# Archestra — Secure AI Platform (MCP Governance)

> Archestra centralizes MCP servers with governance and observability; verified 3,653★ and ships a Docker quickstart for a private MCP registry.

## Install

Merge the JSON below into your `.mcp.json`:

## Quick Use

```bash
docker pull archestra/platform:latest
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:latest
# Then follow the platform quickstart docs for first-login and registry setup.
```

## Intro

Archestra centralizes MCP servers with governance and observability; verified 3,653★ and ships a Docker quickstart for a private MCP registry.

**Best for:** Platform teams who need policy control, auditability, and safer MCP adoption org-wide

**Works with:** Docker (quickstart) and Archestra-managed MCP servers (self-hosted or third-party)

**Setup time:** 10-20 minutes

### Key facts (verified)

- GitHub: 3653 stars · 706 forks · pushed 2026-05-13.
- License: AGPL-3.0 · owner avatar + repo URL verified via GitHub API.
- README-backed entrypoint: `docker run -p 9000:9000 -p 3000:3000 archestra/platform:latest`.

## Main

- Use the Docker quickstart to validate the core loop first: platform UI up, then register 1 MCP server and prove it runs outside individual laptops.

- Treat governance as the product: define which MCP tools can access which credentials and data paths, and require reviews before adding new servers to the registry.

- Plan for observability from day one: record token/tool usage and keep an audit trail for high-risk MCP tools and prompt-injection scenarios (README links to observability + guardrails docs).

### Source-backed notes

- README provides a Docker quickstart with `ARCHESTRA_QUICKSTART=true` and binds ports 9000 and 3000.
- README positions the platform as a private MCP registry + Kubernetes-native orchestrator for running MCP servers and managing state/API keys/OAuth.
- README lists production readiness notes, including a latency benchmark callout ("45ms at 95p") and references Helm/Terraform deployment docs.

### FAQ

- **Is this an MCP server or a platform?**: A platform: it hosts/governs MCP servers and exposes a user-friendly toolbox around them.
- **Do I need Kubernetes on day one?**: No — README’s Docker quickstart is the easiest first validation; Kubernetes comes when you scale.
- **How do I evaluate security claims?**: Start with the guardrails docs linked in README and test with a safe prompt-injection lab before production.

## Source & Thanks

> Source: https://github.com/archestra-ai/archestra
> License: AGPL-3.0
> GitHub stars: 3653 · forks: 706

---

<!-- ZH -->

## Quick Use

```bash
docker pull archestra/platform:latest
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:latest
# Then follow the platform quickstart docs for first-login and registry setup.
```

## Intro

Archestra 将 MCP server 集中到企业级安全 AI 平台，提供私有 registry、guardrails 与可观测能力；已验证 3,653★，并含 Docker 快速启动（9000/3000）。

**Best for:** 需要策略治理、可审计与组织级 MCP 落地的平台团队

**Works with:** Docker 快速启动 + Archestra 托管的 MCP servers（自建或第三方）

**Setup time:** 10-20 minutes

### Key facts (verified)

- GitHub：3653 stars · 706 forks；最近更新 2026-05-13。
- 许可证：AGPL-3.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中可对照的入口命令：`docker run -p 9000:9000 -p 3000:3000 archestra/platform:latest`。

## Main

- 先用 Docker quickstart 跑通最小闭环：平台 UI 能打开，再注册 1 个 MCP server，确认它能脱离个人电脑稳定运行。

- 把“治理”当作核心价值：明确哪些 MCP 工具能访问哪些凭据与数据路径，并在把新 server 加入私有 registry 前要求审核。

- 从第一天就考虑可观测：记录 token/tool 使用并保留审计链路，特别是高风险工具与提示词注入场景（README 提供 observability 与 guardrails 文档入口）。

### Source-backed notes

- README 提供 Docker quickstart：设置 `ARCHESTRA_QUICKSTART=true`，并绑定 9000/3000 端口。
- README 将其定位为私有 MCP registry + Kubernetes 原生 orchestrator，用于运行 MCP server 并管理状态、API key 与 OAuth。
- README 给出生产化提示：包含延迟 benchmark 提示（“45ms at 95p”）以及 Helm/Terraform 部署文档链接。

### FAQ

- **它是 MCP server 还是平台？**：更像平台：用于集中托管/治理 MCP server，并提供统一入口与工具箱。
- **一开始就必须上 Kubernetes 吗？**：不必。README 的 Docker quickstart 适合先验证；规模化再上 K8s。
- **如何验证安全相关能力？**：从 README 的 guardrails 文档入手，先在安全的注入测试环境里验证，再考虑生产落地。

## Source & Thanks

> Source: https://github.com/archestra-ai/archestra
> License: AGPL-3.0
> GitHub stars: 3653 · forks: 706


---
Source: https://tokrepo.com/en/workflows/archestra-secure-ai-platform-mcp-governance
Author: Agent Toolkit