.Debugf("Registered client %s from provider %s with %d models", clientID, clientProvider, len(rawModelIDs))
 		misc.LogCredentialSeparator()
 		return
@@ -208,6 +302,9 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 				if count, okProv := reg.Providers[oldProvider]; okProv {
 					if count <= toRemove {
 						delete(reg.Providers, oldProvider)
+						if reg.InfoByProvider != nil {
+							delete(reg.InfoByProvider, oldProvider)
+						}
 					} else {
 						reg.Providers[oldProvider] = count - toRemove
 					}
@@ -257,6 +354,12 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 		model := newModels[id]
 		if reg, ok := r.models[id]; ok {
 			reg.Info = cloneModelInfo(model)
+			if provider != "" {
+				if reg.InfoByProvider == nil {
+					reg.InfoByProvider = make(map[string]*ModelInfo)
+				}
+				reg.InfoByProvider[provider] = cloneModelInfo(model)
+			}
 			reg.LastUpdated = now
 			if reg.QuotaExceededClients != nil {
 				delete(reg.QuotaExceededClients, clientID)
@@ -287,12 +390,19 @@ func (r *ModelRegistry) RegisterClient(clientID, clientProvider string, models [
 	if len(rawModelIDs) > 0 {
 		r.clientModels[clientID] = append([]string(nil), rawModelIDs...)
 	}
+	// Update client's own model infos
+	clientInfos := make(map[string]*ModelInfo, len(newModels))
+	for id, m := range newModels {
+		clientInfos[id] = cloneModelInfo(m)
+	}
+	r.clientModelInfos[clientID] = clientInfos
 	if provider != "" {
 		r.clientProviders[clientID] = provider
 	} else {
 		delete(r.clientProviders, clientID)
 	}
 
+	r.triggerModelsRegistered(provider, clientID, models)
 	if len(added) == 0 && len(removed) == 0 && !providerChanged {
 		// Only metadata (e.g., display name) changed; skip separator when no log output.
 		return
@@ -313,11 +423,15 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 		if existing.SuspendedClients == nil {
 			existing.SuspendedClients = make(map[string]string)
 		}
+		if existing.InfoByProvider == nil {
+			existing.InfoByProvider = make(map[string]*ModelInfo)
+		}
 		if provider != "" {
 			if existing.Providers == nil {
 				existing.Providers = make(map[string]int)
 			}
 			existing.Providers[provider]++
+			existing.InfoByProvider[provider] = cloneModelInfo(model)
 		}
 		log.Debugf("Incremented count for model %s, now %d clients", modelID, existing.Count)
 		return
@@ -325,6 +439,7 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 
 	registration := &ModelRegistration{
 		Info:                 cloneModelInfo(model),
+		InfoByProvider:       make(map[string]*ModelInfo),
 		Count:                1,
 		LastUpdated:          now,
 		QuotaExceededClients: make(map[string]*time.Time),
@@ -332,6 +447,7 @@ func (r *ModelRegistry) addModelRegistration(modelID, provider string, model *Mo
 	}
 	if provider != "" {
 		registration.Providers = map[string]int{provider: 1}
+		registration.InfoByProvider[provider] = cloneModelInfo(model)
 	}
 	r.models[modelID] = registration
 	log.Debugf("Registered new model %s from provider %s", modelID, provider)
@@ -357,6 +473,9 @@ func (r *ModelRegistry) removeModelRegistration(clientID, modelID, provider stri
 		if count, ok := registration.Providers[provider]; ok {
 			if count <= 1 {
 				delete(registration.Providers, provider)
+				if registration.InfoByProvider != nil {
+					delete(registration.InfoByProvider, provider)
+				}
 			} else {
 				registration.Providers[provider] = count - 1
 			}
@@ -383,6 +502,25 @@ func cloneModelInfo(model *ModelInfo) *ModelInfo {
 	return &copyModel
 }
 
+func cloneModelInfosUnique(models []*ModelInfo) []*ModelInfo {
+	if len(models) == 0 {
+		return nil
+	}
+	cloned := make([]*ModelInfo, 0, len(models))
+	seen := make(map[string]struct{}, len(models))
+	for _, model := range models {
+		if model == nil || model.ID == "" {
+			continue
+		}
+		if _, exists := seen[model.ID]; exists {
+			continue
+		}
+		seen[model.ID] = struct{}{}
+		cloned = append(cloned, cloneModelInfo(model))
+	}
+	return cloned
+}
+
 // UnregisterClient removes a client and decrements counts for its models
 // Parameters:
 //   - clientID: Unique identifier for the client to remove
@@ -419,6 +557,9 @@ func (r *ModelRegistry) unregisterClientInternal(clientID string) {
 				if count, ok := registration.Providers[provider]; ok {
 					if count <= 1 {
 						delete(registration.Providers, provider)
+						if registration.InfoByProvider != nil {
+							delete(registration.InfoByProvider, provider)
+						}
 					} else {
 						registration.Providers[provider] = count - 1
 					}
@@ -436,12 +577,14 @@ func (r *ModelRegistry) unregisterClientInternal(clientID string) {
 	}
 
 	delete(r.clientModels, clientID)
+	delete(r.clientModelInfos, clientID)
 	if hasProvider {
 		delete(r.clientProviders, clientID)
 	}
 	log.Debugf("Unregistered client %s", clientID)
 	// Separator line after completing client unregistration (after the summary line)
 	misc.LogCredentialSeparator()
+	r.triggerModelsUnregistered(provider, clientID)
 }
 
 // SetModelQuotaExceeded marks a model as quota exceeded for a specific client
@@ -607,6 +750,131 @@ func (r *ModelRegistry) GetAvailableModels(handlerType string) []map[string]any
 	return models
 }
 
+// GetAvailableModelsByProvider returns models available for the given provider identifier.
+// Parameters:
+//   - provider: Provider identifier (e.g., "codex", "gemini", "antigravity")
+//
+// Returns:
+//   - []*ModelInfo: List of available models for the provider
+func (r *ModelRegistry) GetAvailableModelsByProvider(provider string) []*ModelInfo {
+	provider = strings.ToLower(strings.TrimSpace(provider))
+	if provider == "" {
+		return nil
+	}
+
+	r.mutex.RLock()
+	defer r.mutex.RUnlock()
+
+	type providerModel struct {
+		count int
+		info  *ModelInfo
+	}
+
+	providerModels := make(map[string]*providerModel)
+
+	for clientID, clientProvider := range r.clientProviders {
+		if clientProvider != provider {
+			continue
+		}
+		modelIDs := r.clientModels[clientID]
+		if len(modelIDs) == 0 {
+			continue
+		}
+		clientInfos := r.clientModelInfos[clientID]
+		for _, modelID := range modelIDs {
+			modelID = strings.TrimSpace(modelID)
+			if modelID == "" {
+				continue
+			}
+			entry := providerModels[modelID]
+			if entry == nil {
+				entry = &providerModel{}
+				providerModels[modelID] = entry
+			}
+			entry.count++
+			if entry.info == nil {
+				if clientInfos != nil {
+					if info := clientInfos[modelID]; info != nil {
+						entry.info = info
+					}
+				}
+				if entry.info == nil {
+					if reg, ok := r.models[modelID]; ok && reg != nil && reg.Info != nil {
+						entry.info = reg.Info
+					}
+				}
+			}
+		}
+	}
+
+	if len(providerModels) == 0 {
+		return nil
+	}
+
+	quotaExpiredDuration := 5 * time.Minute
+	now := time.Now()
+	result := make([]*ModelInfo, 0, len(providerModels))
+
+	for modelID, entry := range providerModels {
+		if entry == nil || entry.count <= 0 {
+			continue
+		}
+		registration, ok := r.models[modelID]
+
+		expiredClients := 0
+		cooldownSuspended := 0
+		otherSuspended := 0
+		if ok && registration != nil {
+			if registration.QuotaExceededClients != nil {
+				for clientID, quotaTime := range registration.QuotaExceededClients {
+					if clientID == "" {
+						continue
+					}
+					if p, okProvider := r.clientProviders[clientID]; !okProvider || p != provider {
+						continue
+					}
+					if quotaTime != nil && now.Sub(*quotaTime) < quotaExpiredDuration {
+						expiredClients++
+					}
+				}
+			}
+			if registration.SuspendedClients != nil {
+				for clientID, reason := range registration.SuspendedClients {
+					if clientID == "" {
+						continue
+					}
+					if p, okProvider := r.clientProviders[clientID]; !okProvider || p != provider {
+						continue
+					}
+					if strings.EqualFold(reason, "quota") {
+						cooldownSuspended++
+						continue
+					}
+					otherSuspended++
+				}
+			}
+		}
+
+		availableClients := entry.count
+		effectiveClients := availableClients - expiredClients - otherSuspended
+		if effectiveClients < 0 {
+			effectiveClients = 0
+		}
+
+		if effectiveClients > 0 || (availableClients > 0 && (expiredClients > 0 || cooldownSuspended > 0) && otherSuspended == 0) {
+			if entry.info != nil {
+				result = append(result, entry.info)
+				continue
+			}
+			if ok && registration != nil && registration.Info != nil {
+				result = append(result, registration.Info)
+			}
+		}
+	}
+
+	return result
+}
+
 // GetModelCount returns the number of available clients for a specific model
 // Parameters:
 //   - modelID: The model ID to check
@@ -698,12 +966,22 @@ func (r *ModelRegistry) GetModelProviders(modelID string) []string {
 	return result
 }
 
-// GetModelInfo returns the registered ModelInfo for the given model ID, if present.
-// Returns nil if the model is unknown to the registry.
-func (r *ModelRegistry) GetModelInfo(modelID string) *ModelInfo {
+// GetModelInfo returns ModelInfo, prioritizing provider-specific definition if available.
+func (r *ModelRegistry) GetModelInfo(modelID, provider string) *ModelInfo {
 	r.mutex.RLock()
 	defer r.mutex.RUnlock()
 	if reg, ok := r.models[modelID]; ok && reg != nil {
+		// Try provider specific definition first
+		if provider != "" && reg.InfoByProvider != nil {
+			if reg.Providers != nil {
+				if count, ok := reg.Providers[provider]; ok && count > 0 {
+					if info, ok := reg.InfoByProvider[provider]; ok && info != nil {
+						return info
+					}
+				}
+			}
+		}
+		// Fallback to global info (last registered)
 		return reg.Info
 	}
 	return nil
@@ -755,10 +1033,10 @@ func (r *ModelRegistry) convertModelToMap(model *ModelInfo, handlerType string)
 			"owned_by": model.OwnedBy,
 		}
 		if model.Created > 0 {
-			result["created"] = model.Created
+			result["created_at"] = model.Created
 		}
 		if model.Type != "" {
-			result["type"] = model.Type
+			result["type"] = "model"
 		}
 		if model.DisplayName != "" {
 			result["display_name"] = model.DisplayName
@@ -871,3 +1149,44 @@ func (r *ModelRegistry) GetFirstAvailableModel(handlerType string) (string, erro
 
 	return "", fmt.Errorf("no available clients for any model in handler type: %s", handlerType)
 }
+
+// GetModelsForClient returns the models registered for a specific client.
+// Parameters:
+//   - clientID: The client identifier (typically auth file name or auth ID)
+//
+// Returns:
+//   - []*ModelInfo: List of models registered for this client, nil if client not found
+func (r *ModelRegistry) GetModelsForClient(clientID string) []*ModelInfo {
+	r.mutex.RLock()
+	defer r.mutex.RUnlock()
+
+	modelIDs, exists := r.clientModels[clientID]
+	if !exists || len(modelIDs) == 0 {
+		return nil
+	}
+
+	// Try to use client-specific model infos first
+	clientInfos := r.clientModelInfos[clientID]
+
+	seen := make(map[string]struct{})
+	result := make([]*ModelInfo, 0, len(modelIDs))
+	for _, modelID := range modelIDs {
+		if _, dup := seen[modelID]; dup {
+			continue
+		}
+		seen[modelID] = struct{}{}
+
+		// Prefer client's own model info to preserve original type/owned_by
+		if clientInfos != nil {
+			if info, ok := clientInfos[modelID]; ok && info != nil {
+				result = append(result, info)
+				continue
+			}
+		}
+		// Fallback to global registry (for backwards compatibility)
+		if reg, ok := r.models[modelID]; ok && reg.Info != nil {
+			result = append(result, reg.Info)
+		}
+	}
+	return result
+}

internal/registry/model_registry_hook_test.go

@@ -0,0 +1,204 @@
+package registry
+
+import (
+	"context"
+	"sync"
+	"testing"
+	"time"
+)
+
+func newTestModelRegistry() *ModelRegistry {
+	return &ModelRegistry{
+		models:           make(map[string]*ModelRegistration),
+		clientModels:     make(map[string][]string),
+		clientModelInfos: make(map[string]map[string]*ModelInfo),
+		clientProviders:  make(map[string]string),
+		mutex:            &sync.RWMutex{},
+	}
+}
+
+type registeredCall struct {
+	provider string
+	clientID string
+	models   []*ModelInfo
+}
+
+type unregisteredCall struct {
+	provider string
+	clientID string
+}
+
+type capturingHook struct {
+	registeredCh   chan registeredCall
+	unregisteredCh chan unregisteredCall
+}
+
+func (h *capturingHook) OnModelsRegistered(ctx context.Context, provider, clientID string, models []*ModelInfo) {
+	h.registeredCh <- registeredCall{provider: provider, clientID: clientID, models: models}
+}
+
+func (h *capturingHook) OnModelsUnregistered(ctx context.Context, provider, clientID string) {
+	h.unregisteredCh <- unregisteredCall{provider: provider, clientID: clientID}
+}
+
+func TestModelRegistryHook_OnModelsRegisteredCalled(t *testing.T) {
+	r := newTestModelRegistry()
+	hook := &capturingHook{
+		registeredCh:   make(chan registeredCall, 1),
+		unregisteredCh: make(chan unregisteredCall, 1),
+	}
+	r.SetHook(hook)
+
+	inputModels := []*ModelInfo{
+		{ID: "m1", DisplayName: "Model One"},
+		{ID: "m2", DisplayName: "Model Two"},
+	}
+	r.RegisterClient("client-1", "OpenAI", inputModels)
+
+	select {
+	case call := <-hook.registeredCh:
+		if call.provider != "openai" {
+			t.Fatalf("provider mismatch: got %q, want %q", call.provider, "openai")
+		}
+		if call.clientID != "client-1" {
+			t.Fatalf("clientID mismatch: got %q, want %q", call.clientID, "client-1")
+		}
+		if len(call.models) != 2 {
+			t.Fatalf("models length mismatch: got %d, want %d", len(call.models), 2)
+		}
+		if call.models[0] == nil || call.models[0].ID != "m1" {
+			t.Fatalf("models[0] mismatch: got %#v, want ID=%q", call.models[0], "m1")
+		}
+		if call.models[1] == nil || call.models[1].ID != "m2" {
+			t.Fatalf("models[1] mismatch: got %#v, want ID=%q", call.models[1], "m2")
+		}
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for OnModelsRegistered hook call")
+	}
+}
+
+func TestModelRegistryHook_OnModelsUnregisteredCalled(t *testing.T) {
+	r := newTestModelRegistry()
+	hook := &capturingHook{
+		registeredCh:   make(chan registeredCall, 1),
+		unregisteredCh: make(chan unregisteredCall, 1),
+	}
+	r.SetHook(hook)
+
+	r.RegisterClient("client-1", "OpenAI", []*ModelInfo{{ID: "m1"}})
+	select {
+	case <-hook.registeredCh:
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for OnModelsRegistered hook call")
+	}
+
+	r.UnregisterClient("client-1")
+
+	select {
+	case call := <-hook.unregisteredCh:
+		if call.provider != "openai" {
+			t.Fatalf("provider mismatch: got %q, want %q", call.provider, "openai")
+		}
+		if call.clientID != "client-1" {
+			t.Fatalf("clientID mismatch: got %q, want %q", call.clientID, "client-1")
+		}
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for OnModelsUnregistered hook call")
+	}
+}
+
+type blockingHook struct {
+	started chan struct{}
+	unblock chan struct{}
+}
+
+func (h *blockingHook) OnModelsRegistered(ctx context.Context, provider, clientID string, models []*ModelInfo) {
+	select {
+	case <-h.started:
+	default:
+		close(h.started)
+	}
+	<-h.unblock
+}
+
+func (h *blockingHook) OnModelsUnregistered(ctx context.Context, provider, clientID string) {}
+
+func TestModelRegistryHook_DoesNotBlockRegisterClient(t *testing.T) {
+	r := newTestModelRegistry()
+	hook := &blockingHook{
+		started: make(chan struct{}),
+		unblock: make(chan struct{}),
+	}
+	r.SetHook(hook)
+	defer close(hook.unblock)
+
+	done := make(chan struct{})
+	go func() {
+		r.RegisterClient("client-1", "OpenAI", []*ModelInfo{{ID: "m1"}})
+		close(done)
+	}()
+
+	select {
+	case <-hook.started:
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for hook to start")
+	}
+
+	select {
+	case <-done:
+	case <-time.After(200 * time.Millisecond):
+		t.Fatal("RegisterClient appears to be blocked by hook")
+	}
+
+	if !r.ClientSupportsModel("client-1", "m1") {
+		t.Fatal("model registration failed; expected client to support model")
+	}
+}
+
+type panicHook struct {
+	registeredCalled   chan struct{}
+	unregisteredCalled chan struct{}
+}
+
+func (h *panicHook) OnModelsRegistered(ctx context.Context, provider, clientID string, models []*ModelInfo) {
+	if h.registeredCalled != nil {
+		h.registeredCalled <- struct{}{}
+	}
+	panic("boom")
+}
+
+func (h *panicHook) OnModelsUnregistered(ctx context.Context, provider, clientID string) {
+	if h.unregisteredCalled != nil {
+		h.unregisteredCalled <- struct{}{}
+	}
+	panic("boom")
+}
+
+func TestModelRegistryHook_PanicDoesNotAffectRegistry(t *testing.T) {
+	r := newTestModelRegistry()
+	hook := &panicHook{
+		registeredCalled:   make(chan struct{}, 1),
+		unregisteredCalled: make(chan struct{}, 1),
+	}
+	r.SetHook(hook)
+
+	r.RegisterClient("client-1", "OpenAI", []*ModelInfo{{ID: "m1"}})
+
+	select {
+	case <-hook.registeredCalled:
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for OnModelsRegistered hook call")
+	}
+
+	if !r.ClientSupportsModel("client-1", "m1") {
+		t.Fatal("model registration failed; expected client to support model")
+	}
+
+	r.UnregisterClient("client-1")
+
+	select {
+	case <-hook.unregisteredCalled:
+	case <-time.After(2 * time.Second):
+		t.Fatal("timeout waiting for OnModelsUnregistered hook call")
+	}
+}

