- Detect providers that don't support streaming with tools (DashScope, aliyuncs, bigmodel.cn)
- Add stream_from_complete() to use non-streaming mode when tools are present
- Fix convert_response() to prioritize tool_calls over empty content
- Fix ToolUse message JSON serialization (Null -> "{}")
- Skip invalid tool calls with empty names in streaming
Root cause: DashScope Coding Plan API doesn't support stream=true with tools,
causing tool parameters to be lost or malformed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Problem: Agent could not invoke appropriate skills when user asked about
financial reports because LLM didn't know which skills were available.
Root causes:
1. System prompt lacked available skill list
2. SkillManifest struct missing 'triggers' field
3. SKILL.md loader not parsing triggers list
4. "财报" keyword not matching "财务报告" trigger
Changes:
- Add triggers field to SkillManifest struct
- Parse triggers list from SKILL.md frontmatter
- Inject skill list into system prompt in kernel.rs
- Add "财报", "财务数据", "盈利", "营收" triggers to finance-tracker
- Add "财报分析" trigger to analytics-reporter
- Document fix in troubleshooting.md
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add ExecuteSkillTool for LLM to call skills during conversation
- Implement SkillExecutor trait in Kernel for skill execution
- Update AgentLoop to support tool execution with skill_executor
- Add default skills_dir configuration in KernelConfig
- Connect frontend skillMarketStore to backend skill_list command
- Update technical documentation with Skill system architecture
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Security Configuration:
- config/security.toml with shell_exec, file_read, file_write, web_fetch, browser, and mcp settings
- Command whitelist/blacklist for shell execution
- Path restrictions for file operations
- SSRF protection for web fetch
Tool Security Implementation:
- ShellSecurityConfig with whitelist/blacklist validation
- ShellExecTool with actual command execution
- Timeout and output size limits
- Security checks before command execution
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add BrowserHand implementation with:
- BrowserAction enum for all automation actions
- Navigate, Click, Type, Scrape, Screenshot, FillForm
- Wait, Execute (JavaScript), GetSource, GetUrl, GetTitle
- Scroll, Back, Forward, Refresh, Hover, PressKey, Upload
- Hand trait implementation with config and execute
- Integration with existing Tauri browser commands
Browser Hand enables agents to interact with web pages
for navigation, form filling, scraping, and automation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update chatStream method to use real Tauri event-based streaming:
- Add StreamChatEvent types matching Rust backend
- Set up Tauri event listener for 'stream:chunk' events
- Route events to appropriate callbacks (onDelta, onTool, onComplete, onError)
- Clean up listener on completion or error
- Remove simulated streaming fallback
This completes the frontend streaming integration for Chunk 4.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Implement run_streaming() method with async channel
- Stream chunks from LLM driver and emit LoopEvent
- Save assistant message to memory on completion
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add StreamChunk and StreamEvent types for Tauri event emission
- Add stream() method to LlmDriver trait with async-stream
- Implement Anthropic streaming with SSE parsing
- Implement OpenAI streaming with SSE parsing
- Add placeholder stream() for Gemini and Local drivers
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Config Persistence:
- Save reflection config to localStorage
- Load config on startup with fallback defaults
- Auto-sync config changes to backend
Proactive Personality Suggestions (P2):
- Add check_personality_improvement to heartbeat engine
- Detects user correction patterns (啰嗦/简洁, etc.)
- Add check_learning_opportunities to heartbeat engine
- Identifies learning opportunities from conversations
- Both checks generate HeartbeatAlert when thresholds met
These enhancements complete the self-evolution capability chain.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Initialize reflection engine with allow_soul_modification: true
- Sync config changes to backend when loading data
- Ensures reflection can generate identity change proposals
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
P1.1: Identity Change Proposal UI
- Create IdentityChangeProposal.tsx with diff view for SOUL.md changes
- Add approve/reject buttons with visual feedback
- Show evolution history timeline with restore capability
P1.2: Connect Reflection Engine to Identity Proposals
- Update ReflectionLog.tsx to convert reflection proposals to identity proposals
- Add ReflectionIdentityProposal type for non-persisted proposals
- Auto-create identity proposals when reflection detects personality changes
P1.3: Evolution History and Rollback
- Display identity snapshots with timestamps
- One-click restore to previous personality versions
- Visual diff between current and proposed content
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Remove temporary console.log and eprintln! statements added during
troubleshooting the model configuration issue.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Integrate react-window v2 List component for messages > 100
- Add VirtualizedMessageList and VirtualizedMessageRow components
- Use useVirtualizedMessages hook for dynamic height measurement
- Preserve smooth animations for small message counts (< 100)
- Auto-scroll to bottom for both virtualized and non-virtualized modes
Performance improvements:
- Only render visible messages in viewport
- Dynamic height measurement for accurate positioning
- LRU cache for message content
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Document completed tasks and file structure
- Record technical decisions and architecture changes
- Note verification results and next steps
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add type assertions for Valtio snapshot readonly arrays
- Remove unused fromPromise import from hands machine
- Ensures type compatibility with Valtio's useSnapshot
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Chat Domain:
- Add types.ts with Message, Conversation, Agent types
- Add store.ts with Valtio-based state management
- Add hooks.ts with useChatState, useMessages, etc.
- Add index.ts for public API export
Hands Domain:
- Add types.ts with Hand, Trigger, Approval types
- Add machine.ts with XState state machine
- Add store.ts with Valtio-based state management
- Add hooks.ts with useHands, useApprovalQueue, etc.
Shared Module:
- Add types.ts with Result, AsyncResult, PaginatedResponse
- Add error-handling.ts with AppError, NetworkError, etc.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add valtio for Proxy-based state management
- Add xstate and @xstate/react for state machines
- Create domains directory structure (chat, hands, intelligence, skills)
- Create shared directory for common utilities
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add SecurityError class for clear error handling
- Add validateWebSocketSecurity function
- Block ws:// connections to non-localhost hosts
- Add unit tests for security validation logic
Security: Prevents man-in-the-middle attacks on remote connections
by requiring WSS protocol for all non-localhost WebSocket connections.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Encrypt credentials before storing in localStorage when OS keyring unavailable
- Decrypt on retrieval with automatic fallback
- Backward compatible with existing unencrypted data (migration on next set)
- Add comprehensive unit tests (11 test cases)
Security: Credentials are now encrypted using AES-GCM when
OS keyring is unavailable, preventing plaintext exposure.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add test for base64 string output
- Add test for 32-byte key length
- Add test for uniqueness
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
