mcpdirectory.dev
9,129 servers · updated 23 Sept

Sovereign Ai Act MCP server

community Active last commit 16 d 
{
  "mcpServers": {
    "sovereign-ai-act": {
      "command": "npx",
      "args": [
        "-y",
        "sovereign-ai-act-mcp@latest",
        "sovereign-ai-act-mcp"
      ]
    }
  }
}
file .mcp.json (in your project root)install npx -y sovereign-ai-act-mcpversion 1.2.0

What it does

This server exposes Leo, a deterministic classifier for the EU AI Act (Regulation 2024/1689 as amended), allowing AI agents to classify AI systems by risk tier and retrieve binding Articles verbatim from the law. It is for developers building AI applications that need accurate, lawful compliance information about the EU AI Act without hallucination.

  • Classify an AI system description to determine its risk tier (prohibited, high-risk, limited, minimal) and the exact Annex III category and Articles that apply.
  • Look up the full text of any specific Article (1–113) from the EU AI Act to answer compliance questions.
  • Search across all Articles, Recitals and Annexes to find relevant regulatory requirements for a given topic.

Tools this server exposes

ToolDoes
classify_ai_systemAccepts a plain-language AI system description and returns the risk tier, Annex III category, and binding Articles from the EU AI Act.
lookup_articleReturns the verbatim text of any Article (1–113) from the 2024 EU AI Act.
search_eu_ai_actPerforms full-text search across all Articles, Recitals and Annexes of the EU AI Act.
get_compliance_deadlinesReturns application dates in force under Regulation (EU) 2026/1744 and Article 99 fine tiers.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.