283 Commits

• Merge branch 'openai-codex'
  Mario Zechner · 2026-01-05 05:33:48 +01:00
  9a147559c0
• Add extensions option to createAgentSession SDK

  - Accept ExtensionFactory[] for inline extensions (merged with discovery)
  - Mark preloadedExtensions as @internal (CLI implementation detail)
  - Update sdk.md with inline extension example
  - Update CHANGELOG

  Mario Zechner · 2026-01-05 03:38:56 +01:00
  79cb8f0906
• Add customTools option back to createAgentSession SDK

  - Accepts ToolDefinition[] directly (simplified from old { path?, tool } format)
  - Tools are combined with extension-registered tools
  - Updated sdk.md documentation
  - Updated CHANGELOG

  Mario Zechner · 2026-01-05 03:34:36 +01:00
  8da793b1ba
• Merge hooks and custom-tools into unified extensions system (#454)

  Breaking changes:
  - Settings: 'hooks' and 'customTools' arrays replaced with 'extensions'
  - CLI: '--hook' and '--tool' flags replaced with '--extension' / '-e'
  - API: HookMessage renamed to CustomMessage, role 'hookMessage' to 'custom'
  - API: FileSlashCommand renamed to PromptTemplate
  - API: discoverSlashCommands() renamed to discoverPromptTemplates()
  - Directories: commands/ renamed to prompts/ for prompt templates
  
  Migration:
  - Session version bumped to 3 (auto-migrates v2 sessions)
  - Old 'hookMessage' role entries converted to 'custom'
  
  Structural changes:
  - src/core/hooks/ and src/core/custom-tools/ merged into src/core/extensions/
  - src/core/slash-commands.ts renamed to src/core/prompt-templates.ts
  - examples/hooks/ and examples/custom-tools/ merged into examples/extensions/
  - docs/hooks.md and docs/custom-tools.md merged into docs/extensions.md
  
  New test coverage:
  - test/extensions-runner.test.ts (10 tests)
  - test/extensions-discovery.test.ts (26 tests)
  - test/prompt-templates.test.ts

  Mario Zechner · 2026-01-05 01:43:35 +01:00
  c6fc084534
• Implement extension discovery with package.json manifest support

  Discovery rules:
  1. extensions/*.ts or *.js - direct files
  2. extensions/*/index.ts or index.js - subdirectory with index
  3. extensions/*/package.json with pi field - load declared paths
  
  No recursion beyond one level. Complex packages use package.json manifest.
  
  Added PiManifest type for future theme/skill bundling support.
  
  17 tests covering all discovery scenarios.
  
  refs #454

  Mario Zechner · 2026-01-04 23:24:24 +01:00
  9794868b38
• Add unified extensions system (not wired up yet)

  New src/core/extensions/ directory with:
  - types.ts: merged types from hooks and custom-tools
  - loader.ts: single loader for extensions
  - runner.ts: ExtensionRunner for event emission
  - wrapper.ts: tool wrapping utilities
  - index.ts: exports
  
  Key changes from old system:
  - Single ExtensionAPI with registerTool() for LLM-callable tools
  - Tools use ExtensionContext (has UI access)
  - No onSession callback on tools (use pi.on events instead)
  
  refs #454

  Mario Zechner · 2026-01-04 22:51:11 +01:00
  2846c7d190
• Fix event bus async error handling, clear pending messages on session switch, improve SDK docs

  - event-bus.ts: await async handlers to catch errors properly
  - agent-session.ts: clear _pendingNextTurnMessages on newSession/switchSession/branch
  - sdk.ts: make eventBus first (required) param for discoverHooks/discoverCustomTools
  - docs/sdk.md: document eventBus sharing pattern for hook/tool communication

  Mario Zechner · 2026-01-04 22:04:50 +01:00
  be330fdd9c
• feat(coding-agent): add event bus for tool/hook communication (#431)

  * feat(coding-agent): add event bus for tool/hook communication
  
  Adds pi.events API enabling custom tools and hooks to communicate via
  pub/sub. Tools can emit events, hooks can listen. Shared EventBus instance
  created per session in createAgentSession().
  
