About eu-ai-act-compliance-mcp
An MCP server for teams assessing readiness for the EU AI Act, with provision-level measurement data and compliance tools. It covers risk classification, a 42-point audit, documentation support, and penalty estimates; the project states that measurement is not certification.
Key features
- 01Search 417 provisions
- 02Classify AI risk
- 03Run a 42-point audit
- 04Support Article 11 documentation
- 05Estimate potential penalties
Use cases
- 01Assess an AI system’s risk category
- 02Conduct a checklist-based regulatory audit
- 03Prepare documentation for Article 11
- 04Estimate potential penalty amounts
MCP support
The server exposes EU AI Act tools to MCP clients, including quick scans, deadline checks, and AI risk classification. It provides measurement, not compliance certification.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEVerification is free and public: signed measurement cards and live verification at https://councilof.ai — measurement, not certification.
Eu Ai Act Compliance MCP
⚖️ Need EU AI Act readiness for your system, fast? This MCP is the free tool. For a tailored readiness pack + a second opinion from the team behind the CSOAI charter,
Part of the MEOK governance platform · councilof.ai · csoai.org
EU AI Act measurement corpus — 417 frozen provisions at provision-level granularity (the Act has 113 Articles; our corpus is the finer-grained provision map)
EU AI Act measurement corpus — 417 frozen provisions at provision-level granularity (the Act has 113 Articles; our corpus is the finer-grained provision map). Risk classification, 42-point audit, Article 11 docs, penalty calculator. MIT
🚀 Quick Start
# Install via pip
pip install eu_ai_act_compliance_mcp
# Or install via Smithery
npx -y @smithery/cli@latest install eu-ai-act-compliance-mcp --client claude
⚡ Pay-per-call (PAYG) — no subscription
This MCP supports universal pay-per-call billing across the MEOK compliance fleet:
# One-time setup
export MEOK_PAYG_KEY="your_topup_token"
# Every tool call now deducts £0.05 from your balance.
# When balance hits zero, the tool returns a top-up URL.
# Works across all 7 MEOK compliance MCPs with the same token.
- No subscription — top up once, deduct per call.
- £0.05/call default (configurable via
MEOK_PAYG_RATE_GBP). - USDC on Base L2 accepted — set
MEOK_X402_RECEIVERand pay via stablecoin. - Backward-compatible — when
MEOK_PAYG_KEYis unset, behaviour is unchanged.
Get a token: councilof.ai/payg (£10 / £50 / £200 top-up tiers).
✨ Features
- MCP protocol compliant
- Easy installation
- Well-documented API
- Production-ready
- Active maintenance
📖 Documentation
Honesty register (2026-09-02)
Live authority is GET https://councilof.ai/api/gspc — cite totals.public_count and totals.lid from that payload. Do not type axis counts into this README.
- Measurement, not certification. CSOAI Ltd measures and signs; it does not certify, accredit, or act as a notified body.
- MEOK.ai is a shelf / frontend, not a second Council. Printers and agents must read the live GSPC endpoint — never invent scores, never wrangler a stamp.
- Prefer honest UNCHECKABLE over a fake stamp. This package does not certify EU AI Act compliance.
🏢 Enterprise
Need custom development, SLA guarantees, or white-label deployment?
View Pricing → | Contact Sales →
🤝 Part of the MEOK Ecosystem
This server is part of the Council of AI ecosystem — 26 PyPI packages · ~public measurement tools.
| Domain | Purpose |
|---|---|
| councilof.ai | Independent measurement board — live GET /api/gspc |
| safetyof.ai | AI safety & monitoring |
| councilof.ai | Shelf / frontend (not a second Council) |
| cobolbridge.ai | Legacy modernization |
📜 License
MIT © CSOAI-ORG
Built with 💜 by Council of AI · UK Companies House 16939677
Configuration
Add to your claude_desktop_config.json (Claude Desktop) or your MCP client config:
{
"mcpServers": {
"eu-ai-act-compliance-mcp": {
"command": "uvx",
"args": ["eu-ai-act-compliance-mcp"]
}
}
}
Or: pip install eu-ai-act-compliance-mcp then run the eu-ai-act-compliance-mcp command (stdio transport).
Examples
Once configured, ask your assistant, for example:
- "Use
quick_scanto …" - "Use
deadline_checkto …" - "Use
classify_ai_riskto …"
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.