internal/routing/adapter.go

@@ -0,0 +1,39 @@
+// Package routing provides adapter to integrate with existing codebase.
+package routing
+
+import (
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+)
+
+// Adapter bridges the new routing layer with existing auth manager.
+type Adapter struct {
+	router *Router
+	exec   *Executor
+}
+
+// NewAdapter creates a new adapter with the given configuration and auth manager.
+func NewAdapter(cfg *config.Config, authManager *coreauth.Manager) *Adapter {
+	registry := NewRegistry()
+	
+	// TODO: Register OAuth providers from authManager
+	// TODO: Register API key providers from cfg
+	
+	router := NewRouter(registry, cfg)
+	exec := NewExecutor(router)
+	
+	return &Adapter{
+		router: router,
+		exec:   exec,
+	}
+}
+
+// Router returns the underlying router.
+func (a *Adapter) Router() *Router {
+	return a.router
+}
+
+// Executor returns the underlying executor.
+func (a *Adapter) Executor() *Executor {
+	return a.exec
+}

internal/routing/ctxkeys/keys.go

@@ -0,0 +1,11 @@
+package ctxkeys
+
+type key string
+
+const (
+	MappedModel     key = "mapped_model"
+	FallbackModels  key = "fallback_models"
+	RouteCandidates key = "route_candidates"
+	RoutingDecision key = "routing_decision"
+	MappingApplied  key = "mapping_applied"
+)

internal/routing/executor.go

@@ -0,0 +1,111 @@
+package routing
+
+import (
+	"context"
+	"errors"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+	log "github.com/sirupsen/logrus"
+)
+
+// Executor handles request execution with fallback support.
+type Executor struct {
+	router *Router
+}
+
+// NewExecutor creates a new executor with the given router.
+func NewExecutor(router *Router) *Executor {
+	return &Executor{router: router}
+}
+
+// Execute sends the request through the routing decision.
+func (e *Executor) Execute(ctx context.Context, req executor.Request) (executor.Response, error) {
+	decision := e.router.Resolve(req.Model)
+	
+	log.Debugf("routing: %s -> %s (%d candidates)", 
+		decision.RequestedModel, 
+		decision.ResolvedModel, 
+		len(decision.Candidates))
+
+	var lastErr error
+	tried := make(map[string]struct{})
+
+	for i, candidate := range decision.Candidates {
+		key := candidate.Provider.Name() + "/" + candidate.Model
+		if _, ok := tried[key]; ok {
+			continue
+		}
+		tried[key] = struct{}{}
+
+		log.Debugf("routing: trying candidate %d/%d: %s with model %s",
+			i+1, len(decision.Candidates), candidate.Provider.Name(), candidate.Model)
+
+		req.Model = candidate.Model
+		resp, err := candidate.Provider.Execute(ctx, candidate.Model, req)
+		if err == nil {
+			return resp, nil
+		}
+
+		lastErr = err
+		log.Debugf("routing: candidate failed: %v", err)
+
+		// Check if it's a fatal error (not retryable)
+		if isFatalError(err) {
+			break
+		}
+	}
+
+	if lastErr != nil {
+		return executor.Response{}, lastErr
+	}
+	return executor.Response{}, errors.New("no available providers")
+}
+
+// ExecuteStream sends a streaming request through the routing decision.
+func (e *Executor) ExecuteStream(ctx context.Context, req executor.Request) (<-chan executor.StreamChunk, error) {
+	decision := e.router.Resolve(req.Model)
+
+	log.Debugf("routing stream: %s -> %s (%d candidates)",
+		decision.RequestedModel,
+		decision.ResolvedModel,
+		len(decision.Candidates))
+
+	var lastErr error
+	tried := make(map[string]struct{})
+
+	for i, candidate := range decision.Candidates {
+		key := candidate.Provider.Name() + "/" + candidate.Model
+		if _, ok := tried[key]; ok {
+			continue
+		}
+		tried[key] = struct{}{}
+
+		log.Debugf("routing stream: trying candidate %d/%d: %s with model %s",
+			i+1, len(decision.Candidates), candidate.Provider.Name(), candidate.Model)
+
+		req.Model = candidate.Model
+		chunks, err := candidate.Provider.ExecuteStream(ctx, candidate.Model, req)
+		if err == nil {
+			return chunks, nil
+		}
+
+		lastErr = err
+		log.Debugf("routing stream: candidate failed: %v", err)
+
+		if isFatalError(err) {
+			break
+		}
+	}
+
+	if lastErr != nil {
+		return nil, lastErr
+	}
+	return nil, errors.New("no available providers")
+}
+
+// isFatalError returns true if the error is not retryable.
+func isFatalError(err error) bool {
+	// TODO: implement based on error type
+	// For now, all errors are retryable
+	return false
+}

internal/routing/extractor.go

@@ -0,0 +1,59 @@
+package routing
+
+import (
+	"strings"
+
+	"github.com/tidwall/gjson"
+)
+
+// ModelExtractor extracts model names from request data.
+type ModelExtractor interface {
+	// Extract returns the model name from the request body and gin parameters.
+	// The ginParams map contains route parameters like "action" and "path".
+	Extract(body []byte, ginParams map[string]string) (string, error)
+}
+
+// DefaultModelExtractor is the standard implementation of ModelExtractor.
+type DefaultModelExtractor struct{}
+
+// NewModelExtractor creates a new DefaultModelExtractor.
+func NewModelExtractor() *DefaultModelExtractor {
+	return &DefaultModelExtractor{}
+}
+
+// Extract extracts the model name from the request.
+// It checks in order:
+// 1. JSON body "model" field (OpenAI, Claude format)
+// 2. "action" parameter for Gemini standard format (e.g., "gemini-pro:generateContent")
+// 3. "path" parameter for AMP CLI Gemini format (e.g., "/publishers/google/models/gemini-3-pro:streamGenerateContent")
+func (e *DefaultModelExtractor) Extract(body []byte, ginParams map[string]string) (string, error) {
+	// First try to parse from JSON body (OpenAI, Claude, etc.)
+	if result := gjson.GetBytes(body, "model"); result.Exists() && result.Type == gjson.String {
+		return result.String(), nil
+	}
+
+	// For Gemini requests, model is in the URL path
+	// Standard format: /models/{model}:generateContent -> :action parameter
+	if action, ok := ginParams["action"]; ok && action != "" {
+		// Split by colon to get model name (e.g., "gemini-pro:generateContent" -> "gemini-pro")
+		parts := strings.Split(action, ":")
+		if len(parts) > 0 && parts[0] != "" {
+			return parts[0], nil
+		}
+	}
+
+	// AMP CLI format: /publishers/google/models/{model}:method -> *path parameter
+	// Example: /publishers/google/models/gemini-3-pro-preview:streamGenerateContent
+	if path, ok := ginParams["path"]; ok && path != "" {
+		// Look for /models/{model}:method pattern
+		if idx := strings.Index(path, "/models/"); idx >= 0 {
+			modelPart := path[idx+8:] // Skip "/models/"
+			// Split by colon to get model name
+			if colonIdx := strings.Index(modelPart, ":"); colonIdx > 0 {
+				return modelPart[:colonIdx], nil
+			}
+		}
+	}
+
+	return "", nil
+}

internal/routing/extractor_test.go

@@ -0,0 +1,214 @@
+package routing
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestModelExtractor_ExtractFromJSONBody(t *testing.T) {
+	extractor := NewModelExtractor()
+
+	tests := []struct {
+		name     string
+		body     []byte
+		want     string
+		wantErr  bool
+	}{
+		{
+			name: "extract from JSON body with model field",
+			body: []byte(`{"model":"gpt-4.1"}`),
+			want: "gpt-4.1",
+		},
+		{
+			name: "extract claude model from JSON body",
+			body: []byte(`{"model":"claude-3-5-sonnet-20241022"}`),
+			want: "claude-3-5-sonnet-20241022",
+		},
+		{
+			name: "extract with additional fields",
+			body: []byte(`{"model":"gpt-4","messages":[{"role":"user","content":"hello"}]}`),
+			want: "gpt-4",
+		},
+		{
+			name: "empty body returns empty",
+			body: []byte{},
+			want: "",
+		},
+		{
+			name: "no model field returns empty",
+			body: []byte(`{"messages":[]}`),
+			want: "",
+		},
+		{
+			name: "model is not string returns empty",
+			body: []byte(`{"model":123}`),
+			want: "",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := extractor.Extract(tt.body, nil)
+			if tt.wantErr {
+				assert.Error(t, err)
+				return
+			}
+			assert.NoError(t, err)
+			assert.Equal(t, tt.want, got)
+		})
+	}
+}
+
+func TestModelExtractor_ExtractFromGeminiActionParam(t *testing.T) {
+	extractor := NewModelExtractor()
+
+	tests := []struct {
+		name      string
+		body      []byte
+		ginParams map[string]string
+		want      string
+	}{
+		{
+			name:      "extract from action parameter - gemini-pro",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"action": "gemini-pro:generateContent"},
+			want:      "gemini-pro",
+		},
+		{
+			name:      "extract from action parameter - gemini-ultra",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"action": "gemini-ultra:chat"},
+			want:      "gemini-ultra",
+		},
+		{
+			name:      "empty action returns empty",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"action": ""},
+			want:      "",
+		},
+		{
+			name:      "action without colon returns full value",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"action": "gemini-model"},
+			want:      "gemini-model",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := extractor.Extract(tt.body, tt.ginParams)
+			assert.NoError(t, err)
+			assert.Equal(t, tt.want, got)
+		})
+	}
+}
+
+func TestModelExtractor_ExtractFromGeminiV1Beta1Path(t *testing.T) {
+	extractor := NewModelExtractor()
+
+	tests := []struct {
+		name      string
+		body      []byte
+		ginParams map[string]string
+		want      string
+	}{
+		{
+			name:      "extract from v1beta1 path - gemini-3-pro",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"path": "/publishers/google/models/gemini-3-pro:streamGenerateContent"},
+			want:      "gemini-3-pro",
+		},
+		{
+			name:      "extract from v1beta1 path with preview",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"path": "/publishers/google/models/gemini-3-pro-preview:generateContent"},
+			want:      "gemini-3-pro-preview",
+		},
+		{
+			name:      "path without models segment returns empty",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"path": "/publishers/google/gemini-3-pro:streamGenerateContent"},
+			want:      "",
+		},
+		{
+			name:      "empty path returns empty",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"path": ""},
+			want:      "",
+		},
+		{
+			name:      "path with /models/ but no colon returns empty",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"path": "/publishers/google/models/gemini-3-pro"},
+			want:      "",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := extractor.Extract(tt.body, tt.ginParams)
+			assert.NoError(t, err)
+			assert.Equal(t, tt.want, got)
+		})
+	}
+}
+
+func TestModelExtractor_ExtractPriority(t *testing.T) {
+	extractor := NewModelExtractor()
+
+	// JSON body takes priority over gin params
+	t.Run("JSON body takes priority over action param", func(t *testing.T) {
+		body := []byte(`{"model":"gpt-4"}`)
+		params := map[string]string{"action": "gemini-pro:generateContent"}
+		got, err := extractor.Extract(body, params)
+		assert.NoError(t, err)
+		assert.Equal(t, "gpt-4", got)
+	})
+
+	// Action param takes priority over path param
+	t.Run("action param takes priority over path param", func(t *testing.T) {
+		body := []byte(`{}`)
+		params := map[string]string{
+			"action": "gemini-action:generate",
+			"path":   "/publishers/google/models/gemini-path:streamGenerateContent",
+		}
+		got, err := extractor.Extract(body, params)
+		assert.NoError(t, err)
+		assert.Equal(t, "gemini-action", got)
+	})
+}
+
+func TestModelExtractor_NoModelFound(t *testing.T) {
+	extractor := NewModelExtractor()
+
+	tests := []struct {
+		name      string
+		body      []byte
+		ginParams map[string]string
+	}{
+		{
+			name:      "empty body and no params",
+			body:      []byte{},
+			ginParams: nil,
+		},
+		{
+			name:      "body without model and no params",
+			body:      []byte(`{"messages":[]}`),
+			ginParams: map[string]string{},
+		},
+		{
+			name:      "irrelevant params only",
+			body:      []byte(`{}`),
+			ginParams: map[string]string{"other": "value"},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := extractor.Extract(tt.body, tt.ginParams)
+			assert.NoError(t, err)
+			assert.Empty(t, got)
+		})
+	}
+}

internal/routing/provider.go

@@ -0,0 +1,80 @@
+// Package routing provides unified model routing for all provider types.
+package routing
+
+import (
+	"context"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+)
+
+// ProviderType indicates the type of provider.
+type ProviderType string
+
+const (
+	ProviderTypeOAuth  ProviderType = "oauth"
+	ProviderTypeAPIKey ProviderType = "api_key"
+	ProviderTypeVertex ProviderType = "vertex"
+)
+
+// Provider is the unified interface for all provider types (OAuth, API key, etc.).
+type Provider interface {
+	// Name returns the unique provider identifier.
+	Name() string
+
+	// Type returns the provider type.
+	Type() ProviderType
+
+	// SupportsModel returns true if this provider can handle the given model.
+	SupportsModel(model string) bool
+
+	// Available returns true if the provider is available for the model (not quota exceeded).
+	Available(model string) bool
+
+	// Priority returns the priority for this provider (lower = tried first).
+	Priority() int
+
+	// Execute sends the request to the provider.
+	Execute(ctx context.Context, model string, req executor.Request) (executor.Response, error)
+
+	// ExecuteStream sends a streaming request to the provider.
+	ExecuteStream(ctx context.Context, model string, req executor.Request) (<-chan executor.StreamChunk, error)
+}
+
+// ProviderCandidate represents a provider + model combination to try.
+type ProviderCandidate struct {
+	Provider Provider
+	Model    string // The actual model name to use (may be different from requested due to aliasing)
+}
+
+// Registry manages all available providers.
+type Registry struct {
+	providers []Provider
+}
+
+// NewRegistry creates a new provider registry.
+func NewRegistry() *Registry {
+	return &Registry{
+		providers: make([]Provider, 0),
+	}
+}
+
+// Register adds a provider to the registry.
+func (r *Registry) Register(p Provider) {
+	r.providers = append(r.providers, p)
+}
+
+// FindProviders returns all providers that support the given model and are available.
+func (r *Registry) FindProviders(model string) []Provider {
+	var result []Provider
+	for _, p := range r.providers {
+		if p.SupportsModel(model) && p.Available(model) {
+			result = append(result, p)
+		}
+	}
+	return result
+}
+
+// All returns all registered providers.
+func (r *Registry) All() []Provider {
+	return r.providers
+}

internal/routing/providers/apikey.go

@@ -0,0 +1,156 @@
+package providers
+
+import (
+	"context"
+	"errors"
+	"net/http"
+	"strings"
+	"sync"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/routing"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+)
+
+// APIKeyProvider wraps API key configs as routing.Provider.
+type APIKeyProvider struct {
+	name     string
+	provider string // claude, gemini, codex, vertex
+	keys     []APIKeyEntry
+	mu       sync.RWMutex
+	client   HTTPClient
+}
+
+// APIKeyEntry represents a single API key configuration.
+type APIKeyEntry struct {
+	APIKey  string
+	BaseURL string
+	Models  []config.ClaudeModel // Using ClaudeModel as generic model alias
+}
+
+// HTTPClient interface for making HTTP requests.
+type HTTPClient interface {
+	Do(req *http.Request) (*http.Response, error)
+}
+
+// NewAPIKeyProvider creates a new API key provider.
+func NewAPIKeyProvider(name, provider string, client HTTPClient) *APIKeyProvider {
+	return &APIKeyProvider{
+		name:     name,
+		provider: provider,
+		keys:     make([]APIKeyEntry, 0),
+		client:   client,
+	}
+}
+
+// Name returns the provider name.
+func (p *APIKeyProvider) Name() string {
+	return p.name
+}
+
+// Type returns ProviderTypeAPIKey.
+func (p *APIKeyProvider) Type() routing.ProviderType {
+	return routing.ProviderTypeAPIKey
+}
+
+// SupportsModel checks if the model is supported by this provider.
+func (p *APIKeyProvider) SupportsModel(model string) bool {
+	p.mu.RLock()
+	defer p.mu.RUnlock()
+
+	for _, key := range p.keys {
+		for _, m := range key.Models {
+			if strings.EqualFold(m.Alias, model) || strings.EqualFold(m.Name, model) {
+				return true
+			}
+		}
+	}
+	return false
+}
+
+// Available always returns true for API keys (unless explicitly disabled).
+func (p *APIKeyProvider) Available(model string) bool {
+	return p.SupportsModel(model)
+}
+
+// Priority returns the priority (API key is lower priority than OAuth).
+func (p *APIKeyProvider) Priority() int {
+	return 20
+}
+
+// Execute sends the request using the API key.
+func (p *APIKeyProvider) Execute(ctx context.Context, model string, req executor.Request) (executor.Response, error) {
+	key := p.selectKey(model)
+	if key == nil {
+		return executor.Response{}, ErrNoMatchingAPIKey
+	}
+
+	// Resolve the actual model name from alias
+	actualModel := p.resolveModel(key, model)
+
+	// Execute via HTTP client
+	return p.executeHTTP(ctx, key, actualModel, req)
+}
+
+// ExecuteStream sends a streaming request.
+func (p *APIKeyProvider) ExecuteStream(ctx context.Context, model string, req executor.Request) (
+	<-chan executor.StreamChunk, error) {
+	key := p.selectKey(model)
+	if key == nil {
+		return nil, ErrNoMatchingAPIKey
+	}
+
+	actualModel := p.resolveModel(key, model)
+	return p.executeHTTPStream(ctx, key, actualModel, req)
+}
+
+// AddKey adds an API key entry.
+func (p *APIKeyProvider) AddKey(entry APIKeyEntry) {
+	p.mu.Lock()
+	defer p.mu.Unlock()
+	p.keys = append(p.keys, entry)
+}
+
+// selectKey selects a key that supports the model.
+func (p *APIKeyProvider) selectKey(model string) *APIKeyEntry {
+	p.mu.RLock()
+	defer p.mu.RUnlock()
+
+	for _, key := range p.keys {
+		for _, m := range key.Models {
+			if strings.EqualFold(m.Alias, model) || strings.EqualFold(m.Name, model) {
+				return &key
+			}
+		}
+	}
+	return nil
+}
+
+// resolveModel resolves alias to actual model name.
+func (p *APIKeyProvider) resolveModel(key *APIKeyEntry, requested string) string {
+	for _, m := range key.Models {
+		if strings.EqualFold(m.Alias, requested) {
+			return m.Name
+		}
+	}
+	return requested
+}
+
+// executeHTTP makes the HTTP request.
+func (p *APIKeyProvider) executeHTTP(ctx context.Context, key *APIKeyEntry, model string, req executor.Request) (executor.Response, error) {
+	// TODO: implement actual HTTP execution
+	// This is a placeholder - actual implementation would build HTTP request
+	return executor.Response{}, errors.New("not yet implemented")
+}
+
+// executeHTTPStream makes a streaming HTTP request.
+func (p *APIKeyProvider) executeHTTPStream(ctx context.Context, key *APIKeyEntry, model string, req executor.Request) (
+	<-chan executor.StreamChunk, error) {
+	// TODO: implement actual HTTP streaming
+	return nil, errors.New("not yet implemented")
+}
+
+// Errors
+var (
+	ErrNoMatchingAPIKey = errors.New("no API key supports the requested model")
+)

