1108 Commits

• feat(caching): implement Claude prompt caching with multi-turn support

  - Add ensureCacheControl() to auto-inject cache breakpoints
  - Cache tools (last tool), system (last element), and messages (2nd-to-last user turn)
  - Add prompt-caching-2024-07-31 beta header
  - Return original payload on sjson error to prevent corruption
  - Include verification test for caching logic
  
  Enables up to 90% cost reduction on cached tokens.
  
  Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

  Martin Schneeweiss · 2026-01-29 22:59:33 +01:00
  3a43ecb19b
• Merge pull request #1276 from router-for-me/thinking

  feat(thinking): enable thinking toggle for qwen3 and deepseek models

  Luis Pater · 2026-01-28 11:16:54 +08:00
  9e5b1d24e8
• refactor: consolidate channel send logic with context-safe handlers

  Optimize channel operations by introducing reusable context-aware send functions (`send` and `sendErr`) across `wsrelay`, `handlers`, and `cliproxy`. Ensure graceful handling of canceled contexts during stream operations.

  Luis Pater · 2026-01-28 10:58:35 +08:00
  e93e05ae25
• feat(thinking): enable thinking toggle for qwen3 and deepseek models

  Fix #1245

  hkfires · 2026-01-28 09:54:05 +08:00
  c8c27325dc
• Merge pull request #1170 from BianBianY/main

  feat: optimization enable/disable auth files

  Luis Pater · 2026-01-28 09:34:35 +08:00
  bbb55a8ab4
• chore(registry): disable gemini 2.5 flash image preview model
  hkfires · 2026-01-27 18:33:13 +08:00
  88a0f095e8
• chore(registry): comment out rev19-uic3-1p model config
  hkfires · 2026-01-27 18:33:13 +08:00
  c65f64dce0
• feat(api): add management model definitions endpoint
  hkfires · 2026-01-27 18:33:12 +08:00
  d18cd217e1
• fix(gemini): force type to string for enum fields to fix Antigravity Gemini API error (Relates to #1260)
  Darley · 2026-01-27 11:14:08 +03:30
  decddb521e
• feat(auth): add support for request_retry and disable_cooling overrides

  Implement `request_retry` and `disable_cooling` metadata overrides for authentication management. Update retry and cooling logic accordingly across `Manager`, Antigravity executor, and file synthesizer. Add tests to validate new behaviors.

  Luis Pater · 2026-01-26 21:59:08 +08:00
  70897247b2
• refactor(runtime): implement retry logic for Antigravity executor with improved error handling and capacity management
  Luis Pater · 2026-01-26 06:22:46 +08:00
  2af4a8dc12
• feat(executor): add request_id to error logs

  Extract error.message from JSON error responses when summarizing error bodies for debug logs

  hkfires · 2026-01-25 21:31:46 +08:00
  f30ffd5f5e
• test(claude): update expectations for system message handling
  hkfires · 2026-01-25 08:31:29 +08:00
  5743b78694
• Merge pull request #1132 from XYenon/fix/gemini-models-displayname-override

  fix(gemini): preserve displayName and description in models list

  Luis Pater · 2026-01-25 03:40:04 +08:00
  2e6a2b655c
• Merge pull request #1179 from mallendeo/main

  fix(claude): skip built-in tools in OAuth tool prefix

  Luis Pater · 2026-01-25 03:31:58 +08:00
  cb47ac21bf
• Merge pull request #1183 from Darley-Wey/fix/api-align

  fix(api): enhance ClaudeModels response to align with api.anthropic.com

  Luis Pater · 2026-01-25 03:30:14 +08:00
  a1394b4596
• Merge branch 'upstream-main'
  Yang Bian · 2026-01-24 16:28:08 +08:00
  f7bfa8a05c
• fix(api): enhance ClaudeModels response to align with api.anthropic.com
  Darley · 2026-01-24 04:41:08 +03:30
  46c6fb1e7a
• fix(auth): improve antigravity token exchange errors
  hkfires · 2026-01-24 09:04:15 +08:00
  9f9fec5d4c
• fix(auth): validate antigravity token userinfo email
  hkfires · 2026-01-24 08:33:52 +08:00
  e95be10485
• fix(auth): correct antigravity oauth redirect and expiry
  hkfires · 2026-01-24 08:33:52 +08:00
  f3d58fa0ce
• refactor(auth): export Gemini constants and use in handler
  hkfires · 2026-01-24 08:33:52 +08:00
  8c0eaa1f71
• refactor(auth): export Codex constants and slim down handler
  hkfires · 2026-01-24 08:33:52 +08:00
  405df58f72
• refactor(api): slim down RequestAnthropicToken to use internal/auth
  hkfires · 2026-01-24 08:33:51 +08:00
  e7f13aa008
• refactor(auth): export Claude OAuth constants for reuse
  hkfires · 2026-01-24 08:33:51 +08:00
  7cb6a9b89a
• refactor(api): slim down RequestAntigravityToken to use internal/auth
  hkfires · 2026-01-24 08:33:51 +08:00
  9aa5344c29
• refactor(auth): extract Antigravity OAuth constants to internal/auth
  hkfires · 2026-01-24 08:33:51 +08:00
  c65407ab9f
• refactor(auth): implement Antigravity AuthService in internal/auth
  hkfires · 2026-01-24 08:33:51 +08:00
  9e59685212
• refactor(auth): replace sanitizeAntigravityFileName with antigravity.CredentialFileName
  hkfires · 2026-01-24 08:33:51 +08:00
  4a4dfaa910
• Fixed: #1077

