Documentation
OpenAgent is a persistent AI agent framework. You run an Agent Server on a machine you control, then talk to it from the Desktop App or CLI — all pointing at the same agent, sharing the same memory and tools.
It's model-agnostic: Claude, GLM, Ollama, LM Studio, vLLM, or any OpenAI-compatible provider. Swap the model without losing memory or tool state. Run multiple independent agents side-by-side, each in its own folder.
Start here
- Getting Started — download, install, first run
- Configuration reference
Core concepts
- Invitation System & Networking — Iroh P2P transport, coordinator, device certs, invite tickets
- Models — pick a provider
- MCP tools — filesystem, editor, browser, web search, and more
- Gateway — WebSocket + REST surface over Iroh QUIC
- Channels — Telegram, Discord, WhatsApp
- Memory & vault — the markdown-based Obsidian vault
- Scheduler & Dream Mode — recurring tasks and background work
Clients & operations
- Desktop App
- Architecture — component map and system flow
- Deployment