  - EventBus interface with emit() and on() methods
  - on() returns unsubscribe function
  - Threaded through hook and tool loaders
  - Documented in hooks.md and custom-tools.md
  
  * fix(coding-agent): wrap event handlers to catch errors
  
  * docs: note async handler error handling for event bus
  
  * feat(coding-agent): add sendMessage to tools, nextTurn delivery mode
  
  - Custom tools now have pi.sendMessage() for direct agent notifications
  - New deliverAs: 'nextTurn' queues messages for next user prompt
  - Fix: hooks and tools now share the same eventBus (was isolated before)
  
  * fix(coding-agent): nextTurn delivery should always queue, even when streaming

  Nico Bailon · 2026-01-04 21:36:19 +01:00
  9c9e6822e3
• feat(hooks): expose setTitle to HookUIContext (#446)
  Aliou Diallo · 2026-01-04 20:55:52 +01:00
  5cb7af1ddc
• feat(ai): add OpenAI Codex OAuth + responses provider
  Ahmed Kamal · 2026-01-04 21:11:19 +02:00
  1650041a63
• fix(tools): wrap ALL registry tools with hooks, not just active ones

  wrappedToolRegistry was only containing activeToolsArray (4 tools).
  Now wraps all tools from the registry so hooks can enable any tool.

  Mario Zechner · 2026-01-04 18:47:57 +01:00
  31438fdf2a
• fix(tools): tool registry now contains ALL built-in tools

  - createAllTools() populates registry with all 7 built-in tools
  - --tools flag only sets initially active tools (default: read/bash/edit/write)
  - Hooks can enable any tool from registry via setActiveTools()
  - System prompt rebuilds with correct tool guidelines when tools change
  - Document tsx module resolution workaround in README

  Mario Zechner · 2026-01-04 18:44:41 +01:00
  2849623afc
• fix(hooks): add stack traces to hook errors, fix tools.ts theme bug

  - HookError now includes optional stack field
  - Hook error display shows stack trace in dim color below error message
  - tools.ts: create SettingsListTheme using the theme passed to ctx.ui.custom()
    instead of using getSettingsListTheme() which depends on global theme

  Mario Zechner · 2026-01-04 18:33:28 +01:00
  6390ff87ef
• feat(hooks): add systemPromptAppend to before_agent_start, full tool registry

  - before_agent_start handlers can return systemPromptAppend to dynamically
    append text to the system prompt for that turn
  - Multiple hooks' systemPromptAppend strings are concatenated
  - Multiple hooks' messages are now all injected (not just first)
  - Tool registry now contains ALL built-in tools (read, bash, edit, write,
    grep, find, ls) regardless of --tools flag
  - --tools only sets initially active tools, hooks can enable any via
    setActiveTools()
  - System prompt automatically rebuilds when tools change, updating tool
    descriptions and guidelines
  - Add pirate.ts example hook demonstrating systemPromptAppend
  - Update hooks.md with systemPromptAppend documentation

  Mario Zechner · 2026-01-04 18:21:26 +01:00
  e4dd21a3b2
• fix(hooks): deep copy messages in context event before passing to hooks

  The context event handler documentation promised a deep copy but the
  implementation passed the original array reference. This could cause
  hooks to accidentally mutate session messages.
  
  Uses structuredClone() for fast native deep copying.

  Mario Zechner · 2026-01-04 18:13:30 +01:00
  ddf8bfceee
• refactor: address PR feedback - merge setWidget, use KeyId for shortcuts

  1. Merge setWidget and setWidgetComponent into single overloaded method
     - Accepts either string[] or component factory function
     - Uses single Map<string, Component> internally
     - String arrays wrapped in Container with Text components
  
  2. Use KeyId type for registerShortcut instead of plain string
     - Import Key from @mariozechner/pi-tui
     - Update plan-mode example to use Key.shift('p')
     - Type-safe shortcut registration
  
  3. Fix tool API docs
     - Both built-in and custom tools can be enabled/disabled
     - Removed incorrect 'custom tools always active' statement
  
  4. Use matchesKey instead of matchShortcut (already done in rebase)

  Helmut Januschka · 2026-01-04 18:13:30 +01:00
  8ecb1d6c0b
• fix: use robust matchShortcut from TUI library

  - Add matchShortcut() function to @mariozechner/pi-tui
  - Handles Kitty protocol, legacy terminal sequences, and lock keys
  - Supports special keys (enter, tab, space, backspace, escape)
  - Replace custom implementation in interactive-mode.ts
  - Remove unused imports

  Helmut Januschka · 2026-01-04 18:13:29 +01:00
  c1d7f8d962
• fix: remove inline imports and debug logging

  - Convert all inline import() types to top-level imports
  - Remove debug console.error statements from plan-mode hook

  Helmut Januschka · 2026-01-04 18:13:29 +01:00
  277d7bd83b
• fix(hooks): fix ContextEventResult.messages type to AgentMessage[]

  - Was incorrectly typed as Message[] which caused filtered messages to be ignored
  - Context event filter in plan-mode hook should now properly remove stale [PLAN MODE ACTIVE] messages

