cso1z/Feishu-MCP

What it solves

This project enables AI coding assistants (like Cursor, Windsurf, and Cline) to interact directly with Feishu (Lark) documents and tasks. It bridges the gap between AI agents and enterprise productivity tools, allowing users to automate document creation, content editing, and task management without leaving their AI-powered development environment.

How it works

It implements the Model Context Protocol (MCP), a standard that allows AI models to securely access external data and tools. The project functions as an MCP server that translates natural language requests from an AI agent into specific API calls for Feishu. It supports two authentication modes: 'tenant' (application-level) and 'user' (user-level), with the latter providing deeper access to tasks and personal information. It also includes a standalone CLI tool, feishu-tool, for executing these operations directly from a terminal or within custom scripts.

Who it’s for

It is designed for developers and power users who use AI coding assistants and want to integrate their documentation and task management workflows directly into their AI-driven development process.

Highlights

• Document Automation: Create, edit, search, and manage Feishu documents and folders directly via AI.
• Task Management: Perform CRUD operations on Feishu tasks, including sub-task and member management.
• Flexible Integration: Supports multiple modes, including MCP server (for AI editors) and a standalone CLI (for scripts).
• Rich Content Support: Handles complex document elements like Markdown imports, Mermaid diagrams, images, tables, and mathematical formulas.
• Multi-User Support: Allows multiple users to share a single service instance through unique user-key identifiers.