Ross ROSS = Recommend OSS · open-source software intelligence for agents

archestra-ai/archestra

Enterprise AI Platform with guardrails, MCP registry, gateway & orchestrator observed · 2026-08-28

github.com/archestra-ai/archestra · homepage · TypeScript · NOASSERTION (other) observed · 2026-08-28

Health v2 · maintenance only

81/100

  • Activity 99
  • Release rhythm 87
  • Longevity 29

Flags: no_license

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 0
  • age_days: 414
  • days_rel: 8
  • days_push: 7
  • n_releases_24m: 268

Full methodology

Adoption not part of the score

4232 stars · 1182 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Archestra is an open-source, self-hosted enterprise AI platform combining an agentic chat UI, agent runtime, LLM gateway/proxy, MCP gateway, private MCP registry, and Kubernetes-based MCP orchestrator. It ships with enterprise features like SSO/RBAC, prompt-injection-resistant guardrails, sandboxed code execution, OpenTelemetry tracing, and per-team cost tracking.

Use cases

  • self-host an internal ChatGPT-like assistant for my company
  • run MCP servers securely in Kubernetes with per-user identity
  • put a gateway in front of OpenAI, Anthropic, and other LLM providers with cost limits
  • give non-technical teams an AI chat with SSO and RBAC
  • host a private MCP registry for my team's tools
  • add guardrails against prompt injection to agent tool calls
  • proxy Claude Code or Cursor through one enterprise endpoint
  • build and schedule autonomous agents with MCP tools and triggers

When to choose

  • you need a centralized, self-hosted AI platform serving both engineers and non-technical staff
  • you want MCP gateway, registry, and orchestration with OAuth on-behalf-of semantics
  • enterprise requirements like SSO, RBAC, audit logs, observability, and cost tracking are mandatory
  • you want drop-in LLM/MCP proxies that work with LangChain, n8n, Vercel AI, or Pydantic AI

When to avoid

  • you only need a lightweight LLM API client or SDK in application code
  • you want a fully permissive-licensed project (it is AGPL 3.0 with enterprise licensing)
  • you need a minimal single-purpose tool rather than a full multi-component platform
  • you cannot run Kubernetes or containerized infrastructure for the orchestrator

Facets

application · maturity active

agent-framework mcp rag chatbot chat-interface api-gateway auth authorization monitoring tracing security self-hosted workflow-automation scheduling webhook artificial-intelligence large-language-models chatbots self-hosted developer-tools security self-hosted cloud mcp-gateway mcp-registry mcp-orchestrator llm-gateway llm-proxy a2a agent-platform enterprise-ai guardrails sso rbac opentelemetry prometheus kubernetes-operator agentic-chat agent-skills cost-tracking ai-agents retrieval-augmented-generation devops docker kubernetes web-server typescript

10 sources

Member repositories

RepositoryRoleHealth v2
archestra-ai/archestramain81

For agents

markdown · JSON · MCP: product_card(name="archestra-ai/archestra")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem