Skip to content
MCP server

Marrow AI: Control Layer for AI Agents

Policy check before risky agent actions, with proof recorded afterwards

Runs locallyStatus and history
Install
Pricing
Freemium
Platforms
CLI
Runs
Locally
Version
3.9.114
Listing updated
3 days ago

About Marrow AI: Control Layer for AI Agents

Marrow AI: Control Layer for AI Agents is a freemium MCP server for security. It runs locally on your machine and is started by your AI client.

Marrow sits between an agent and consequential actions such as a deploy, merge, publish or customer message. Before acting, the agent calls marrow_agent_runtime; on plans that enforce, it waits on review_required or block, and on other plans Marrow warns.

Afterwards marrow_commit closes the action with outcome and proof, so the next agent starts from what was verified. Approval comes from the operator at the host's own prompt or from the account owner, and locked categories require verified approval.

It needs a Marrow agent key kept in the environment, and an installer sets up hooks for Claude Code, Cursor and Codex. It checks only actions that pass through its hook, runner or MCP gate.

Install

Run in a terminal

claude mcp add marrow --env MARROW_API_KEY='<MARROW_API_KEY>' -- npx -y @getmarrow/mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "marrow": {
      "command": "npx",
      "args": [
        "-y",
        "@getmarrow/mcp"
      ],
      "env": {
        "MARROW_API_KEY": "<MARROW_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "marrow": {
      "command": "npx",
      "args": [
        "-y",
        "@getmarrow/mcp"
      ],
      "env": {
        "MARROW_API_KEY": "<MARROW_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "marrow": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@getmarrow/mcp"
      ],
      "env": {
        "MARROW_API_KEY": "<MARROW_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "marrow": {
      "command": "npx",
      "args": [
        "-y",
        "@getmarrow/mcp"
      ],
      "env": {
        "MARROW_API_KEY": "<MARROW_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "marrow": {
      "command": "npx",
      "args": [
        "-y",
        "@getmarrow/mcp"
      ],
      "env": {
        "MARROW_API_KEY": "<MARROW_API_KEY>"
      }
    }
  }
}

Configuration

  • MARROW_API_KEYNamed Marrow agent API key. Keep it in trusted secret storage.Required

Tools this server provides (2)

Taken from the project documentation; the running server may differ.

marrow_agent_runtime
Policy check called before a risky action.
marrow_commit
Closes an action with its outcome and proof.

What it does

Pre-action check
Policy review before risky steps.
Proof capture
Records outcome and evidence.
Locked categories
Require verified approval.
Client hooks
Installer writes hooks for Claude Code, Cursor and Codex.
Free tier
One agent and 500 decisions a month.

Good for

  • Gate production deploys by agents
  • Keep an audit of agent decisions
  • Require approval for customer messages
  • Share proven outcomes between agents

Pricing

Freemium, from $49 a month

Free; Team $49/mo; Business $199/mo

Prices change. Check getmarrow.ai before you buy.

Marrow AI: Control Layer for AI Agents FAQ

What does it cost?

Free covers 1 agent; Team is $49 a month and Business $199.

Does it cover every action?

Only those through its hook, runner or MCP gate.

Where do I keep the key?

In the environment or ~/.marrow/env, not config or chat.

Is Marrow AI: Control Layer for AI Agents free?

Marrow AI: Control Layer for AI Agents has a free plan, with paid plans for more usage or features. Paid plans start at $49 a month.

How do I add Marrow AI: Control Layer for AI Agents to Claude Code?

Run this command in your terminal: claude mcp add marrow --env MARROW_API_KEY='<MARROW_API_KEY>' -- npx -y @getmarrow/mcp

What are the best alternatives to Marrow AI: Control Layer for AI Agents?

The closest alternatives to Marrow AI: Control Layer for AI Agents are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Marrow AI: Control Layer for AI Agents

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools