- Add onBeforeSend callback to ChatPanel and AgentInterface
- Add onBeforeToolCall callback (for future permission dialogs)
- Create NavigationMessage custom message type
- Add browserMessageTransformer that converts nav messages to <system> tags
- Track last submitted URL and tab index
- Auto-insert navigation message when URL/tab changes on user submit
- Navigation message shows favicon + page title in UI
- LLM receives: <system>Navigated to [title] (tab X): [url]</system>
- Implement CustomMessages interface for type-safe message extension via declaration merging
- Add MessageRenderer<T> with generic typing for custom message rendering
- Add messageTransformer to Agent for filtering/transforming messages before LLM
- Move message filtering from transports to Agent (separation of concerns)
- Add message renderer registry with typed role support
- Update web-ui example with SystemNotificationMessage demo
- Add custom transformer that converts notifications to <system> tags
- Add SessionListDialog onDelete callback for active session cleanup
- Handle non-existent session IDs in URL (redirect to new session)
- Update both web-ui example and browser extension with session fixes
Major architectural improvements:
- Renamed AgentSession → Agent (state/ → agent/)
- Removed id field from AgentState
- Fixed transport abstraction to pass messages directly instead of using callbacks
- Eliminated circular dependencies in transport creation
Transport changes:
- Changed signature: run(messages, userMessage, config, signal)
- Removed getMessages callback from ProviderTransport and AppTransport
- Transports now filter attachments internally
Session storage:
- Added SessionRepository with IndexedDB backend
- Auto-save sessions after first exchange
- Auto-generate titles from first user message
- Session list dialog with search and delete
- Persistent storage permission dialog
- Browser extension now auto-loads last session
UI improvements:
- ChatPanel creates single AgentInterface instance in setAgent()
- Added drag & drop file upload to MessageEditor
- Fixed artifacts panel auto-opening on session load
- Added "Drop files here" i18n strings
- Changed "Continue Without Saving" → "Continue Anyway"
Web example:
- Complete rewrite of main.ts with clean architecture
- Added check script to package.json
- Session management with URL state
- Editable session titles
Browser extension:
- Added full session storage support
- History and new session buttons
- Auto-load most recent session on open
- Session titles in header
- Import Select component from mini-lit and Brain icon from lucide
- Add thinkingLevel property to track current thinking level (off/minimal/low/medium/high)
- Enable showThinkingSelector (was disabled before)
- Add onThinkingChange callback for thinking level changes
- Render Select component with thinking level options when model supports reasoning
- Position thinking selector on left side next to attachment button
- Add i18n translations for Off/Minimal/Low/Medium/High in English and German
- Rename showThinking → showThinkingSelector in AgentInterface and ChatPanel for consistency
- Remove showDebugToggle property from AgentInterface (unused)
- Clean up browser-javascript tool output message (remove CSP note)
Storage Architecture:
- New pluggable storage system with backends (LocalStorage, ChromeStorage, IndexedDB)
- SettingsRepository for app settings (proxy config, etc.)
