Agentos MCP Server

Build and manage policy-compliant AI agents with safety enforcement and compliance checking

Local serverstdioPython

What is the Agentos MCP server?

Agentos MCP server is a locally run integration for AI assistants that speak the Model Context Protocol. Build and manage policy-compliant AI agents with safety enforcement and compliance checking.

What you get

1,680+Tests Passing 12Framework Integrations 170K+Combined Stars of Integrated Projects <0.1ms p99Governance Latency Benchmarks 9More Framework Proposals Under Review

Agent OS applies operating system concepts to AI agent governance. Instead of relying on prompts to enforce safety ("please don't do dangerous things"), it provides application-level middleware that intercepts and validates agent actions before execution.

What the assistant can call

Once Agentos is connected, these are the calls the assistant has available:

  • name — block-destructive-sql
  • Framework — Stars
  • Dify — 65K ⭐
  • LlamaIndex — 47K ⭐
  • LangGraph — 24K ⭐
  • OpenClaw — —
  • AutoGen — 54K ⭐
  • CrewAI — 44K ⭐
  • Haystack — 22K ⭐
  • smolagents — 25K ⭐
  • PydanticAI — 15K ⭐
  • LlamaGuard — Content classification

Setting it up

ai-agent-governance on PyPI is all you need. Most clients run it directly, so configuration is a few lines and a restart.

Choosing this one

Among the database access options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Agentos's toolset — name, Framework, Dify and 11 more — is a fair guide to whether it matches your workflow. It is maintained by imran-siddique; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Before you rely on it

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Agentos.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the agentos mcp server does with a few real requests.

Available tools

ToolWhat it does
nameblock-destructive-sql
FrameworkStars
Dify65K ⭐
LlamaIndex47K ⭐
LangGraph24K ⭐
OpenClaw
AutoGen54K ⭐
CrewAI44K ⭐
Haystack22K ⭐
smolagents25K ⭐
PydanticAI15K ⭐
LlamaGuardContent classification
RiskCoverage
LayerPackage

How to install the Agentos MCP server

{
  "mcpServers": {
    "agentos": {
      "command": "uvx",
      "args": ["ai-agent-governance"]
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Example prompts to try

  • Use Agentos to name.
  • Use Agentos to Framework.
  • Use Agentos to Dify.

Frequently asked questions

It connects Agentos to MCP-compatible AI assistants such as Claude and Cursor, exposing 14 tools (name, Framework, Dify, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Agentos directly.