internal/routing/providers/oauth.go

@@ -0,0 +1,132 @@
+package providers
+
+import (
+	"context"
+	"errors"
+	"sync"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/routing"
+	coreauth "github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/auth"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+)
+
+// OAuthProvider wraps OAuth-based auths as routing.Provider.
+type OAuthProvider struct {
+	name      string
+	auths     []*coreauth.Auth
+	mu        sync.RWMutex
+	executor  coreauth.ProviderExecutor
+}
+
+// NewOAuthProvider creates a new OAuth provider.
+func NewOAuthProvider(name string, exec coreauth.ProviderExecutor) *OAuthProvider {
+	return &OAuthProvider{
+		name:     name,
+		auths:    make([]*coreauth.Auth, 0),
+		executor: exec,
+	}
+}
+
+// Name returns the provider name.
+func (p *OAuthProvider) Name() string {
+	return p.name
+}
+
+// Type returns ProviderTypeOAuth.
+func (p *OAuthProvider) Type() routing.ProviderType {
+	return routing.ProviderTypeOAuth
+}
+
+// SupportsModel checks if any auth supports the model.
+func (p *OAuthProvider) SupportsModel(model string) bool {
+	p.mu.RLock()
+	defer p.mu.RUnlock()
+
+	// OAuth providers typically support models via oauth-model-alias
+	// The actual model support is determined at execution time
+	return true
+}
+
+// Available checks if there's an available auth for the model.
+func (p *OAuthProvider) Available(model string) bool {
+	p.mu.RLock()
+	defer p.mu.RUnlock()
+
+	for _, auth := range p.auths {
+		if p.isAuthAvailable(auth, model) {
+			return true
+		}
+	}
+	return false
+}
+
+// Priority returns the priority (OAuth is preferred over API key).
+func (p *OAuthProvider) Priority() int {
+	return 10
+}
+
+// Execute sends the request using an available OAuth auth.
+func (p *OAuthProvider) Execute(ctx context.Context, model string, req executor.Request) (executor.Response, error) {
+	auth := p.selectAuth(model)
+	if auth == nil {
+		return executor.Response{}, ErrNoAvailableAuth
+	}
+
+	return p.executor.Execute(ctx, auth, req, executor.Options{})
+}
+
+// ExecuteStream sends a streaming request.
+func (p *OAuthProvider) ExecuteStream(ctx context.Context, model string, req executor.Request) (<-chan executor.StreamChunk, error) {
+	auth := p.selectAuth(model)
+	if auth == nil {
+		return nil, ErrNoAvailableAuth
+	}
+
+	return p.executor.ExecuteStream(ctx, auth, req, executor.Options{})
+}
+
+// AddAuth adds an auth to this provider.
+func (p *OAuthProvider) AddAuth(auth *coreauth.Auth) {
+	p.mu.Lock()
+	defer p.mu.Unlock()
+	p.auths = append(p.auths, auth)
+}
+
+// RemoveAuth removes an auth from this provider.
+func (p *OAuthProvider) RemoveAuth(authID string) {
+	p.mu.Lock()
+	defer p.mu.Unlock()
+
+	filtered := make([]*coreauth.Auth, 0, len(p.auths))
+	for _, auth := range p.auths {
+		if auth.ID != authID {
+			filtered = append(filtered, auth)
+		}
+	}
+	p.auths = filtered
+}
+
+// isAuthAvailable checks if an auth is available for the model.
+func (p *OAuthProvider) isAuthAvailable(auth *coreauth.Auth, model string) bool {
+	// TODO: integrate with model_registry for quota checking
+	// For now, just check if auth exists
+	return auth != nil
+}
+
+// selectAuth selects an available auth for the model.
+func (p *OAuthProvider) selectAuth(model string) *coreauth.Auth {
+	p.mu.RLock()
+	defer p.mu.RUnlock()
+
+	for _, auth := range p.auths {
+		if p.isAuthAvailable(auth, model) {
+			return auth
+		}
+	}
+	return nil
+}
+
+// Errors
+var (
+	ErrNoAvailableAuth = errors.New("no available OAuth auth for model")
+)

internal/routing/rewriter.go

@@ -0,0 +1,159 @@
+package routing
+
+import (
+	"bytes"
+	"net/http"
+	"strings"
+
+	"github.com/tidwall/gjson"
+	"github.com/tidwall/sjson"
+	log "github.com/sirupsen/logrus"
+)
+
+// ModelRewriter handles model name rewriting in requests and responses.
+type ModelRewriter interface {
+	// RewriteRequestBody rewrites the model field in a JSON request body.
+	// Returns the modified body or the original if no rewrite was needed.
+	RewriteRequestBody(body []byte, newModel string) ([]byte, error)
+
+	// WrapResponseWriter wraps an http.ResponseWriter to rewrite model names in the response.
+	// Returns the wrapped writer and a cleanup function that must be called after the response is complete.
+	WrapResponseWriter(w http.ResponseWriter, requestedModel, resolvedModel string) (http.ResponseWriter, func())
+}
+
+// DefaultModelRewriter is the standard implementation of ModelRewriter.
+type DefaultModelRewriter struct{}
+
+// NewModelRewriter creates a new DefaultModelRewriter.
+func NewModelRewriter() *DefaultModelRewriter {
+	return &DefaultModelRewriter{}
+}
+
+// RewriteRequestBody replaces the model name in a JSON request body.
+func (r *DefaultModelRewriter) RewriteRequestBody(body []byte, newModel string) ([]byte, error) {
+	if !gjson.GetBytes(body, "model").Exists() {
+		return body, nil
+	}
+	result, err := sjson.SetBytes(body, "model", newModel)
+	if err != nil {
+		return body, err
+	}
+	return result, nil
+}
+
+// WrapResponseWriter wraps a response writer to rewrite model names.
+// The cleanup function must be called after the handler completes to flush any buffered data.
+func (r *DefaultModelRewriter) WrapResponseWriter(w http.ResponseWriter, requestedModel, resolvedModel string) (http.ResponseWriter, func()) {
+	rw := &responseRewriter{
+		ResponseWriter:  w,
+		body:            &bytes.Buffer{},
+		requestedModel:  requestedModel,
+		resolvedModel:   resolvedModel,
+	}
+	return rw, func() { rw.flush() }
+}
+
+// responseRewriter wraps http.ResponseWriter to intercept and modify the response body.
+type responseRewriter struct {
+	http.ResponseWriter
+	body           *bytes.Buffer
+	requestedModel string
+	resolvedModel  string
+	isStreaming    bool
+	wroteHeader    bool
+	flushed        bool
+}
+
+// Write intercepts response writes and buffers them for model name replacement.
+func (rw *responseRewriter) Write(data []byte) (int, error) {
+	// Ensure header is written
+	if !rw.wroteHeader {
+		rw.WriteHeader(http.StatusOK)
+	}
+
+	// Detect streaming on first write
+	if rw.body.Len() == 0 && !rw.isStreaming {
+		contentType := rw.Header().Get("Content-Type")
+		rw.isStreaming = strings.Contains(contentType, "text/event-stream") ||
+			strings.Contains(contentType, "stream")
+	}
+
+	if rw.isStreaming {
+		n, err := rw.ResponseWriter.Write(rw.rewriteStreamChunk(data))
+		if err == nil {
+			if flusher, ok := rw.ResponseWriter.(http.Flusher); ok {
+				flusher.Flush()
+			}
+		}
+		return n, err
+	}
+	return rw.body.Write(data)
+}
+
+// WriteHeader captures the status code and delegates to the underlying writer.
+func (rw *responseRewriter) WriteHeader(code int) {
+	if !rw.wroteHeader {
+		rw.wroteHeader = true
+		rw.ResponseWriter.WriteHeader(code)
+	}
+}
+
+// flush writes the buffered response with model names rewritten.
+func (rw *responseRewriter) flush() {
+	if rw.flushed {
+		return
+	}
+	rw.flushed = true
+
+	if rw.isStreaming {
+		if flusher, ok := rw.ResponseWriter.(http.Flusher); ok {
+			flusher.Flush()
+		}
+		return
+	}
+	if rw.body.Len() > 0 {
+		data := rw.rewriteModelInResponse(rw.body.Bytes())
+		if _, err := rw.ResponseWriter.Write(data); err != nil {
+			log.Warnf("response rewriter: failed to write rewritten response: %v", err)
+		}
+	}
+}
+
+// modelFieldPaths lists all JSON paths where model name may appear.
+var modelFieldPaths = []string{"model", "modelVersion", "response.modelVersion", "message.model"}
+
+// rewriteModelInResponse replaces all occurrences of the resolved model with the requested model.
+func (rw *responseRewriter) rewriteModelInResponse(data []byte) []byte {
+	if rw.requestedModel == "" || rw.resolvedModel == "" || rw.requestedModel == rw.resolvedModel {
+		return data
+	}
+
+	for _, path := range modelFieldPaths {
+		if gjson.GetBytes(data, path).Exists() {
+			data, _ = sjson.SetBytes(data, path, rw.requestedModel)
+		}
+	}
+	return data
+}
+
+// rewriteStreamChunk rewrites model names in SSE stream chunks.
+func (rw *responseRewriter) rewriteStreamChunk(chunk []byte) []byte {
+	if rw.requestedModel == "" || rw.resolvedModel == "" || rw.requestedModel == rw.resolvedModel {
+		return chunk
+	}
+
+	// SSE format: "data: {json}\n\n"
+	lines := bytes.Split(chunk, []byte("\n"))
+	for i, line := range lines {
+		if bytes.HasPrefix(line, []byte("data: ")) {
+			jsonData := bytes.TrimPrefix(line, []byte("data: "))
+			if len(jsonData) > 0 && jsonData[0] == '{' {
+				// Rewrite JSON in the data line
+				rewritten := rw.rewriteModelInResponse(jsonData)
+				lines[i] = append([]byte("data: "), rewritten...)
+			}
+		}
+	}
+
+	return bytes.Join(lines, []byte("\n"))
+}

internal/routing/rewriter_test.go

@@ -0,0 +1,342 @@
+package routing
+
+import (
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestModelRewriter_RewriteRequestBody(t *testing.T) {
+	rewriter := NewModelRewriter()
+
+	tests := []struct {
+		name       string
+		body       []byte
+		newModel   string
+		wantModel  string
+		wantChange bool
+	}{
+		{
+			name:       "rewrites model field in JSON body",
+			body:       []byte(`{"model":"gpt-4.1","messages":[]}`),
+			newModel:   "claude-local",
+			wantModel:  "claude-local",
+			wantChange: true,
+		},
+		{
+			name:       "rewrites with empty body returns empty",
+			body:       []byte{},
+			newModel:   "gpt-4",
+			wantModel:  "",
+			wantChange: false,
+		},
+		{
+			name:       "handles missing model field gracefully",
+			body:       []byte(`{"messages":[{"role":"user"}]}`),
+			newModel:   "gpt-4",
+			wantModel:  "",
+			wantChange: false,
+		},
+		{
+			name:       "preserves other fields when rewriting",
+			body:       []byte(`{"model":"old-model","temperature":0.7,"max_tokens":100}`),
+			newModel:   "new-model",
+			wantModel:  "new-model",
+			wantChange: true,
+		},
+		{
+			name:       "handles nested JSON structure",
+			body:       []byte(`{"model":"gpt-4","messages":[{"role":"user","content":"hello"}],"stream":true}`),
+			newModel:   "claude-3-opus",
+			wantModel:  "claude-3-opus",
+			wantChange: true,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result, err := rewriter.RewriteRequestBody(tt.body, tt.newModel)
+			require.NoError(t, err)
+
+			if tt.wantChange {
+				assert.NotEqual(t, string(tt.body), string(result), "body should have been modified")
+			}
+
+			if tt.wantModel != "" {
+				// Parse result and check model field
+				model, _ := NewModelExtractor().Extract(result, nil)
+				assert.Equal(t, tt.wantModel, model)
+			}
+		})
+	}
+}
+
+func TestModelRewriter_WrapResponseWriter(t *testing.T) {
+	rewriter := NewModelRewriter()
+
+	t.Run("response writer wraps without error", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+		require.NotNil(t, wrapped)
+		require.NotNil(t, cleanup)
+		defer cleanup()
+	})
+
+	t.Run("rewrites model in non-streaming response", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		// Write a response with the resolved model
+		response := []byte(`{"model":"claude-local","content":"hello"}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		_, err := wrapped.Write(response)
+		require.NoError(t, err)
+
+		// Cleanup triggers the rewrite
+		cleanup()
+
+		// Check the response was rewritten to the requested model
+		body := recorder.Body.Bytes()
+		assert.Contains(t, string(body), `"model":"gpt-4"`)
+		assert.NotContains(t, string(body), `"model":"claude-local"`)
+	})
+
+	t.Run("no-op when requested equals resolved", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "gpt-4")
+
+		response := []byte(`{"model":"gpt-4","content":"hello"}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		_, err := wrapped.Write(response)
+		require.NoError(t, err)
+
+		cleanup()
+
+		body := recorder.Body.Bytes()
+		assert.Contains(t, string(body), `"model":"gpt-4"`)
+	})
+
+	t.Run("rewrites modelVersion field", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		response := []byte(`{"modelVersion":"claude-local","content":"hello"}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		_, err := wrapped.Write(response)
+		require.NoError(t, err)
+
+		cleanup()
+
+		body := recorder.Body.Bytes()
+		assert.Contains(t, string(body), `"modelVersion":"gpt-4"`)
+	})
+
+	t.Run("handles streaming responses", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		// Set streaming content type
+		wrapped.Header().Set("Content-Type", "text/event-stream")
+
+		// Write SSE chunks with resolved model
+		chunk1 := []byte("data: {\"model\":\"claude-local\",\"delta\":\"hello\"}\n\n")
+		_, err := wrapped.Write(chunk1)
+		require.NoError(t, err)
+
+		chunk2 := []byte("data: {\"model\":\"claude-local\",\"delta\":\" world\"}\n\n")
+		_, err = wrapped.Write(chunk2)
+		require.NoError(t, err)
+
+		cleanup()
+
+		// For streaming, data is written immediately with rewrites
+		body := recorder.Body.Bytes()
+		assert.Contains(t, string(body), `"model":"gpt-4"`)
+		assert.NotContains(t, string(body), `"model":"claude-local"`)
+	})
+
+	t.Run("empty body handled gracefully", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		wrapped.Header().Set("Content-Type", "application/json")
+		// Don't write anything
+
+		cleanup()
+
+		body := recorder.Body.Bytes()
+		assert.Empty(t, body)
+	})
+
+	t.Run("preserves other JSON fields", func(t *testing.T) {
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		response := []byte(`{"model":"claude-local","temperature":0.7,"usage":{"prompt_tokens":10}}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		_, err := wrapped.Write(response)
+		require.NoError(t, err)
+
+		cleanup()
+
+		body := recorder.Body.Bytes()
+		assert.Contains(t, string(body), `"temperature":0.7`)
+		assert.Contains(t, string(body), `"prompt_tokens":10`)
+	})
+}
+
+func TestResponseRewriter_ImplementsInterfaces(t *testing.T) {
+	rewriter := NewModelRewriter()
+	recorder := httptest.NewRecorder()
+	wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+	defer cleanup()
+
+	// Should implement http.ResponseWriter
+	assert.Implements(t, (*http.ResponseWriter)(nil), wrapped)
+
+	// Should preserve header access
+	wrapped.Header().Set("X-Custom", "value")
+	assert.Equal(t, "value", recorder.Header().Get("X-Custom"))
+
+	// Should write status
+	wrapped.WriteHeader(http.StatusCreated)
+	assert.Equal(t, http.StatusCreated, recorder.Code)
+}
+
+func TestResponseRewriter_Flush(t *testing.T) {
+	t.Run("flush writes buffered content", func(t *testing.T) {
+		rewriter := NewModelRewriter()
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		response := []byte(`{"model":"claude-local","content":"test"}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		wrapped.Write(response)
+
+		// Before cleanup, response should be empty (buffered)
+		assert.Empty(t, recorder.Body.Bytes())
+
+		// After cleanup, response should be written
+		cleanup()
+		assert.NotEmpty(t, recorder.Body.Bytes())
+	})
+
+	t.Run("multiple flush calls are safe", func(t *testing.T) {
+		rewriter := NewModelRewriter()
+		recorder := httptest.NewRecorder()
+		wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+		response := []byte(`{"model":"claude-local"}`)
+		wrapped.Header().Set("Content-Type", "application/json")
+		wrapped.Write(response)
+
+		// First cleanup
+		cleanup()
+		firstBody := recorder.Body.Bytes()
+
+		// Second cleanup should not write again
+		cleanup()
+		secondBody := recorder.Body.Bytes()
+
+		assert.Equal(t, firstBody, secondBody)
+	})
+}
+
+func TestResponseRewriter_StreamingWithDataLines(t *testing.T) {
+	rewriter := NewModelRewriter()
+	recorder := httptest.NewRecorder()
+	wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+	wrapped.Header().Set("Content-Type", "text/event-stream")
+
+	// SSE format with multiple data lines
+	chunk := []byte("data: {\"model\":\"claude-local\"}\n\ndata: {\"model\":\"claude-local\",\"done\":true}\n\n")
+	wrapped.Write(chunk)
+
+	cleanup()
+
+	body := recorder.Body.Bytes()
+	// Both data lines should have model rewritten
+	assert.Contains(t, string(body), `"model":"gpt-4"`)
+	assert.NotContains(t, string(body), `"model":"claude-local"`)
+}
+
+func TestModelRewriter_RoundTrip(t *testing.T) {
+	// Simulate a full request -> response cycle with model rewriting
+	rewriter := NewModelRewriter()
+
+	// Step 1: Rewrite request body
+	originalRequest := []byte(`{"model":"gpt-4","messages":[{"role":"user","content":"hello"}]}`)
+	rewrittenRequest, err := rewriter.RewriteRequestBody(originalRequest, "claude-local")
+	require.NoError(t, err)
+
+	// Verify request was rewritten
+	extractor := NewModelExtractor()
+	requestModel, _ := extractor.Extract(rewrittenRequest, nil)
+	assert.Equal(t, "claude-local", requestModel)
+
+	// Step 2: Simulate response with resolved model
+	recorder := httptest.NewRecorder()
+	wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+	response := []byte(`{"model":"claude-local","content":"Hello! How can I help?"}`)
+	wrapped.Header().Set("Content-Type", "application/json")
+	wrapped.Write(response)
+	cleanup()
+
+	// Verify response was rewritten back
+	body, _ := io.ReadAll(recorder.Result().Body)
+	responseModel, _ := extractor.Extract(body, nil)
+	assert.Equal(t, "gpt-4", responseModel)
+}
+
+func TestModelRewriter_NonJSONBody(t *testing.T) {
+	rewriter := NewModelRewriter()
+
+	// Binary/non-JSON body should be returned unchanged
+	body := []byte{0x00, 0x01, 0x02, 0x03}
+	result, err := rewriter.RewriteRequestBody(body, "gpt-4")
+	require.NoError(t, err)
+	assert.Equal(t, body, result)
+}
+
+func TestModelRewriter_InvalidJSON(t *testing.T) {
+	rewriter := NewModelRewriter()
+
+	// Invalid JSON without model field should be returned unchanged
+	body := []byte(`not valid json`)
+	result, err := rewriter.RewriteRequestBody(body, "gpt-4")
+	require.NoError(t, err)
+	assert.Equal(t, body, result)
+}
+
+func TestResponseRewriter_StatusCodePreserved(t *testing.T) {
+	rewriter := NewModelRewriter()
+	recorder := httptest.NewRecorder()
+	wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+	wrapped.WriteHeader(http.StatusAccepted)
+	wrapped.Write([]byte(`{"model":"claude-local"}`))
+	cleanup()
+
+	assert.Equal(t, http.StatusAccepted, recorder.Code)
+}
+
+func TestResponseRewriter_HeaderFlushed(t *testing.T) {
+	rewriter := NewModelRewriter()
+	recorder := httptest.NewRecorder()
+	wrapped, cleanup := rewriter.WrapResponseWriter(recorder, "gpt-4", "claude-local")
+
+	wrapped.Header().Set("Content-Type", "application/json")
+	wrapped.Header().Set("X-Request-ID", "abc123")
+	wrapped.Write([]byte(`{"model":"claude-local"}`))
+	cleanup()
+
+	result := recorder.Result()
+	assert.Equal(t, "application/json", result.Header.Get("Content-Type"))
+	assert.Equal(t, "abc123", result.Header.Get("X-Request-ID"))
+}