- ProviderKeysRepository for API key management
- AppStorage with global accessors (getAppStorage, setAppStorage, initAppStorage)
Transport Refactoring:
- Renamed DirectTransport → ProviderTransport (calls LLM providers with optional CORS proxy)
- Renamed ProxyTransport → AppTransport (uses app server with user auth)
- Updated TransportMode: "direct" → "provider", "proxy" → "app"
CORS Proxy Integration:
- ProviderTransport checks proxy.enabled/proxy.url from storage
- When enabled, modifies model baseUrl to route through proxy: {proxyUrl}/?url={originalBaseUrl}
- ProviderKeyInput test function also honors proxy settings
- Settings dialog with Proxy tab (Switch toggle, URL input, explanatory description)
Anthropic Prompt Caching:
- System prompt cached with cache_control markers (both OAuth and regular API keys)
- Last user message cached to cache conversation history
- Saves 90% on input tokens for cached content (10x cost reduction)
Settings Dialog Improvements:
- Configurable tab system with SettingsTab base class
- ApiKeysTab and ProxyTab as custom elements
- Switch toggle for proxy enable (instead of Checkbox)
- Explanatory paragraphs for each tab
- ApiKeyPromptDialog reuses ProviderKeyInput component
Removed:
- Deprecated ApiKeysDialog (replaced by ProviderKeyInput in SettingsDialog)
- Old storage-adapter and key-store (replaced by new storage architecture)
Major changes:
- Migrate browser-extension to use web-ui package (85% code reduction)
- Add JailJS content script with ES6+ transform support
- Expose DOM constructors (Event, KeyboardEvent, etc.) to JailJS
- Support top-level await by wrapping code in async IIFE
- Add returnFile() support in JailJS execution
- Refactor KeyStore into pluggable storage-adapter pattern
- Make ChatPanel configurable with sandboxUrlProvider and additionalTools
- Update jailjs to 0.1.1
Files deleted (33 duplicate files):
- All browser-extension components, dialogs, state, tools, utils
- Now using web-ui versions via @mariozechner/pi-web-ui
Files added:
- packages/browser-extension/src/content.ts (JailJS content script)
- packages/web-ui/src/state/storage-adapter.ts
- packages/web-ui/src/state/key-store.ts
Browser extension now has only 5 source files (down from 38).
- Created javascript-repl.ts tool with SandboxedIframe for code execution
- Integrated into ChatPanel replacing calculate and getCurrentTime tools
- REPL provides full browser API access with external library support
- Auto-registers renderer for tool output display
- Uses srcdoc instead of sandbox.html (no CSP restrictions in web apps)
Fixed loadContent() and execute() to set srcdoc at the correct time:
- loadContent(): Now sets srcdoc after creating iframe (was missing entirely)
- execute(): Now sets srcdoc after setting up message listeners and creating iframe (was setting before iframe existed)
This prevents missing postMessage events from the sandboxed iframe.
- Create Pi Reader browser extension for Chrome/Firefox
- Chrome uses Side Panel API, Firefox uses Sidebar Action API
- Supports both browsers with separate manifests and unified codebase
- Built with mini-lit components and Tailwind CSS v4
- Features model selection dialog with Ollama support
- Hot reload development server watches both browser builds
- Add useDefineForClassFields: false to fix LitElement reactivity
- Add 'aborted' as a distinct stop reason separate from 'error'
- Change AssistantMessage.error to errorMessage for clarity
- Update error event to include reason field ('error' | 'aborted')
- Map provider-specific safety/refusal reasons to 'error' stop reason
- Reorganize utility functions into utils/ directory
- Rename agent.ts to agent-loop.ts for better clarity
- Fix error handling in all providers to properly distinguish abort from error
- Added partial-json package for parsing incomplete JSON during streaming
- Tool call arguments now contain partially parsed JSON during toolcall_delta events
- Enables progressive UI updates (e.g., showing file paths before content is complete)
- Arguments are always valid objects (minimum empty {}), never undefined
- Full validation still occurs at toolcall_end when arguments are complete
- Updated all providers (Anthropic, OpenAI Completions/Responses) to use parseStreamingJson
- Added comprehensive documentation and examples in README
- Added test to verify arguments are always defined during streaming
- Switch from Zod to TypeBox for tool parameter schemas
- TypeBox schemas can be serialized/deserialized as JSON
- Use AJV for runtime validation instead of Zod's parse
- Add StringEnum helper for Google API compatibility (avoids anyOf/const patterns)
- Export Type and Static from main package for convenience
- Update all tests and documentation to reflect TypeBox usage
- Reduce default max tokens for Anthropic to 1/3 of model max
- Fix Google provider to properly handle empty tools array
- Ensure toolConfig is undefined when no tools are present
- Replace JSON Schema with Zod schemas for tool parameter definitions
- Add runtime validation for all tool calls at provider level
- Create shared validation module with detailed error formatting
