This guide gets the current OpenDot platform, database, API, and local voice runtime running.Documentation Index
Fetch the complete documentation index at: https://docs.opendot.ai/llms.txt
Use this file to discover all available pages before exploring further.
1. Run with Docker Compose
From the repository root:.env before testing live voice sessions. For
now, only OPENAI_API_KEY and DEEPGRAM_API_KEY need to be added; the rest can
stay at the defaults. Create those keys from the
Deepgram Console and the
OpenAI platform.
Open:
http://localhost:8787. The default local run uses OpenDot email/password auth; Supabase Auth can be enabled later with SUPABASE_URL, VITE_SUPABASE_URL, VITE_SUPABASE_ANON_KEY, and a Supabase POSTGRES_URI.
2. Create and test an agent
- Open Agent Studio.
- Create a draft agent with a name and description.
- Open Configuration and review the default pipeline.
- Open Browser Test.
- Click Connect.
- Click Start mic and speak.
- Stop speaking and wait for the assistant text and audio chunks.
3. Optional pnpm workflow
Use this workflow when you want each service in its own terminal. From the repository root:.env before using the runtime:
4. Optional device check
Open Dot Device to claim a spoken device code, refresh runtime availability, and bind the selected agent config to the device. The default runtime endpoints are:Next steps
- Platform architecture explains the whole repo and runtime boundary.
- Configuration explains the VAD, STT, LLM, and TTS stage controls.
- Database inspection shows how to inspect PostgreSQL rows with Drizzle Studio.
- Authentication covers local email/password auth and Supabase Auth.
- Render deployment covers hosted preview deployment.