internal/routing/router.go

@@ -0,0 +1,267 @@
+package routing
+
+import (
+	"sort"
+	"strings"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/thinking"
+)
+
+// Router resolves models to provider candidates.
+type Router struct {
+	registry      *Registry
+	modelMappings map[string]string      // normalized from -> to
+	oauthAliases  map[string][]string    // normalized model -> []alias
+}
+
+// NewRouter creates a new router with the given configuration.
+func NewRouter(registry *Registry, cfg *config.Config) *Router {
+	r := &Router{
+		registry:      registry,
+		modelMappings: make(map[string]string),
+		oauthAliases:  make(map[string][]string),
+	}
+
+	if cfg != nil {
+		r.loadModelMappings(cfg.AmpCode.ModelMappings)
+		r.loadOAuthAliases(cfg.OAuthModelAlias)
+	}
+
+	return r
+}
+
+// LegacyRoutingDecision contains the resolved routing information.
+// Deprecated: Will be replaced by RoutingDecision from types.go in T-013.
+type LegacyRoutingDecision struct {
+	RequestedModel string              // Original model from request
+	ResolvedModel  string              // After model-mappings
+	Candidates     []ProviderCandidate // Ordered list of providers to try
+}
+
+// Resolve determines the routing decision for the requested model.
+// Deprecated: Will be updated to use RoutingRequest and return *RoutingDecision in T-013.
+func (r *Router) Resolve(requestedModel string) *LegacyRoutingDecision {
+	// 1. Extract thinking suffix
+	suffixResult := thinking.ParseSuffix(requestedModel)
+	baseModel := suffixResult.ModelName
+
+	// 2. Apply model-mappings
+	targetModel := r.applyMappings(baseModel)
+
+	// 3. Find primary providers
+	candidates := r.findCandidates(targetModel, suffixResult)
+
+	// 4. Add fallback aliases
+	for _, alias := range r.oauthAliases[strings.ToLower(targetModel)] {
+		candidates = append(candidates, r.findCandidates(alias, suffixResult)...)
+	}
+
+	// 5. Sort by priority
+	sort.Slice(candidates, func(i, j int) bool {
+		return candidates[i].Provider.Priority() < candidates[j].Provider.Priority()
+	})
+
+	return &LegacyRoutingDecision{
+		RequestedModel: requestedModel,
+		ResolvedModel:  targetModel,
+		Candidates:     candidates,
+	}
+}
+
+// ResolveV2 determines the routing decision for a routing request.
+// It uses the new RoutingRequest and RoutingDecision types.
+func (r *Router) ResolveV2(req RoutingRequest) *RoutingDecision {
+	// 1. Extract thinking suffix
+	suffixResult := thinking.ParseSuffix(req.RequestedModel)
+	baseModel := suffixResult.ModelName
+	thinkingSuffix := ""
+	if suffixResult.HasSuffix {
+		thinkingSuffix = "(" + suffixResult.RawSuffix + ")"
+	}
+
+	// 2. Check for local providers
+	localCandidates := r.findLocalCandidates(baseModel, suffixResult)
+
+	// 3. Apply model-mappings if needed
+	mappedModel := r.applyMappings(baseModel)
+	mappingCandidates := r.findLocalCandidates(mappedModel, suffixResult)
+
+	// 4. Determine route type based on preferences and availability
+	var decision *RoutingDecision
+
+	if req.ForceModelMapping && mappedModel != baseModel && len(mappingCandidates) > 0 {
+		// FORCE MODE: Use mapping even if local provider exists
+		decision = r.buildMappingDecision(req.RequestedModel, mappedModel, mappingCandidates, thinkingSuffix, mappingCandidates[1:])
+	} else if req.PreferLocalProvider && len(localCandidates) > 0 {
+		// DEFAULT MODE with local preference: Use local provider first
+		decision = r.buildLocalProviderDecision(req.RequestedModel, localCandidates, thinkingSuffix)
+	} else if len(localCandidates) > 0 {
+		// DEFAULT MODE: Local provider available
+		decision = r.buildLocalProviderDecision(req.RequestedModel, localCandidates, thinkingSuffix)
+	} else if mappedModel != baseModel && len(mappingCandidates) > 0 {
+		// DEFAULT MODE: No local provider, but mapping available
+		decision = r.buildMappingDecision(req.RequestedModel, mappedModel, mappingCandidates, thinkingSuffix, mappingCandidates[1:])
+	} else {
+		// No local provider, no mapping - use amp credits proxy
+		decision = &RoutingDecision{
+			RouteType:     RouteTypeAmpCredits,
+			ResolvedModel: req.RequestedModel,
+			ShouldProxy:   true,
+		}
+	}
+
+	return decision
+}
+
+// findLocalCandidates finds local provider candidates for a model.
+func (r *Router) findLocalCandidates(model string, suffixResult thinking.SuffixResult) []ProviderCandidate {
+	var candidates []ProviderCandidate
+
+	for _, p := range r.registry.All() {
+		if !p.SupportsModel(model) {
+			continue
+		}
+
+		// Apply thinking suffix if needed
+		actualModel := model
+		if suffixResult.HasSuffix && !thinking.ParseSuffix(model).HasSuffix {
+			actualModel = model + "(" + suffixResult.RawSuffix + ")"
+		}
+
+		if p.Available(actualModel) {
+			candidates = append(candidates, ProviderCandidate{
+				Provider: p,
+				Model:    actualModel,
+			})
+		}
+	}
+
+	// Sort by priority
+	sort.Slice(candidates, func(i, j int) bool {
+		return candidates[i].Provider.Priority() < candidates[j].Provider.Priority()
+	})
+
+	return candidates
+}
+
+// buildLocalProviderDecision creates a decision for local provider routing.
+func (r *Router) buildLocalProviderDecision(requestedModel string, candidates []ProviderCandidate, thinkingSuffix string) *RoutingDecision {
+	resolvedModel := requestedModel
+	if thinkingSuffix != "" {
+		// Ensure thinking suffix is preserved
+		sr := thinking.ParseSuffix(requestedModel)
+		if !sr.HasSuffix {
+			resolvedModel = requestedModel + thinkingSuffix
+		}
+	}
+
+	var fallbackModels []string
+	if len(candidates) > 1 {
+		for _, c := range candidates[1:] {
+			fallbackModels = append(fallbackModels, c.Model)
+		}
+	}
+
+	return &RoutingDecision{
+		RouteType:      RouteTypeLocalProvider,
+		ResolvedModel:  resolvedModel,
+		ProviderName:   candidates[0].Provider.Name(),
+		FallbackModels: fallbackModels,
+		ShouldProxy:    false,
+	}
+}
+
+// buildMappingDecision creates a decision for model mapping routing.
+func (r *Router) buildMappingDecision(requestedModel, mappedModel string, candidates []ProviderCandidate, thinkingSuffix string, fallbackCandidates []ProviderCandidate) *RoutingDecision {
+	// Apply thinking suffix to resolved model if needed
