Skip to content
MCP server

GRC Lake

Open-source evidence lake that grades security controls for audits

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
self-hosted, CLI, Web
Runs
Locally
Version
0.3.26
Listing updated
3 days ago

About GRC Lake

GRC Lake is an open-source MCP server for security. It runs locally on your machine and is started by your AI client.

Compliance work usually ends in screenshots and spreadsheets. This project instead collects evidence from cloud, identity, endpoint and code systems, or reads OCSF tables you already keep in a security lake, and tests it against versioned control rules. Results trace back to source records and rule versions, and snapshots form a hash-chained ledger that can be re-verified.

Reusable safeguards map to several frameworks such as SOC 2, ISO 27001 and NIST, with reviewed and proposed mappings kept apart. Reviewed decisions and exportable auditor workpapers are retained alongside the results.

The MCP entry is a local stdio server installed from PyPI. The registry text says connectors are read-only, writes stay local, and only people approve. Environment variables choose the lake directory and a local or remote mode; remote mode uses an API URL and key. A synthetic demo runs with one uvx command, and a GitHub Action posts a readiness table to a CI job.

Install

Run in a terminal

claude mcp add grc-lake --env GRC_LAKE_API_KEY='<GRC_LAKE_API_KEY>' -- uvx grc-lake

Add to claude_desktop_config.json

{
  "mcpServers": {
    "grc-lake": {
      "command": "uvx",
      "args": [
        "grc-lake"
      ],
      "env": {
        "GRC_LAKE_API_KEY": "<GRC_LAKE_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "grc-lake": {
      "command": "uvx",
      "args": [
        "grc-lake"
      ],
      "env": {
        "GRC_LAKE_API_KEY": "<GRC_LAKE_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "grc-lake": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "grc-lake"
      ],
      "env": {
        "GRC_LAKE_API_KEY": "<GRC_LAKE_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "grc-lake": {
      "command": "uvx",
      "args": [
        "grc-lake"
      ],
      "env": {
        "GRC_LAKE_API_KEY": "<GRC_LAKE_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "grc-lake": {
      "command": "uvx",
      "args": [
        "grc-lake"
      ],
      "env": {
        "GRC_LAKE_API_KEY": "<GRC_LAKE_API_KEY>"
      }
    }
  }
}

Configuration

  • GRC_LAKE_LAKEPath to a local lake directory (default ./lake)
  • GRC_LAKE_MCP_MODElocal, remote or auto (default); use remote on shared machines
  • GRC_LAKE_API_URLGRC Lake API URL for remote mode
  • GRC_LAKE_API_KEYAPI key for remote mode

What it does

Traceable assessments
Links results to source records, rule versions and recorded mapping decisions.
Hash-chained snapshots
Stores evidence snapshots in a ledger you can re-verify.
OCSF support
Queries OCSF tables already sitting in your lake.
Multi-framework mapping
Maps one control to several frameworks, separating reviewed from proposed mappings.
CI readiness gate
A GitHub Action posts a framework readiness table and can gate merges.
Synthetic demo
Loads a fictional company so you can explore the console before connecting anything.

Good for

  • Show an auditor how a failing control was evidenced
  • Gate a merge on framework readiness
  • Reuse one set of safeguards across SOC 2 and ISO 27001
  • Give an agent read access to assessment evidence
  • Try the console on sample data before a pilot

Pricing

Open source

Open source and self-hosted; no price published

Prices change. Check koda-ai-studio.github.io before you buy.

GRC Lake FAQ

How do I run the demo?

Run uvx --from "grc-lake[server]" grc-lake demo; it loads a synthetic company into a temporary lake and opens the console on localhost:8787.

Is it self-hosted?

Yes. The README describes it as open-source and self-hosted, with Docker and source options.

How does the MCP server run?

Over stdio, installed with pip install "grc-lake[server,mcp]" and started as grc-lake-mcp.

Does the hosted demo touch my systems?

No. It is read-only, runs on recorded synthetic data and does not connect to your systems.

Is GRC Lake free?

Yes. GRC Lake is open source, so you can use and self-host it for free.

How do I add GRC Lake to Claude Code?

Run this command in your terminal: claude mcp add grc-lake --env GRC_LAKE_API_KEY='<GRC_LAKE_API_KEY>' -- uvx grc-lake

What are the best alternatives to GRC Lake?

The closest alternatives to GRC Lake are REA, Agent Roast, Midplane, Leash and Tinplate Edge AI Safety, Security & EU Compliance.

Alternatives to GRC Lake

  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. Agent Roast

    MCP server

    Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  3. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  4. Leash

    MCP server

    Review what your AI touched, approve risky actions and undo supported changes

    • 1 tool
    • Remote
    • Free
    • 1 GitHub star
    • Security
  5. SecureCloud

    MCP server

    SentinelOne and IT security catalog, prices, quotes and checkout links for Israel

    • 7 tools
    • Remote
    • Free
    • Security