Archestra Community Edition

Archestra Community Edition is a self-hosted AI platform for teams managing access to company tools and models. It gives coding assistants, internal agents, and staff chat a shared place to enforce policies and track activity.

• Model gateway: Route requests to approved models, keep provider API keys on the server, and set spending limits per person, team, or agent.
• MCP tools: Manage a shared registry and gateway for MCP servers, with controls over tool access, approvals, and network connections.
• Agent runtime: Run agents as managed, sandboxed workloads with scheduled, email, or webhook triggers, reusable skills, and sub-agent delegation.
• Staff workspace: Give non-technical users chat, shared projects, apps, and RAG knowledge bases for working with company information.
• Audit and monitoring: Trace model requests, tool calls, policy decisions, and costs, with OpenTelemetry and Prometheus support for existing monitoring systems.

Teams can keep clients such as Claude Code, Codex, Cursor, and n8n. The platform connects to self-hosted models and cloud providers including OpenAI, Anthropic, Azure OpenAI, Vertex AI, and Bedrock. Cloud model requests go to the selected provider even when Archestra runs on your infrastructure.

The community platform is free to self-host under AGPL-3.0, with Docker and Kubernetes deployment options. SSO, OIDC, RBAC, white-labeling, and advanced knowledge bases with access control require a separate enterprise license.