{
  "markdown": "<div align=\"center\">\n\n# ◆ Sovereign AI Act — MCP server\n\n**Give your AI agent a *trustworthy* EU AI Act expert.**\n\n[![npm version](https://img.shields.io/npm/v/sovereign-ai-act-mcp?color=4DD6E0&label=npm)](https://www.npmjs.com/package/sovereign-ai-act-mcp)\n[![npm downloads](https://img.shields.io/npm/dm/sovereign-ai-act-mcp?color=4DD6E0)](https://www.npmjs.com/package/sovereign-ai-act-mcp)\n[![license](https://img.shields.io/npm/l/sovereign-ai-act-mcp?color=C9A961)](./LICENSE)\n[![MCP](https://img.shields.io/badge/Model_Context_Protocol-compatible-4DD6E0)](https://modelcontextprotocol.io)\n\n</div>\n\nThis [Model Context Protocol](https://modelcontextprotocol.io) server exposes **Leo** — the deterministic classifier for **Regulation (EU) 2024/1689 (the EU AI Act), as amended by Regulation (EU) 2026/1744 (the Digital Omnibus on AI, in force 27 July 2026)** — as callable tools. Every answer is grounded **verbatim** in the official law. **Leo never guesses.**\n\nWhen a user asks an AI *\"is my AI system high-risk under the EU AI Act?\"*, the honest answer must cite the actual law — not a hallucination. This server lets your agent do exactly that.\n\n## 🛠️ Tools\n\n| Tool | What it does |\n|------|--------------|\n| `classify_ai_system` | Plain-language AI description → risk tier (prohibited / high-risk / limited / minimal) + the exact Annex III category and binding Articles. |\n| `lookup_article` | Verbatim text of any Article (1–113 of the 2024 text; amended articles are flagged on the site). |\n| `search_eu_ai_act` | Full-text search across Articles, Recitals and Annexes. |\n| `get_compliance_deadlines` | The application dates in force under Regulation (EU) 2026/1744 (statuses computed at call time) and the Article 99 fine tiers. |\n\n## 🚀 Install (Claude Desktop)\n\nAdd to your `claude_desktop_config.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"sovereign-ai-act\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"sovereign-ai-act-mcp\"]\n    }\n  }\n}\n```\n\nRestart Claude, then ask: *\"Is an AI that screens job applicants high-risk under the EU AI Act?\"* — Claude calls Leo and answers with the exact Articles.\n\nWorks the same in **Cursor**, **VS Code**, **Windsurf**, **Cline**, and any MCP-capable client.\n\n## ⚡ Try the API directly (no install, no key)\n\n```bash\ncurl -X POST https://www.regulatoryai.eu/api/classify \\\n  -H \"content-type: application/json\" \\\n  -d '{\"description\":\"AI that screens job applicants and ranks CVs\"}'\n```\n\n```json\n{\n  \"ok\": true,\n  \"tier\": \"high_risk\",\n  \"classifications\": [{\n    \"tier\": \"high_risk\",\n    \"rule\": \"Employment, recruitment & worker management (Annex III.4)\",\n    \"annex\": \"III.4\",\n    \"articles\": [6, 9, 10, 11, 13, 14, 15],\n    \"severity\": \"HIGH — conformity assessment + risk management + ...\"\n  }]\n}\n```\n\nAlso: `GET /api/article/{1-113}` · `GET /api/search?q=...`\n\n## 💎 Why trust it\n\n- **Verbatim law.** All 113 Articles, 180 Recitals and 13 Annexes loaded word-for-word from the EU Publications Office (CELEX 32024R1689).\n- **Deterministic.** A rule engine, not a guess — every verdict cites the Article that binds you.\n- **EU-sovereign.** Operated by Dominion Intelligence AB, hosted in the EU.\n- **24 languages.** Ask in your own language.\n\n## ⚙️ Config\n\n- `SOVEREIGN_API_BASE` — override the API base URL (default `https://www.regulatoryai.eu`).\n- No API key required.\n\n---\n\nIndicative classification grounded in the official text — **not legal advice**.\nPowered by [**Sovereign AI Act**](https://www.regulatoryai.eu/for-ai/) · © Dominion Intelligence AB\n",
  "bytes": 3512,
  "sha": "7ac515afcb72e9d621df6502e36733975caf33bfbc85f799f1c259f656803bcc",
  "repo_slug": "saidbazyar/sovereign-ai-act-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_eu_regulatoryai_sovereign_ai_act_mcp_cb90fb8b/readme"
}