DebugDruid/mcp_gdpr_checker
GDPR MCP Compliance Checker
A tool for checking whether Model Context Protocol (MCP) servers - and the companies or developers behind them - have published privacy policies, GDPR documentation, security information, and contact details. Uses real-time web search (via OpenAI) rather than a static database, so results reflect what's actually published right now.

Features
- Direct URL Analysis: Analyze MCP servers from GitHub, Smithery, MCP Market, MCP.so, or NPM URLs
- Real Web Search: Uses OpenAI's WebSearchTool to find privacy policies and GDPR documentation as they exist today, run concurrently for faster results
- Intelligent Context Extraction: Distinguishes the MCP's actual developer/company from the hosting platform (e.g. GitHub's own privacy policy is not attributed to every project merely hosted there)
- AI-Enhanced Analysis: Rule-based scoring plus AI-generated recommendations
- Email Templates: Ready-to-send templates for contacting MCP maintainers about gaps found
- Per-Session Privacy: Consent, rate limiting, and data summaries are isolated per visitor - no shared state between users
- Minimal Data Footprint: No database, no persistent storage; analysis runs in memory and is discarded when the session ends
How to Use
- Enter MCP URL: Paste a direct URL to an MCP server:
- GitHub:
https://github.com/owner/repo - Smithery:
https://server.smithery.ai/@owner/mcp/mcp - MCP Market:
https://mcpmarket.com/server/mcp-name - MCP.so:
https://mcp.so/server/mcp-name/owner - NPM:
https://www.npmjs.com/package/@company/mcp
- Check the consent box and click "Analyze MCP". This sends the MCP name/URL you enter to OpenAI for the search - see Privacy & GDPR Compliance below. A typical analysis takes roughly 30-60 seconds.
- Review Results:
- Privacy policies, GDPR documentation, security info, and contact details found
- Who's actually responsible under GDPR for the MCP you're evaluating vs. for deploying it yourself
- Email template for contacting maintainers
- Next steps and recommendations
Example URLs to Try
https://github.com/github/github-mcp-server
https://github.com/makenotion/notion-mcp-server
https://github.com/stripe/agent-toolkit
https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking
https://mcp.so/server/edgeone-pages-mcp/TencentEdgeOne
https://www.npmjs.com/package/@archbee/mcpRunning Locally
git clone https://github.com/CharleyRutledge/GDPR-MCP-compliance-checker.git
cd GDPR-MCP-compliance-checker
pip install -e ".[dev]" # installs runtime + test/lint dependencies
echo "OPENAI_API_KEY=sk-..." > .env # see Environment variables below
python app.py # serves on http://localhost:7860pip install -e . (without [dev]) installs just the runtime dependencies. requirements.txt is kept as a synced mirror of pyproject.toml's dependency list for tools that expect one.
Environment variables
Testing
pytest # full suite (228 tests)
black --check --diff . # formatting
isort --check-only --diff . # import order
flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics # syntax/critical errors (must pass)CI (.github/workflows/ci.yml) runs lint, tests, and a security scan on every push to main and every PR.
Privacy & GDPR Compliance of This Tool Itself
This tool practices what it checks for. In brief:
- Consent and session data are isolated per visitor - no shared file, no cross-user leakage.
- The OpenAI transfer is required, not optional: every analysis sends the MCP name/URL you enter to OpenAI for web search and AI analysis. The consent checkbox discloses this explicitly.
- No persistent storage: everything runs in memory and is discarded when your session ends, the server restarts, or you click "Revoke Consent."
- Full data subject rights are built in: "View My Data," "Export My Data" (in-memory Excel export, no server-side file), and "Revoke Consent" (erasure) are all live in the interface.
- Each processing purpose has one specific GDPR Art. 6(1) legal basis - see the full breakdown in `docs/privacy_policy.md`.
Read the complete policy at `docs/privacy_policy.md`, or click "View Privacy Policy" in the running app.
Important: whether an MCP's author published a privacy policy is a useful signal, but the organization that deploys an MCP server as part of its own AI agent - not the MCP's author - carries GDPR liability for whatever personal data flows through it. This tool provides guidance, not legal advice; consult a qualified professional for definitive compliance guidance.
Copyright
Copyright © 2025 GDPR MCP Compliance Checker. All Rights Reserved.
This software is proprietary. See LICENSE for full terms.
Support
For issues or questions, visit the GitHub repository.
