archestra

Archestra MCP server provides a centralized platform for running private MCPs with security-first controls, observability, and governance. The Docker-based quickstart pulls the official Archestra platform image and runs the container exposing a web UI and API endpoints for managing MCPs, tokens, API keys, and cost controls. Use the included observability, governance, and dynamic tool features to monitor usage, enforce data access policies, and plug in private MCPs for your organization. After starting the container, you can access the web UI and begin configuring MCPs, registering private MCPs, and leveraging the built-in chat-style interfaces for interacting with MCPs through the orchestrator.

Prerequisites:

• Docker installed on your host (Docker Desktop or Docker Engine).
• Sufficient privileges to run containers and expose ports.

Installation steps:

1. Pull the Archestra platform image: docker pull archestra/platform:latest

2. Run the container in quickstart mode (example with required ports and volumes): docker run -p 9000:9000 -p 3000:3000
   -e ARCHESTRA_QUICKSTART=true
   -v /var/run/docker.sock:/var/run/docker.sock
   -v archestra-postgres-data:/var/lib/postgresql/data
   -v archestra-app-data:/app/data
   archestra/platform

3. Verify the service is up by visiting:

   • UI: http://localhost:9000
   • API/docs: http://localhost:9000/api/docs (or the provided documentation URL in the UI)

4. (Optional) If you want to run a more customized setup, consult the Quickstart Guide linked in the README for environment variables and advanced docker run options.

Tips and considerations:

• The ARCHESTRA_QUICKSTART environment variable enables a simplified startup mode suitable for testing and demos. For production, use your normal deployment workflow and configure persistent storage and proper networking.
• Expose only the needed ports (default 9000 for UI and 3000 for API) and consider securing them behind a reverse proxy with TLS.
• When running with Docker, you can mount volumes to persist data (e.g., /var/lib/postgresql/data and /app/data).
• Use the private MCP registry and Kubernetes orchestration features described in the docs to manage MCP deployment at scale.
• Review environment variables and resource limits to tailor performance and security to your environment.