- Update Agent API with comprehensive event system
- Add agent tests with calculator tool for multi-turn execution
- Add abort test to verify proper handling of aborted requests
- Update documentation with detailed event flow examples
- Rename generate.ts to stream.ts for clarity
- Add 'zai' as a KnownProvider type
- Add ZAI_API_KEY environment variable mapping
- Generate 4 zAI models (glm-4.5-air, glm-4.5v, etc.) using anthropic-messages API
- Add comprehensive test coverage for zAI provider in generate.test.ts and empty.test.ts
- Models support reasoning/thinking capabilities and tool calling
- Google provider sometimes returns duplicate or missing tool call IDs
- Added counter to ensure unique IDs for each tool call
- Check for duplicates and generate new ID when needed
- Fixes issues with multiple tool calls having the same ID
- Anthropic API requires tool call IDs to match pattern ^[a-zA-Z0-9_-]+$
- OpenAI Responses API generates IDs with pipe character (|) which breaks Anthropic
- Added sanitizeToolCallId() to replace invalid characters with underscores
- Fixes cross-provider handoffs from OpenAI Responses to Anthropic
- Added test to verify the fix works
- Replace createLLM with getModel/getModels/getProviders functions
- Rename PROVIDERS to MODELS (internal only, not exposed)
- Add streamSimple/completeSimple for unified reasoning interface
- Update README with new API examples and comprehensive documentation
- Remove model registration (models are now fixed from build time)
- Add proper TypeScript typing for provider-specific options
- Document context serialization, cross-provider handoffs, and browser usage
- Implement QueuedGenerateStream class that extends AsyncIterable with finalMessage() method
- Add new types: GenerateStream, GenerateOptions, GenerateOptionsUnified, GenerateFunction
- Create generateAnthropic function-based implementation replacing class-based approach
- Add comprehensive test suite for the new generate API
- Support streaming events with text, thinking, and tool call deltas
- Map ReasoningEffort to provider-specific options
- Include apiKey in options instead of constructor parameter
- Test handling of empty content arrays
- Test handling of empty string content
- Test handling of whitespace-only content
- All providers handle these edge cases gracefully
- Updated generate-models.ts to fetch these providers directly from models.dev API
- OpenRouter now only used for xAI and other third-party providers
- Fixed test model IDs to match new model names from models.dev
- Removed unused import from google.ts
- Add getApi() method to all providers to identify the API type
- Add api field to AssistantMessage to track which API generated it
- Update transformMessages to check API compatibility instead of model
- Fixes issue where OpenAI Responses API failed when switching models
- Preserves thinking blocks and signatures when staying within same API
- Required header for browser-based access to Anthropic API
- Added to both OAuth and regular API key authentication
- Ensures full browser compatibility
- Add transformMessages utility to handle cross-provider compatibility
- Convert thinking blocks to <thinking> tagged text when switching providers
- Preserve native thinking blocks when staying with same provider/model
- Add comprehensive handoff tests verifying all provider combinations
- Fix OpenAI Completions to return partial results on abort
- Update tool call ID format for Anthropic compatibility
- Document cross-provider handoff capabilities in README
- Anthropic, Google, and OpenAI Responses providers now return partial results when aborted
- Restructured streaming to accumulate content blocks incrementally
- Prevents submission of thinking/toolCall blocks from aborted completions in multi-turn conversations
- Makes UI development easier by providing partial content even when requests are interrupted
- Add inlineSources: true to embed TypeScript source in source maps
- Keeps source maps in separate files (inlineSourceMap: false)
- Enables proper debugging and breakpoints when using packages from npm
- Emit start event with model and provider info after creating stream
- Add abort signal tests for all providers
- Update README abort signal section to reflect non-throwing API
- Fix model references in README examples
- Add 'start' event with model/provider info
- Add 'done' event with stop reason and usage stats
- Add 'error' event handling
- Show more detailed event data in examples
- Add API changes section explaining v0.5.15+ changes
- Update Quick Start example to show content array usage
- Update Tool Calling example to filter tool calls from content blocks
- Update Streaming example to use new onEvent callback
- Fix model IDs in provider-specific examples
- Collect complete output items during streaming instead of building blocks incrementally
- Handle reasoning summary parts with proper newline separation
- Support refusal content in message outputs
- Preserve full reasoning items and message IDs for multi-turn resubmission
- Emit proper streaming events for text and thinking deltas
Mario Zechner