  Helmut Januschka · 2026-01-04 18:13:29 +01:00
  274d4a6247
• feat(hooks): add setWidgetComponent for custom TUI components

  - New ctx.ui.setWidgetComponent(key, factory) method
  - Allows custom Component to render as widget without taking focus
  - Unlike custom(), widget components render inline above editor
  - Components are disposed when cleared or replaced
  - Falls back to no-op in RPC/print modes

  Helmut Januschka · 2026-01-04 18:13:29 +01:00
  ce88ebcd68
• refactor(hooks): address PR feedback

  - Rename getTools/setTools to getActiveTools/setActiveTools
  - Add getAllTools to enumerate all configured tools
  - Remove text_delta event (use turn_end/agent_end instead)
  - Add shortcut conflict detection:
    - Skip shortcuts that conflict with built-in shortcuts (with warning)
    - Log warnings when hooks register same shortcut (last wins)
  - Add note about prompt cache invalidation in setActiveTools
  - Update plan-mode hook to use agent_end for [DONE:id] parsing

  Helmut Januschka · 2026-01-04 18:13:29 +01:00
  4ecf3f9422
• feat(hooks): add text_delta event for streaming text monitoring

  - New text_delta hook event fires for each chunk of streaming text
  - Enables real-time monitoring of agent output
  - Plan-mode hook now updates todo progress as [DONE:id] tags stream in
  - Each todo item has unique ID for reliable tracking

  Helmut Januschka · 2026-01-04 18:13:28 +01:00
  d1eea3ac4e
• feat(hooks): add setWidget API for multi-line status displays

  - ctx.ui.setWidget(key, lines) for multi-line displays above editor
  - Widgets appear below 'Working...' indicator, above editor
  - Supports ANSI styling including strikethrough
  - Added theme.strikethrough() method
  - Plan-mode hook now shows todo list with checkboxes
  - Completed items show checked box and strikethrough text

  Helmut Januschka · 2026-01-04 18:13:28 +01:00
  dc44816051
• feat(coding-agent): add hook API for CLI flags, shortcuts, and tool control

  Hook API additions:
  - pi.getTools() / pi.setTools(toolNames) - dynamically enable/disable tools
  - pi.registerFlag(name, options) / pi.getFlag(name) - register custom CLI flags
  - pi.registerShortcut(shortcut, options) - register keyboard shortcuts
  
  Plan mode hook (examples/hooks/plan-mode.ts):
  - /plan command or Shift+P shortcut to toggle
  - --plan CLI flag to start in plan mode
  - Read-only tools: read, bash, grep, find, ls
  - Bash restricted to non-destructive commands (blocks rm, mv, git commit, etc.)
  - Interactive prompt after each response: execute, stay, or refine
  - Shows plan indicator in footer when active
  - State persists across sessions

  Helmut Januschka · 2026-01-04 18:13:28 +01:00
  c956a726ed
• WIP: Add hook API for dynamic tool control with plan-mode hook example

  - Add pi.getTools() and pi.setTools(toolNames) to HookAPI
  - Hooks can now enable/disable tools dynamically
  - Changes take effect on next agent turn
  
  New example hook: plan-mode.ts
  - Claude Code-style read-only exploration mode
  - /plan command toggles plan mode on/off
  - Plan mode tools: read, bash, grep, find, ls
  - Edit/write tools disabled in plan mode
  - Injects context telling agent about restrictions
  - After each response, prompts to execute/stay/refine
  - State persists across sessions

  Helmut Januschka · 2026-01-04 18:13:28 +01:00
  059292ead1
• fix(coding-agent): resolved UTF-8 corruption in bash executor output (#433)

  - Fixed UTF-8 text corruption in bash executor by replacing Buffer.toString() with streaming TextDecoder.

  Can Bölük · 2026-01-04 03:15:50 +01:00
  6ddfd1be13
• Use same header color for Available Tools as System Prompt
  Mario Zechner · 2026-01-03 23:05:32 +01:00
  f49d2aac12
• Show first 10 lines of system prompt with click to expand for full content
  Mario Zechner · 2026-01-03 23:04:23 +01:00
  f9838bb5d4
• Add click-to-expand for system prompt in HTML export
  Mario Zechner · 2026-01-03 23:02:30 +01:00
  f5e38c4007
• Fix missing spacer between text and tool calls in HTML export
  Mario Zechner · 2026-01-03 23:00:35 +01:00
  f090ec214a
• Fix missing spacer between text and tool calls in HTML export
  Mario Zechner · 2026-01-03 22:56:36 +01:00
  e98524f1fc
• Merge branch 'feat/ctrl-n-external-editor' - configurable keybindings (#405)
  Mario Zechner · 2026-01-03 22:51:12 +01:00
  f2b89d5ec5
• Fix slash commands and hook commands during streaming