  refactor(translator): improve tools handling by separating functionDeclarations and googleSearch nodes

  Luis Pater · 2026-01-24 05:51:11 +08:00
  0d6ecb0191
• fix(claude): skip built-in tools in OAuth tool prefix
  Mauricio Allende · 2026-01-23 21:29:39 +00:00
  f16461bfe7
• Merge pull request #1125 from NightHammer1000/dev

  Filter out Top_P when Temp is set on Claude

  Luis Pater · 2026-01-24 02:03:33 +08:00
  873d41582f
• Merge pull request #1137 from augustVino/fix/remove_empty_systemmsg

  fix(translator): ensure system message is only added if it contains c…

  Luis Pater · 2026-01-24 02:02:18 +08:00
  6fb7d85558
• fix(auth): normalize plan type filenames to lowercase
  hkfires · 2026-01-23 20:13:09 +08:00
  d5e3e32d58
• feat: optimization enable/disable auth files
  Yang Bian · 2026-01-23 18:03:09 +08:00
  c8620d1633
• feat(executor): apply payload rules using requested model
  hkfires · 2026-01-23 16:38:41 +08:00
  ecc850bfb7
• Merge pull request #1102 from aldinokemal/main

  feat(management): add PATCH endpoint to enable/disable auth files

  Chén Mù · 2026-01-23 09:05:24 +08:00
  19b4ef33e0
• fix(executor): adjust model-specific request payload
  hkfires · 2026-01-22 20:28:08 +08:00
  7ca045d8b9
• refactor(util): reorder gemini schema cleaner helpers
  hkfires · 2026-01-22 18:38:48 +08:00
  abfca6aab2
• test: align signature cache tests with cache behavior

  Co-Authored-By: Warp <agent@warp.dev>

  sowar1987 · 2026-01-22 17:12:47 +08:00
  9c2992bfb2
• refactor: reuse placeholder reason description

  Co-Authored-By: Warp <agent@warp.dev>

  sowar1987 · 2026-01-22 17:12:47 +08:00
  269a1c5452
• test: update signature cache tests

  Revert gemini translator changes for scheme A
  
  Co-Authored-By: Warp <agent@warp.dev>

  sowar1987 · 2026-01-22 17:12:47 +08:00
  22ce65ac72
• Fix Gemini function-calling INVALID_ARGUMENT by relaxing Gemini tool validation and cleaning schema
  sowar1987 · 2026-01-22 17:11:07 +08:00
  a2f8f59192
• fix(gemini): preserve displayName and description in models list

  Previously GeminiModels handler unconditionally overwrote displayName
  and description with the model name, losing the original values defined
  in model definitions (e.g., 'Gemini 3 Pro Preview').
  
  Now only set these fields as fallback when they are missing or empty.

  XYenon · 2026-01-22 15:19:27 +08:00
  8c7c446f33
• refactor(translator): enhance signature handling in Claude and Gemini requests, streamline cache usage and remove unnecessary tests
  hkfires · 2026-01-21 20:21:49 +08:00
  c8884f5e25
• refactor(cache, translator): refine signature caching logic and tests, replace session-based logic with model group handling
  Luis Pater · 2026-01-21 18:30:05 +08:00
  d9c6317c84
• fix(translator): ensure system message is only added if it contains content
  Vino · 2026-01-21 16:45:50 +08:00
  d29ec95526
• refactor(cache, translator): remove session ID from signature caching and clean up logic
  Luis Pater · 2026-01-21 13:37:10 +08:00
  ef4508dbc8
• refactor(translator): remove session ID logic from signature caching and associated tests
  Luis Pater · 2026-01-21 12:45:07 +08:00
  f775e46fe2
• refactor(cache): simplify signature caching by removing sessionID parameter
  Luis Pater · 2026-01-21 12:38:05 +08:00
  65ad5c0c9d