MCP Server (Mortgage Comparison Platform)

@confersolutions

リポジトリ

基本情報

カテゴリ

その他

ライセンス

MIT license

トランスポート

stdio

作者

confersolutions

MCP Server について

Canonical MCP server for parsing Loan Estimate (LE) and Closing Disclosure (CD) PDFs into MISMO-compliant JSON with LLM-enriched context. Built for AI-driven mortgage automation, compliance, and decisioning.

MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-mortgage-server": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        ".venv"
      ]
    }
  }
}

MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is MCP Server (Mortgage Comparison Platform)?

A FastAPI-based server that provides mortgage document parsing and comparison tools through a standardized API, designed to be integrated with AI frameworks such as CrewAI, AutoGen, and LangChain. Currently in beta (v0.1.0) with a basic "hello" tool as a proof of concept; mortgage parsing tools are planned for future releases.

How to use MCP Server (Mortgage Comparison Platform)?

Clone the repository, create a virtual environment, install dependencies (fastapi, uvicorn, slowapi, python-dotenv, plus optional framework packages), create a .env file with API_KEY, and run python server.py. Use the /health, /tools, and /call endpoints, authenticating with the X-API-Key header.

Key features of MCP Server (Mortgage Comparison Platform)

FastAPI server with API key authentication, rate limiting, and CORS
Integrations with CrewAI, AutoGen, and LangChain
Extensible architecture for adding mortgage parsing tools
Open source under MIT license, maintained by Confer Solutions
Currently provides a "hello" tool for testing framework integrations
Future roadmap includes Loan Estimate and Closing Disclosure parsing

Use cases of MCP Server (Mortgage Comparison Platform)

Testing AI framework integrations with the provided "hello" tool
Building AI agents that can call mortgage comparison APIs (when parsing tools are available)
Developing and extending open-source mortgage analysis tooling
Prototyping multi-agent mortgage workflows using supported frameworks

よくある質問

What does the server currently do?

It implements a basic "hello" tool as a proof of concept for testing integrations with CrewAI, AutoGen, and LangChain. Mortgage document parsing and comparison tools are planned but not yet available.

What runtime dependencies are required?

Python with FastAPI, uvicorn, slowapi, and python-dotenv. Integration frameworks (crewai, autogen, langchain) are optional and only needed if using those examples.

How is authentication handled?

API key authentication is required for all endpoints except `/health`. The key is sent via the `X-API-Key` header. Rate limiting is also implemented (default 120 requests per minute per IP).

Where is data stored?

The server does not store any data; it processes requests and returns responses in memory. No database or persistent storage is mentioned.

What AI frameworks are supported out of the box?

CrewAI, AutoGen, and LangChain. Example integration code is provided in the repository under `examples/test_all_integrations.py`.

「その他」の他のコンテンツ

Glasswarp

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov

Yuppsy

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code