| # 🎂 MCP 1st Birthday Hackathon Submission | |
| ## EU AI Act Compliance Agent | |
| ### 🔗 Links | |
| | Resource | URL | | |
| | --------------- | ------------------------------------------------------------------------------- | | |
| | **Live Demo** | [HF Space](https://huggingface.co/spaces/MCP-1st-Birthday/eu-ai-act-compliance) | | |
| | **GitHub** | [Repository](https://github.com/your-org/eu-ai-act-compliance) | | |
| | **Social Post** | [Twitter/X Post](#) | | |
| --- | |
| ## 📋 Submission Details | |
| ### Tracks | |
| - ✅ **Track 1: Building MCP** - MCP Server with compliance tools | |
| - ✅ **Track 2: MCP in Action** - Agentic application with Gradio UI | |
| ### Tags | |
| ``` | |
| mcp, agents, eu-ai-act, compliance, legal-tech, gradio, | |
| track-1-mcp-servers, track-2-agentic-applications | |
| ``` | |
| --- | |
| ## 🎯 Project Overview | |
| The **EU AI Act Compliance Agent** helps organizations navigate the European Union's AI Act (Regulation 2024/1689) — the world's first comprehensive AI regulation framework. | |
| ### The Problem | |
| - 📋 **Complex Classification** — AI systems must be classified by risk level | |
| - 📝 **Documentation** — Extensive technical documentation required | |
| - 🔍 **Transparency** — Clear disclosure obligations | |
| - ⏰ **Tight Deadlines** — Phased implementation starting 2025 | |
| ### Our Solution | |
| Three MCP tools + AI Agent for automated compliance: | |
| | Tool | Purpose | | |
| | ----------------------- | ------------------------------- | | |
| | `discover_organization` | Research & profile organization | | |
| | `discover_ai_services` | Find & classify AI systems | | |
| | `assess_compliance` | Generate compliance assessment | | |
| --- | |
| ## 🛠️ Tech Stack | |
| - **Gradio 6** — Interactive web UI | |
| - **xAI Grok** — AI reasoning and tool calling | |
| - **Tavily AI** — Web research | |
| - **Model Context Protocol** — Tool integration | |
| --- | |
| ## ✨ Key Features | |
| ### Track 1: MCP Server | |
| 1. **Organization Discovery** — Real-time web research using Tavily | |
| 2. **AI System Classification** — Risk tiers per EU AI Act Annex III | |
| 3. **Compliance Assessment** — Gap analysis with documentation templates | |
| ### Track 2: AI Agent | |
| 1. **Conversational Interface** — Natural language interaction | |
| 2. **Tool Orchestration** — Intelligent multi-tool workflows | |
| 3. **Document Generation** — Ready-to-use compliance templates | |
| 4. **Real-time Streaming** — Progressive response display | |
| --- | |
| ## 📊 Demo Workflow | |
| ``` | |
| User: "Analyze OpenAI's EU AI Act compliance" | |
| Agent: | |
| ├── 🔧 discover_organization("OpenAI") | |
| │ └── ✅ Found: AI company, Expert maturity, Provider role | |
| ├── 🔧 discover_ai_services(orgContext) | |
| │ └── ✅ Found: 5 AI systems (2 high-risk, 3 limited-risk) | |
| ├── 🔧 assess_compliance(orgContext, servicesContext) | |
| │ └── ✅ Score: 65/100, 3 critical gaps identified | |
| └── 📝 Generated compliance report with recommendations | |
| ``` | |
| --- | |
| ## 🏆 Why This Matters | |
| 1. **Real Regulation** — EU AI Act is live, affecting millions of organizations | |
| 2. **Practical Tools** — Automates tedious compliance workflows | |
| 3. **Educational** — Helps understand complex legal requirements | |
| 4. **Actionable** — Generates usable documentation templates | |
| --- | |
| ## 👥 Team | |
| **Team EU Compliance** | |
| Building the future of AI governance 🇪🇺 | |
| **** | |
| <div align="center"> | |
| **Built with ❤️ for the MCP 1st Birthday Hackathon** | |
| 🎂 Happy 1st Birthday, MCP! 🎂 | |
| </div> | |