The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Gibs MCP listing page.
Regulatory compliance for AI-powered development tools — classify AI systems, check obligations, and get article-level citations across 240+ articles and 3 major regulations.
Full Documentation · Get API Key · API Reference
Gibs MCP Server connects your AI development environment directly to a regulatory compliance knowledge base. Currently covering:
| Regulation | Scope | Articles |
|---|---|---|
| EU AI Act | AI system classification, prohibited practices, obligations by risk level | 113 articles + 13 annexes |
| GDPR | Data protection, processing obligations, data subject rights | 99 articles |
| DORA | ICT risk management, incident reporting, third-party oversight for financial entities | 64 articles + 12 delegated/implementing acts |
Every response includes article-level citations to binding legal text, with real-time corpus updates as regulations evolve.
Sign up at gibs.dev and grab your API key from the dashboard.
Add this to your Claude Desktop configuration file (claude_desktop_config.json):
Pass your API key as the user_api_key parameter when calling any tool, or set it in your environment:
classify_ai_systemClassify an AI system under AI Act risk levels (unacceptable, high-risk, limited, minimal) with full legal reasoning.
| Parameter | Type | Required | Description |
|---|---|---|---|
description | string | Yes | What the AI system does (10–5000 chars) |
data_types | list[string] | No | Types of data processed (e.g., ["biometric", "personal"]) |
decision_scope | string | No | What decisions the system influences |
sector | string | No | Industry sector (e.g., "healthcare", "finance", "hr") |
Example prompt:
Classify my facial recognition attendance system for schools
What you get back:
check_complianceAsk any regulatory compliance question and get a structured answer with article-level citations.
| Parameter | Type | Required | Description |
|---|---|---|---|
question | string | Yes | Compliance question (10–2000 chars) |
system_description | string | No | Context about your AI system or organization |
regulation | string | No | Target regulation: "ai_act", "gdpr", "dora", or "both" (default: "both") |
Example prompts:
What are the GDPR requirements for a data breach notification?
Returns Article 33 (notification to supervisory authority) and Article 34 (communication to data subjects) obligations, including the 72-hour notification timeline, required content, and exceptions.
Does DORA apply to a fintech startup processing payments?
Returns Article 2 scope analysis with financial entity classification, applicable ICT risk management obligations, and relevant delegated act requirements.
health_checkCheck if the compliance API is available. No parameters required.
| SSE Endpoint | https://mcp.gibs.dev/sse |
| Protocol | Model Context Protocol (MCP) over Server-Sent Events |
| Authentication | API key from gibs.dev dashboard |
Any MCP-compatible client can connect to the Gibs server. Tested with:
pip install gibsMIT License. See LICENSE for details.
Built by Gibbr AB — making regulatory compliance accessible for developers.