c6b04f302a
* fix(server): stop a steered prompt from deadlocking on a permission Claude parks its turn inside the blocking canUseTool callback while a plan approval or a tool prompt waits for an answer. The SDK stops to read its input stream there, so a message sent at the open card is queued and never read. The user must reject the card or interrupt the agent by hand, and then send the message a second time. Deny the pending permissions after the steer is admitted. The prompt is already in the SDK input queue when the permission callback returns, so Claude reads it in the same turn. Plan mode stays on and no turn restarts. One pass is not enough. A turn that made several tool calls opens its next request 10 to 300 milliseconds after the previous one is answered, so a single snapshot advances the batch by one and parks the turn again. The Claude adapter already tracks the steers the SDK has not read, so expose that as an optional AgentSession.hasUnreadSteer and answer each new request while that method reports the prompt as unread. The release stops the moment the prompt is read, so a permission the agent asks for after it acts on the message still reaches the user. Only the two entrypoints that carry a message from a person set the flag. A scheduled prompt, a child-finished notification, and an agent-to-agent prompt must not answer a card the user is looking at. Denial left no timeline row, so a denied plan disappeared from the transcript. Claude and Codex now push a plan_approval row with approved: false. For Codex the row comes from resolvePlanPermission, which is the path that dismissPendingPlanApprovals also uses. In the app, queue mode held the message until the turn ended, and a parked turn never ends. The composer now sends while a permission is pending, and the button label agrees with the submit path. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(server): keep a post-prompt permission out of the release The release loop answered the pending requests before it asked whether the agent had read the prompt. The agent can read the prompt during a poll interval and then open a permission request because of it. The next pass denied that request, so a decision that belongs to the user was made for them. The loop now asks first and answers second. The first pass stays unconditional, because a provider that does not report unread steers must still get one release. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(server): make steer permission clearing provider-owned Keep the daemon policy explicit while letting each provider queue the steer and release permissions that block it. This removes timing-based polling and preserves rejected plans as canonical timeline rows. * fix(server): preserve omitted steer options Only construct provider steer options when permission clearing is requested. System prompts otherwise keep the existing undefined options contract. --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: Mohamed Boudra <boudra.moha@gmail.com>
Voice Assistant
A voice-controlled terminal assistant that runs as a single local service.
Quick Start
# Install dependencies
npm install
# Copy environment variables
cp .env.example .env
# Edit .env and add your API keys (OpenAI, Deepgram)
# Run development servers
npm run dev
# Open browser to http://localhost:5173
Architecture
- Express Server (port 3000) - Serves API and built UI in production
- Vite Dev Server (port 5173) - Hot-reload React UI in development
- WebSocket (
/ws) - Real-time bidirectional communication - Agent - STT → LLM → TTS pipeline with terminal control
- Daemon - tmux-based terminal management (in-process)
Development
# Run both servers (recommended)
npm run dev
# Or run separately:
npm run dev:server # Express on port 3000
npm run dev:ui # Vite on port 5173
# Type checking
npm run typecheck
# Build for production
npm run build
# Start production server
npm start
Project Status
✅ Completed (Phases 1-2):
- Package setup and configuration
- Express server with WebSocket
- React UI with Vite
- WebSocket client with ping/pong testing
⏳ In Progress (Phase 3):
- Terminal control (tmux integration)
📋 Planned (Phases 4-9):
- LLM integration (OpenAI GPT-4)
- Agent orchestrator
- Speech-to-Text (Deepgram)
- Text-to-Speech (OpenAI)
- Audio streaming
- UI polish
See IMPLEMENTATION_PLAN.md for complete details.
Environment Variables
OPENAI_API_KEY=your-openai-key-here # GPT-4 and TTS
DEEPGRAM_API_KEY=your-deepgram-key-here # Streaming STT
STT_MODEL=whisper-1 # Optional: override to gpt-4o-transcribe, etc.
STT_CONFIDENCE_THRESHOLD=-3.0 # Optional: reject low-confidence clips
STT_DEBUG_AUDIO_DIR=.stt-debug # Optional: persist raw dictation audio for debugging
PASEO_HOME=~/.paseo # Runtime state directory (agents/, etc.)
PASEO_LISTEN=127.0.0.1:6767 # Listen address (host:port or /path/to/socket)
PASEO_HOME defaults to ~/.paseo and isolates runtime artifacts like agents/. PASEO_LISTEN controls the daemon listen address. For blue/green testing you can run a parallel server without touching production state:
PASEO_HOME=~/.paseo-blue PASEO_LISTEN=127.0.0.1:7777 npm run dev
Tech Stack
- Server: Express, TypeScript, ws (WebSocket)
- Client: React 18, Vite, TypeScript
- Terminal: tmux (via child_process)
- AI: OpenAI (LLM + TTS), Deepgram (STT)
Testing
Currently manual testing via:
- Start servers:
npm run dev - Open http://localhost:5173
- Test WebSocket connection (green status indicator)
- Click "Send Ping" button to test communication
More testing guidance as features are implemented.
License
MIT