cfr-compliance-mcp
A production-quality MCP server that exposes the official eCFR API as structured tools for AI-driven contract compliance, enabling retrieval of relevant Code of Federal Regulations by clauses extracted from contracts.
README
cfr-compliance-mcp
A production-quality MCP (Model Context Protocol) server exposing the official eCFR (Electronic Code of Federal Regulations) API as structured tools, built for an AI-driven contract compliance pipeline.
Status: in active development. The MCP server's foundation, client, cache, and parsing layers are complete. Validation models, the 8 MCP tools, and the server entrypoint are still being built. See
HANDOFF/PROJECT_HANDOFF.mdfor the full current status.
What this is
This package is the MCP server component of a larger pipeline:
Contract (PDF/DOCX) → Clause Extraction → Agno Team/Agents → this MCP Server
→ Official eCFR REST API → Relevant CFR Retrieval → Compliance Reasoning
→ Clause-wise Compliance Report
It has no dependency on any third-party MCP server — it talks directly to the official, public, unauthenticated eCFR REST API (https://www.ecfr.gov). See HANDOFF/PROJECT_HANDOFF.md Section 9 and the research addendum for why.
Requirements
- Python 3.12
- uv for dependency management
Setup
git clone <this repo>
cd cfr-compliance-mcp
uv sync
cp .env.example .env # adjust settings if needed; defaults work out of the box
Running (once server.py exists)
uv run cfr-compliance-mcp
Project layout
src/cfr_compliance_mcp/
├── config.py, logging_config.py, exceptions.py, constants.py # foundation
├── clients/ # generic HTTP transport + eCFR-specific API client
├── cache/ # backend-agnostic caching (in-memory today, Redis-ready)
├── parsing/ # raw eCFR XML -> clean text + citation metadata
├── models/ # (pending) Pydantic request/response validation
├── tools/ # (pending) the 8 MCP tools
└── server.py # (pending) FastMCP entrypoint
Full architectural rationale is in HANDOFF/ARCHITECTURE.md. Chronological build history is in HANDOFF/PROJECT_PROGRESS.md. A single authoritative current-state reference (what's built, why, what's next) is HANDOFF/PROJECT_HANDOFF.md — read that first if you're picking this project up.
Testing
uv run pytest
(Test suite is not yet built — see HANDOFF/PROJECT_HANDOFF.md for status.)
Development conventions
- Full type hints,
strictmypy. rufffor linting (uv run ruff check .).- All configuration via
.env/config.Settings— neveros.environdirectly. - All logging via
logging_config.get_logger(__name__)— logs go to stderr only (the MCPstdiotransport uses stdout for the protocol itself; never write there). - See
HANDOFF/PROJECT_HANDOFF.mdSection 17 for the full conventions list.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.