Happy slice of mcp pizzaMCP.pizza
Fire in da houseTop Tip:Paying $100+ per month for Perplexity, MidJourney, Runway, ChatGPT and other tools is crazy - get all your AI tools in one site starting at $15 per month with Galaxy AI Fire in da houseCheck it out free

jiang-feishu-mcp

MCP.Pizza Chef: upJiang

jiang-feishu-mcp is an MCP server that connects Feishu cloud documents with AI clients like Cursor and Claude. It exposes Feishu document spaces and document lists through the Model Context Protocol, enabling seamless integration and real-time access to Feishu Docs within AI-enhanced workflows. It supports fetching all cloud document spaces and documents in specified spaces, facilitating direct interaction with Feishu Docs via standardized MCP communication.

Use This MCP server To

Fetch all Feishu cloud document spaces for AI clients Retrieve document lists from specific Feishu spaces Integrate Feishu Docs into AI copilots like Cursor Enable real-time document access in AI workflows Standardize Feishu Docs access via MCP protocol Host a local MCP server for Feishu document interaction

README

飞书文档 MCP

一个用于连接飞书文档的Model Context Protocol (MCP) 服务，可以在Cursor等AI客户端中直接访问飞书文档空间和文档列表。

功能特点

  • 支持获取飞书全部云文档空间列表
  • 支持获取指定空间的文档列表
  • 通过MCP协议与AI客户端无缝连接

什么是MCP

MCP (Model Context Protocol) 是一个将自定义服务与各种LLM客户端（如Claude、Cursor）无缝连接的协议。它允许AI直接调用和使用我们的服务功能。

  • MCP客户端：AI应用程序（如Claude Desktop或Cursor），负责发起请求并与服务器通信
  • MCP服务器：本项目，暴露飞书文档功能，通过标准化协议与客户端交互

安装与运行

从源码安装

# 克隆项目
git clone https://github.com/yourusername/jiang-feishu-mcp.git
cd jiang-feishu-mcp

# 安装依赖
pnpm install

# 启动服务
pnpm start

服务器将在本地7777端口启动。

配置

在使用前，您需要创建一个.env文件配置飞书应用凭证：

# 复制示例配置文件
cp .env.example .env

然后编辑.env文件，填入您的飞书应用凭证：

FEISHU_APP_ID=your_app_id_here
FEISHU_APP_SECRET=your_app_secret_here

# 可选：设置端口号 (默认为7777)
PORT=7777

# 可选：设置文档保存路径 (默认为 ./docs)
DOCS_SAVE_PATH=./docs

获取飞书应用凭证的步骤：

  1. 访问飞书开放平台并登录
  2. 创建一个企业自建应用
  3. 在应用详情页获取App ID和App Secret
  4. 确保开启了云文档相关权限（文档、表格、云空间读取权限）

在Cursor中使用

  1. 启动服务器：pnpm start
  2. 打开Cursor的设置页面
  3. 找到"Model Context Protocol"设置
  4. 添加新的MCP Server，URL填入：http://localhost:7777/mcp
  5. 保存设置

现在，您可以在Cursor中使用以下工具：

  • list-spaces: 列出所有文档空间
  • list-documents: 列出所有或指定空间的文档

示例用法

在Cursor中，您可以这样使用MCP工具：

请列出我所有的飞书文档空间

请列出空间ID为"XYZ123"中的所有文档

技术架构

该项目使用了以下技术：

  • Node.js 和 TypeScript：开发环境
  • MCP SDK：实现MCP服务器接口
  • Express：提供HTTP服务
  • 飞书开放API：访问飞书文档内容

注意事项

  • 应用需要有访问飞书云文档的权限

许可证

MIT

jiang-feishu-mcp FAQ

How do I configure jiang-feishu-mcp to connect to my Feishu account?
You need to create a .env file with your Feishu App ID and App Secret, obtained from the Feishu Open Platform, and ensure cloud document permissions are enabled.
What port does jiang-feishu-mcp run on by default?
By default, the server runs on port 7777, but you can customize this in the .env configuration file.
Can jiang-feishu-mcp be used with multiple AI clients?
Yes, it supports any MCP-compatible AI clients such as Claude, Cursor, and others, enabling seamless Feishu Docs integration.
How do I install and start the jiang-feishu-mcp server?
Clone the GitHub repo, install dependencies with pnpm, and start the server using pnpm start; it will listen on the configured port.
Does jiang-feishu-mcp support document editing or only reading?
Currently, it supports fetching document spaces and lists for reading and integration; editing capabilities depend on future updates.
Is jiang-feishu-mcp secure for enterprise use?
Security depends on your Feishu app credentials management and network setup; MCP protocol ensures scoped and observable interactions.
Can I customize the document save path?
Yes, you can set DOCS_SAVE_PATH in the .env file to specify where documents are saved locally.