@reaatech/create-voice-agent

Status: Pre-1.0 — APIs may change in minor versions.

Scaffolding CLI that bootstraps a complete voice-agent-kit project — pipeline config, STT/TTS providers, telephony or WebRTC transport, an MCP client, and a ready-to-run server.

Usage

npm create @reaatech/voice-agent@latest my-agent
# or
npx @reaatech/create-voice-agent my-agent
# or
pnpm create @reaatech/voice-agent my-agent

Run with no project name for the interactive prompts, or pass flags to skip them.

Options

| Option | Description | |--------|-------------| | [project-name] | Directory to create the project in | | --quick | Non-interactive — use defaults | | --stt <provider> | STT provider (e.g. deepgram, openai, assemblyai) | | --tts <provider> | TTS provider (e.g. deepgram, elevenlabs, cartesia) | | --telephony <provider> | Telephony provider: twilio, telnyx, none | | --transport <type> | Transport: twilio or webrtc | | --mcp <endpoint> | MCP endpoint URL | | --skip-install | Don't install dependencies after scaffolding |

Example

npx @reaatech/create-voice-agent my-agent \
  --quick \
  --stt deepgram \
  --tts elevenlabs \
  --telephony twilio \
  --transport twilio

Related Packages

• @reaatech/voice-agent-core — Pipeline, session, latency
• @reaatech/voice-agent-simulator — Local dev runner

License

MIT