  - Hook commands now execute immediately during streaming (they manage their own LLM interaction via pi.sendMessage())
  - File-based slash commands are expanded and queued via steer/followUp during streaming
  - prompt() accepts new streamingBehavior option ('steer' or 'followUp') for explicit queueing during streaming
  - steer() and followUp() now expand file-based slash commands and error on hook commands
  - RPC prompt command accepts optional streamingBehavior field
  - Updated docs: rpc.md, sdk.md, CHANGELOG.md
  
  fixes #420

  Mario Zechner · 2026-01-03 15:36:54 +01:00
  e9cf3c1835
• feat(coding-agent): add $ARGUMENTS syntax for slash commands (#418)

  * feat(coding-agent): add $ARGUMENTS syntax for slash commands
  
  * test(coding-agent): add tests for slash command argument substitution

  Evgeniy Skuridin · 2026-01-03 14:08:39 +01:00
  8917a1f853
• feat: configurable keybindings for all editor and app actions

  All keybindings configurable via ~/.pi/agent/keybindings.json
  
  Editor actions:
  - cursorUp, cursorDown, cursorLeft, cursorRight
  - cursorWordLeft, cursorWordRight, cursorLineStart, cursorLineEnd
  - deleteCharBackward, deleteCharForward, deleteWordBackward
  - deleteToLineStart, deleteToLineEnd
  - newLine, submit, tab
  - selectUp, selectDown, selectConfirm, selectCancel
  
  App actions:
  - interrupt, clear, exit, suspend
  - cycleThinkingLevel, cycleModelForward, cycleModelBackward
  - selectModel, expandTools, toggleThinking, externalEditor
  
  Also adds support for numpad Enter key (Kitty protocol codepoint 57414
  and SS3 M sequence)
  
  Example emacs-style keybindings.json:
  {
    "cursorUp": ["up", "ctrl+p"],
    "cursorDown": ["down", "ctrl+n"],
    "cursorLeft": ["left", "ctrl+b"],
    "cursorRight": ["right", "ctrl+f"],
    "deleteCharForward": ["delete", "ctrl+d"],
    "cycleModelForward": "ctrl+o"
  }

  Helmut Januschka · 2026-01-03 08:23:56 +01:00
  8f2682578b
• Add shell commands without context contribution (!! prefix)

  Use !!command to execute bash commands that are shown in the TUI and
  saved to session history but excluded from LLM context, compaction
  summaries, and branch summaries.
  
  - Add excludeFromContext field to BashExecutionMessage
  - Filter excluded messages in convertToLlm()
  - Parse !! prefix in interactive mode
  - Use dim border color for excluded commands
  
  fixes #414

  Mario Zechner · 2026-01-03 04:14:35 +01:00
  746ec9eb01
• feat(coding-agent): configurable double-escape action (tree vs branch)

  Add doubleEscapeAction setting to choose whether double-escape with an
  empty editor opens /tree (default) or /branch.
  
  - Add setting to Settings interface and SettingsManager
  - Add to /settings UI for easy toggling
  - Update interactive-mode to respect the setting
  - Document in README.md settings table
  
  fixes #404

  Mario Zechner · 2026-01-03 01:59:08 +01:00
  42b1e06ad1
• Fix slow /model selector by deferring OAuth token refresh

  getAvailable() now uses hasAuth() which checks if auth is configured
  without triggering OAuth token refresh. Refresh happens later when
  the model is actually used.

  Mario Zechner · 2026-01-03 01:20:11 +01:00
  17b3a14bfa
• Add Vertex AI provider with ADC support

  - Implement google-vertex provider in packages/ai
  - Support ADC (Application Default Credentials) via @google/generative-ai
  - Add Gemini model catalog for Vertex AI
  - Update packages/coding-agent to handle google-vertex provider

  Anton Kuzmenko · 2026-01-03 01:11:03 +01:00
  214e7dae15
• Enhance provider override to support baseUrl-only mode

  Builds on #406 to support simpler proxy use case:
  - Override just baseUrl to route built-in provider through proxy
  - All built-in models preserved, no need to redefine them
  - Full replacement still works when models array is provided

  Mario Zechner · 2026-01-03 01:06:08 +01:00
  d747ec6e23
• Allow models.json to override built-in providers (#406)

  * Allow models.json to override built-in providers
  
  When a provider is defined in models.json with the same name as a
  built-in provider (e.g., 'anthropic', 'google'), the built-in models
  for that provider are completely replaced by the custom definition.
  
