MCP Documentation Server
A server that provides organized documentation content for various applications using the Model Context Protocol, enabling AI assistants to access quickstart guides and code examples.
README
MCP Documentation Server
This project serves documentation for various applications using the Model Context Protocol (MCP). It demonstrates how to organize and expose documentation content, including quickstart guides and code examples, via MCP tools.
📚 Documentation Structure
Application documentation is organized under the content/ directory. For the XYZ application, content is located in content/xyz/ and includes:
quickstart.md: A quickstart guide for integrating with the XYZ application.code_examples.md: Various code examples demonstrating how to use the XYZ application with MCP tools.
🛠️ MCP Tools
MCP tools are defined in app/mcp/mcp_routes.py to expose the documentation content:
quickstart: Provides the quickstart guide for a specified application.code_examples: Provides code examples for a specified application.
⚙️ Setup Instructions
-
Initialize the environment
uv init uv venv source .venv/bin/activate uv pip install fastapi uvicorn openai fastmcp uvicorn app.main:app -
Start the application server
Run the provided start script:
chmod +x scripts/start.sh ./scripts/start.sh
Connect to MCP Server using Claude Desktop
To connect Claude Desktop to this MCP documentation server, you need to add a new server configuration in Claude's settings.
- Open Claude Desktop settings.
- Navigate to the 'Developer' section.
- Under 'MCP Server Configs', add the following JSON configuration:
{
"AppDocsServer": {
"command": "npx",
"args": [
"mcp-remote",
"http://localhost:8000/mcp-app-docs-server/mcp"
]
}
}
- Save the settings. Claude Desktop should now be able to discover and use the tools provided by this server.
Running the MCP Inspector
To test and inspect the MCP tools exposed by this server, you can use the MCP inspector tool. Run the following command in your terminal:
npx @modelcontextprotocol/inspector node build/index.js
This will start the inspector, allowing you to interact with the quickstart and code_examples tools provided by this server.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。