MCP.so
ログイン

Claude AI Documentation Assistant MCP Server

@XPE-7

基本情報

カテゴリ

AI とエージェント

トランスポート

stdio

作者

XPE-7

Claude AI Documentation Assistant MCP Server について

概要はまだありません

#claude-ai#documentation-assistant#mcp-server

Claude AI Documentation Assistant 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-server-xpe-7": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        "venv"
      ]
    }
  }
}

Claude AI Documentation Assistant 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 🤖 Claude AI Documentation Assistant 📚?

It is an MCP server that enhances Claude with documentation search capabilities for AI and ML libraries. It integrates with the Serper API for web search and the Claude Desktop app, targeting users who need fast, intelligent retrieval of technical documentation.

How to use 🤖 Claude AI Documentation Assistant 📚?

Clone the repository, create a virtual environment, install dependencies, and set the SERPER_API_KEY in a .env file. Start the server with python main.py, then connect the Claude Desktop app via the Tools section by adding a local tool.

Key features of 🤖 Claude AI Documentation Assistant 📚

  • Smart documentation search across multiple AI/ML libraries
  • Seamless integration with Claude’s reasoning capabilities
  • Intelligent web search using the Serper API
  • Fast response times optimized for quick retrieval
  • Extendable architecture to add more documentation sources

Use cases of 🤖 Claude AI Documentation Assistant 📚

  • Ask Claude to explain how to use FAISS with LangChain
  • Search and retrieve documentation for LlamaIndex or OpenAI
  • Look up API details or usage examples from supported libraries
  • Quickly find specific explanations without manual browsing

よくある質問

1

What dependencies are required?

Python 3.8 or higher, a Claude Pro subscription, a Serper API key, and the Claude Desktop application.

2

How do I connect to the MCP server?

Open the Claude Desktop app, go to Settings > Tools, click “Add Tool”, select “Connect to a local tool”, and follow the prompts to connect to the running server.

3

What documentation sources are supported?

Currently it supports LangChain (python.langchain.com/docs), LlamaIndex (docs.llamaindex.ai/en/stable), and OpenAI (platform.openai.com/docs). More can be added by editing the `docs_urls` dictionary in `main.py`.

4

What are common issues and how to fix them?

“Connection refused” usually means the server is not running. Timeout errors may indicate poor internet or need for increased timeout. API key issues require verifying the Serper key in the `.env` file.

5

How does the server handle data and searches?

When Claude invokes the `get_docs` tool, the server searches the web via the Serper API, scrapes relevant documentation pages, and returns the content to Claude for analysis. No documentation data is stored locally beyond the search results.

コメント

「AI とエージェント」の他のコンテンツ