Bonsai BackendAPI Documentation

Design, deploy, and manage conversational AI experiences — voice and text, at scale.

Conversation Design

Build multi-stage conversational flows with agents, classifiers, context transformers, knowledge bases, and fully customisable actions.

REST API

Full-featured HTTP API for managing projects, agents, stages, classifiers, tools, knowledge, providers, users, and more.

Real-Time Channels

Two channel options for live AI sessions — WebSocket for maximum compatibility, WebRTC for lower audio latency. Both share the same conversation protocol.

Multi-Provider Support

Integrate with OpenAI, Anthropic, Google Gemini, Azure, ElevenLabs, Deepgram, Cartesia, and more — swap providers without changing your code.

Scripting & Extensibility

Execute sandboxed JavaScript actions, call external webhooks, and invoke LLM-powered tools from within any conversation stage.

Authentication & Permissions

JWT-based operator authentication with role-based access control (RBAC) and project-scoped API keys for WebSocket clients.