Free · No API key · Model Context Protocol
EU AI Act tools for your AI assistant
Add one URL to Claude, ChatGPT, Cursor or any MCP client, then ask “is this system high-risk under the EU AI Act?”. Your assistant calls the same classifier as this site and answers with the tier, the article and the documents you need.
Server URL (Streamable HTTP)
https://eurogpsr.com/api/mcp
Claude (web and desktop)
Settings → Connectors → Add custom connector, and paste the server URL.
Claude Code
claude mcp add --transport http eurocomply https://eurogpsr.com/api/mcp
Cursor, VS Code and other clients
{
"mcpServers": {
"eurocomply": { "url": "https://eurogpsr.com/api/mcp" }
}
}ChatGPT
Where custom connectors are available, add a new connector with the server URL and no authentication.
What it can do
Four tools
classify_ai_system
Risk tier (prohibited, high, limited, minimal), compliance score, findings with the legal article, required documents and a link to a saved report.
estimate_ai_act_fine
Maximum fine under Articles 99 and 101 for a violation type and turnover, with the SME lower-of rule.
generate_ai_disclosure_notice
Article 50 “you are talking to an AI” notices for chat, voice and generated content, in 8 EU languages.
list_ai_act_deadlines
Every application date after the Digital Omnibus, with who it affects, the legal basis and days left.
Things to ask
- “We rank job applicants with an LLM for clients in Germany. What does the AI Act require from us?”
- “What is our maximum fine for a missing chatbot disclosure with €2M turnover?”
- “Write the AI disclosure for our support bot in Spanish, German and French.”
Prefer a form? Use the free assessment, the fine calculator or the notice generator.
FAQ
Frequently asked questions
What is an MCP server?
The Model Context Protocol is an open standard that lets AI assistants call external tools. Adding this server gives your assistant four EU AI Act tools it can use while you talk to it.
Do I need an account or API key?
No. The server is public and free. It runs the same logic as the assessment and tools on this website.
What data does it receive?
Only the arguments your assistant sends: a handful of facts about the system (type, sector, a few yes/no answers) and, optionally, a company name. The classification is saved so you can open the report link. No conversation content is sent.
Is the result legal advice?
No. It is a triage signal that tells you which tier you are probably in and which documents that tier requires. Have high-risk systems reviewed by a qualified professional.
General information about Regulation (EU) 2024/1689, updated 4 October 2026. Not legal advice.