Lark_doc

@Xeonice

About Lark_doc

base on origin lark mcp-server modified, add dictionary support

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "lark_doc": {
      "command": "/path/to/your/uvx",
      "args": [
        "mcp-lark-doc-manage"
      ],
      "env": {
        "LARK_APP_ID": "your_app_id",
        "LARK_APP_SECRET": "your_app_secret",
        "OAUTH_HOST": "localhost",
        "OAUTH_PORT": "9997",
        "FOLDER_TOKEN": "your_folder_token",
        "DEBUG": "1"
      }
    }
  }
}

Tools

Retrieve document content from Lark

documentUrl (string) - The URL of the Lark document

Document content in text format

Doc URLs: https://xxx.feishu.cn/docx/xxxxx

Search documents in Lark Wiki

query (string) - Search keywords

JSON string containing search results with following fields:

List contents of a specified folder

page_size (int, optional) - Number of results to return (default: 10)

JSON string containing file list with following fields:

Create a new Lark document with content

title (string) - Document title

JSON string containing:

Supports Markdown content conversion

Overview

What is Lark_doc?

Lark_doc is a Model Context Protocol (MCP) server that enables AI assistants to search, access, and manage Lark (Feishu) documents. It is designed for users who have a Lark Enterprise Application and need to integrate document retrieval, content extraction, and creation into LLM workflows.

How to use Lark_doc?

Install via uvx mcp-lark-doc-manage and set environment variables: LARK_APP_ID, LARK_APP_SECRET, FOLDER_TOKEN, OAUTH_HOST, OAUTH_PORT. Configure the MCP server in Claude desktop with the provided JSON block. The server exposes four tools: get_lark_doc_content, search_wiki, list_folder_content, and create_doc.

Key features of Lark_doc

Lark Doc and Wiki document support
OAuth authentication with automatic token refresh
Automatic document type detection and ID extraction
Markdown to document content conversion
Folder content listing and wiki search
Comprehensive error handling and reporting

Use cases of Lark_doc

Retrieve document content from Lark for LLM analysis
Search wiki documents by keywords
List contents of a specified folder
Create new Lark documents with Markdown content
Integrate documents into wiki spaces

FAQ from Lark_doc

What permissions are required for Lark_doc?

The app needs wiki:wiki:readonly, wiki:node:read, docx:document:readonly, search:docs:read, and drive:drive:readonly permissions.

How do I obtain the App ID and App Secret?

Go to your app in the Lark Developer Console, find "App ID" under "Credentials & Basic Info"; App Secret is also there and can be viewed by clicking "View".

How do I find the folder token?

Open the target folder in Lark, copy the URL, and the last segment after "folder/" is the folder token. Alternatively, use the Drive API.

What tools does Lark_doc provide?

It provides four tools: get_lark_doc_content, search_wiki, list_folder_content, and create_doc.

How do I handle common errors like "Authorization timeout"?

The authorization timeout means

Frequently asked questions

What permissions are required for Lark_doc?

The app needs `wiki:wiki:readonly`, `wiki:node:read`, `docx:document:readonly`, `search:docs:read`, and `drive:drive:readonly` permissions.

How do I obtain the App ID and App Secret?

Go to your app in the Lark Developer Console, find "App ID" under "Credentials & Basic Info"; App Secret is also there and can be viewed by clicking "View".