Overview

Nist Rmf Ai MCP provides AI-powered tools via the Model Context Protocol (MCP).

Tools

Tool	Description
assess_risk_profile	Assess an AI system against the full NIST AI RMF 1.0 framework.
map_ai_impact	Map AI system impacts across people, organizations, and ecosystems.
generate_risk_controls	Generate NIST-aligned control recommendations for identified AI risks.
crosswalk_to_eu_ai_act	Map NIST AI RMF functions and subcategories to EU AI Act articles.
create_risk_report	Generate a complete NIST AI RMF compliance report in markdown.
check_trustworthy_characteristics	Evaluate AI system against NIST's 7 trustworthy AI characteristics.
predict_risk_neural	Neural network-based risk prediction that improves from every compliance check.
neural_insights	Get aggregate learning insights from the neural compliance model.
quick_scan	One-line system description to instant NIST AI RMF risk profile. No API key need
framework_overview	Returns the NIST AI RMF GOVERN/MAP/MEASURE/MANAGE structure. No parameters neede

Installation

pip install meok-nist-rmf-ai-mcp

Usage with Claude Desktop

Add to your Claude Desktop MCP config (claude_desktop_config.json):

{
  "mcpServers": {
    "nist-rmf-ai-mcp": {
      "command": "python",
      "args": ["-m", "meok_nist_rmf_ai_mcp.server"]
    }
  }
}

Usage with FastMCP

from mcp.server.fastmcp import FastMCP

# This server exposes 10 tool(s) via MCP
# See server.py for full implementation

License

MIT © MEOK AI Labs