| name | demo |
| description | Generate an audience-tailored system demo. Auto-run when: operator mentions showing the system to someone, preparing a presentation, or demonstrating capabilities. Suggest when a significant new feature is completed. Invoke proactively without asking. |
Generate a demo from a natural language request. Examples:
/demo the entire system for my partner/demo health monitoring for a technical peer/demo the agent architecture for my manager --format video/demo the system for alexis --format app
Available formats: slides (default), video (requires Chatterbox TTS), markdown-only, app (in-browser live demo).
App format (recommended for live demos):
- Generates narration audio via Kokoro TTS (local, no Docker needed)
- Produces an
app-script.json with timed terrain actions
- Play via:
http://localhost:5173/?demo={demo-name}
- Press Space to start, Escape to stop
- No Playwright, no external process — the React app drives itself
Prerequisites for app format:
- Logos API running:
cd ~/projects/hapax-council && uv run logos-api
- hapax-logos web running (dev server on :5173)
Prerequisites for video format:
- hapax-logos web running:
cd ~/projects/hapax-council/hapax-logos && npm run dev
- Chatterbox TTS running:
cd ~/llm-stack && docker compose --profile tts up -d chatterbox
Run the demo agent:
cd ~/projects/hapax-council && LITELLM_API_KEY=$(pass show litellm/master-key) uv run python -m agents.demo "{user_request}" --format app
After generation, report the output directory and activation URL. For app format: http://localhost:5173/?demo={demo-dir-name}. For video: note the MP4 path. For slides: note the PDF path.