+	resolvedModel := mappedModel
+	if thinkingSuffix != "" {
+		sr := thinking.ParseSuffix(mappedModel)
+		if !sr.HasSuffix {
+			resolvedModel = mappedModel + thinkingSuffix
+		}
+	}
+
+	var fallbackModels []string
+	for _, c := range fallbackCandidates {
+		fallbackModels = append(fallbackModels, c.Model)
+	}
+
+	// Also add oauth aliases as fallbacks
+	baseMapped := thinking.ParseSuffix(mappedModel).ModelName
+	for _, alias := range r.oauthAliases[strings.ToLower(baseMapped)] {
+		// Check if this alias has providers
+		aliasCandidates := r.findLocalCandidates(alias, thinking.SuffixResult{ModelName: alias})
+		for _, c := range aliasCandidates {
+			fallbackModels = append(fallbackModels, c.Model)
+		}
+	}
+
+	return &RoutingDecision{
+		RouteType:      RouteTypeModelMapping,
+		ResolvedModel:  resolvedModel,
+		ProviderName:   candidates[0].Provider.Name(),
+		FallbackModels: fallbackModels,
+		ShouldProxy:    false,
+	}
+}
+
+// applyMappings applies model-mappings configuration.
+func (r *Router) applyMappings(model string) string {
+	key := strings.ToLower(strings.TrimSpace(model))
+	if mapped, ok := r.modelMappings[key]; ok {
+		return mapped
+	}
+	return model
+}
+
+// findCandidates finds all provider candidates for a model.
+func (r *Router) findCandidates(model string, suffixResult thinking.SuffixResult) []ProviderCandidate {
+	var candidates []ProviderCandidate
+
+	for _, p := range r.registry.All() {
+		if !p.SupportsModel(model) {
+			continue
+		}
+
+		// Apply thinking suffix if needed
+		actualModel := model
+		if suffixResult.HasSuffix && !thinking.ParseSuffix(model).HasSuffix {
+			actualModel = model + "(" + suffixResult.RawSuffix + ")"
+		}
+
+		if p.Available(actualModel) {
+			candidates = append(candidates, ProviderCandidate{
+				Provider: p,
+				Model:    actualModel,
+			})
+		}
+	}
+
+	return candidates
+}
+
+// loadModelMappings loads model-mappings from config.
+func (r *Router) loadModelMappings(mappings []config.AmpModelMapping) {
+	for _, m := range mappings {
+		from := strings.ToLower(strings.TrimSpace(m.From))
+		to := strings.TrimSpace(m.To)
+		if from != "" && to != "" {
+			r.modelMappings[from] = to
+		}
+	}
+}
+
+// loadOAuthAliases loads oauth-model-alias from config.
+func (r *Router) loadOAuthAliases(aliases map[string][]config.OAuthModelAlias) {
+	for _, entries := range aliases {
+		for _, entry := range entries {
+			name := strings.ToLower(strings.TrimSpace(entry.Name))
+			alias := strings.TrimSpace(entry.Alias)
+			if name != "" && alias != "" && name != alias {
+				r.oauthAliases[name] = append(r.oauthAliases[name], alias)
+			}
+		}
+	}
+}

internal/routing/router_test.go

@@ -0,0 +1,115 @@
+package routing
+
+import (
+	"context"
+	"testing"
+
+	"github.com/router-for-me/CLIProxyAPI/v6/internal/config"
+	"github.com/router-for-me/CLIProxyAPI/v6/sdk/cliproxy/executor"
+	"github.com/stretchr/testify/assert"
+)
+
+// mockProvider is a test double for Provider.
+type mockProvider struct {
+	name           string
+	providerType   ProviderType
+	supportsModels map[string]bool
+	available      bool
+	priority       int
+}
+
+func (m *mockProvider) Name() string                             { return m.name }
+func (m *mockProvider) Type() ProviderType                       { return m.providerType }
+func (m *mockProvider) SupportsModel(model string) bool          { return m.supportsModels[model] }
+func (m *mockProvider) Available(model string) bool              { return m.available }
+func (m *mockProvider) Priority() int                            { return m.priority }
+func (m *mockProvider) Execute(ctx context.Context, model string, req executor.Request) (executor.Response, error) {
+	return executor.Response{}, nil
+}
+func (m *mockProvider) ExecuteStream(ctx context.Context, model string, req executor.Request) (<-chan executor.StreamChunk, error) {
+	return nil, nil
+}
+
+func TestRouter_Resolve_ModelMappings(t *testing.T) {
+	registry := NewRegistry()
+	
+	// Add a provider
+	p := &mockProvider{
+		name:           "test-provider",
+		providerType:   ProviderTypeOAuth,
+		supportsModels: map[string]bool{"target-model": true},
+		available:      true,
+		priority:       1,
+	}
+	registry.Register(p)
+	
+	// Create router with model mapping
+	cfg := &config.Config{
+		AmpCode: config.AmpCode{
+			ModelMappings: []config.AmpModelMapping{
+				{From: "user-model", To: "target-model"},
+			},
+		},
+	}
+	router := NewRouter(registry, cfg)
+	
+	// Resolve
+	decision := router.Resolve("user-model")
+	
+	assert.Equal(t, "user-model", decision.RequestedModel)
+	assert.Equal(t, "target-model", decision.ResolvedModel)
+	assert.Len(t, decision.Candidates, 1)
+	assert.Equal(t, "target-model", decision.Candidates[0].Model)
+}
+
+func TestRouter_Resolve_OAuthAliases(t *testing.T) {
+	registry := NewRegistry()
+	
+	// Add providers
+	p1 := &mockProvider{
+		name:           "oauth-1",
+		providerType:   ProviderTypeOAuth,
+		supportsModels: map[string]bool{"primary-model": true},
+		available:      true,
+		priority:       1,
+	}
+	p2 := &mockProvider{
+		name:           "oauth-2",
+		providerType:   ProviderTypeOAuth,
+		supportsModels: map[string]bool{"fallback-model": true},
+		available:      true,
+		priority:       2,
+	}
+	registry.Register(p1)
+	registry.Register(p2)
+	
+	// Create router with oauth aliases
+	cfg := &config.Config{
+		OAuthModelAlias: map[string][]config.OAuthModelAlias{
+			"test-channel": {
+				{Name: "primary-model", Alias: "fallback-model"},
+			},
+		},
+	}
+	router := NewRouter(registry, cfg)
+	
+	// Resolve
+	decision := router.Resolve("primary-model")
+	
+	assert.Equal(t, "primary-model", decision.ResolvedModel)
+	assert.Len(t, decision.Candidates, 2)
+	// Primary should come first (lower priority value)
+	assert.Equal(t, "primary-model", decision.Candidates[0].Model)
+	assert.Equal(t, "fallback-model", decision.Candidates[1].Model)
+}
+
+func TestRouter_Resolve_NoProviders(t *testing.T) {
+	registry := NewRegistry()
+	cfg := &config.Config{}
+	router := NewRouter(registry, cfg)
+	
+	decision := router.Resolve("unknown-model")
+	
+	assert.Equal(t, "unknown-model", decision.ResolvedModel)
+	assert.Empty(t, decision.Candidates)
+}