Archestra Docs
Archestra connects your agents to tools from MCP servers and context from your enterprise documents, then governs every model and tool call they make.
New here? Follow the Get Started guide to run Archestra locally and connect your first agent.
Core Capabilities
- Agents and Skills: Build autonomous agents with custom tools, knowledge, and skills. Run them in Chat, messaging apps like Slack and Microsoft Teams, or your own code over A2A.
- MCP Gateway & Servers: Expose tools to any MCP-compatible client through a unified, secure gateway, and run sandboxed or self-hosted MCP servers.
- LLM Proxy & Router: Centralize model access across OpenAI, Anthropic, Gemini, Bedrock, and Azure with virtual keys, cost tracking, and automatic failover.
- Knowledge (RAG): Give agents cited answers from connected enterprise sources like Confluence, Google Drive, and SharePoint with automatic source permission sync.
- Governance & Guardrails: Decide which tool calls an agent can make from what it has read, and keep an audit log of every tool and model call.
Explore
Get Started
Run Archestra, connect your clients, and see your first governed request
Agents
Build agents with instructions, tools, skills, subagents, and triggers
MCP
Add MCP servers once, then give agents and MCP clients their tools through governed gateways
LLM Proxy
See and control every model call your company makes, from coding agents to your own apps
Knowledge (RAG)
Give agents and MCP clients cited answers from connected sources and uploaded documents
Chat & Projects
Start conversations, delegate tasks, attach files, and collaborate with autonomous agents in real time
Admin
Deploy Archestra, manage access, and monitor your instance
Integrations
Connect your AI clients and frameworks to Archestra
Contributing
Propose changes and report bugs through GitHub issues, and report vulnerabilities privately
Reference
Look up environment variables, permissions, API endpoints, built-in MCP tools, and the Apps SDK