Archestra
ActiveDescription
Enterprise AI Platform with guardrails, MCP registry, gateway and orchestrator — comprehensive AI agent governance and management.
Key Features
- Kubernetes-native MCP orchestrator — runs MCP servers in K8s managing state, API keys, and OAuth centrally
- Private MCP registry with governance — share self-hosted and third-party MCPs across your organization
- Non-probabilistic security engine preventing data exfiltration via tool guardrails and the Lethal Trifecta defense
- Dual-LLM security sub-agents that isolate dangerous tool responses to prevent prompt injection attacks
- Built-in RAG knowledge base without external vector databases, plus cost monitoring with up to 96% reduction via dynamic optimization
- Full observability with metrics, traces, and logs for per-org, per-agent, and per-team token and tool usage
Use Cases
💡 Centralizing MCP server management across an organization to eliminate per-machine configuration chaos
💡 Enforcing AI security guardrails to prevent data exfiltration from private repositories to public endpoints
💡 Deploying company-wide AI adoption with one-click MCP access for technical and non-technical users
💡 Monitoring and controlling AI costs across teams with per-agent and per-org limits and dynamic model switching
💡 Building production AI agent platforms with 45ms p95 latency, Terraform provider, and Helm chart support
Tags
Categories
Quick Start
Run `docker pull archestra/platform:latest && docker run -p 127.0.0.1:9000:9000 -p 127.0.0.1:3000:3000 -e ARCHESTRA_QUICKSTART=true -v /var/run/docker.sock:/var/run/docker.sock archestra/platform` to start the full platform locally. Access the UI at localhost:3000.