  This enables users to:
  - Use custom base URLs (proxies, self-hosted endpoints)
  - Define a subset of models they want available
  - Customize model configurations for built-in providers
  
  Example usage in ~/.pi/agent/models.json:
  {
    "providers": {
      "anthropic": {
        "baseUrl": "https://my-proxy.example.com/v1",
        "apiKey": "ANTHROPIC_API_KEY",
        "api": "anthropic-messages",
        "models": [...]
      }
    }
  }
  
  * Refactor model-registry for readability
  
  - Extract CustomModelsResult type and emptyCustomModelsResult helper
  - Extract loadBuiltInModels method with clear skip logic
  - Simplify loadModels with destructuring and ternary
  - Reduce repetition in error handling paths
  
  * Refactor model-registry tests for readability
  
  - Extract providerConfig() helper to hide irrelevant model fields
  - Extract writeModelsJson() helper for file writing
  - Extract getModelsForProvider() helper for filtering
  - Move modelsJsonPath to beforeEach
  
  Reduces test file from 262 to 130 lines while maintaining same coverage.

  Yevhen Bobrov · 2026-01-03 00:59:59 +01:00
  243104fa18
• Merge pull request #402 from mitsuhiko/image-resize

  Added automatic image resizing

  Mario Zechner · 2026-01-03 00:39:26 +01:00
  e82af9da47
• feat(coding-agent): expose deliverAs option in hook sendMessage() API

  - pi.sendMessage(msg, options?) now accepts { triggerTurn?, deliverAs? }
  - deliverAs: 'steer' (default) or 'followUp' controls delivery timing
  - Update all mode handlers to pass options through
  - Update file-trigger example to use new API
  - Update CHANGELOG

  Mario Zechner · 2026-01-03 00:13:26 +01:00
  9d8230dfc6
• feat(coding-agent): complete steer()/followUp() migration

  - Update settings-manager with steeringMode/followUpMode (migrates old queueMode)
  - Update sdk.ts to use new mode options
  - Update settings-selector UI to show both modes
  - Add Alt+Enter keybind for follow-up messages
  - Update RPC API: steer/follow_up commands, set_steering_mode/set_follow_up_mode
  - Update rpc-client with new methods
  - Delete dead code: queue-mode-selector.ts
  - Update tests for new API
  - Update mom/context.ts stubs
  - Update web-ui example

  Mario Zechner · 2026-01-03 00:13:25 +01:00
  3ae02a6849
• feat(coding-agent): update AgentSession for steer()/followUp() API

  - Rename queueMessage to steer(), add followUp()
  - Split _pendingMessages into _steeringMessages and _followUpMessages
  - Update sendHookMessage to accept deliverAs option
  - Rename hasQueuedMessages to hasPendingMessages
  - Rename queuedMessageCount to pendingMessageCount
  - Update clearQueue() return type to { steering, followUp }
  - Update UI to show steering vs follow-up messages differently
  
  WIP: settings-manager, sdk, interactive-mode, rpc-mode still need updates

  Mario Zechner · 2026-01-03 00:13:25 +01:00
  58c423ba36
• Add guard against concurrent prompt() calls

  Agent.prompt() and Agent.continue() now throw if called while already
  streaming, preventing race conditions and corrupted state. Use
  queueMessage() to queue messages during streaming, or await the
  previous call.
  
  AgentSession.prompt() has the same guard with a message directing
  users to queueMessage().
  
  Ref #403

  Mario Zechner · 2026-01-02 22:02:24 +01:00
  5ef3cc90d1
• Added automatic image resizing
  Armin Ronacher · 2026-01-02 12:01:56 +01:00
  4a32af2532
• fix: export type guard functions at runtime, fixes #397
  Mario Zechner · 2026-01-02 10:37:28 +01:00
  c10d99d2de
• Fix edit tool failing on files with UTF-8 BOM

  Strip BOM before matching (LLM won't include invisible BOM in oldText),
  restore on write.
  
  Based on #394 by @prathamdby

  Mario Zechner · 2026-01-02 01:57:24 +01:00
  d9adf659ca
• fix(coding-agent): prevent full re-renders during write tool streaming

  Move line count from header to footer to avoid changing the first line
  during streaming, which was triggering full screen re-renders in the
  TUI's differential rendering logic.

  Mario Zechner · 2026-01-02 01:11:06 +01:00
  d51770a63d