Feishu Business Operations

by haohanghao0321-cpu

343 downloads
Not rated
GitHub

About

AI-native business operations MCP server for Feishu/Lark. 8 scenario-based tools for multidimensional tables (search, query, write, create, AI schema design) and approvals (list, alerts, stats).

Details

Author
haohanghao0321-cpu
Downloads
343
Categories
Communication

- Eight tools covering Bitable and approval workflows
- Full CRUD and search on multidimensional tables
- AI-powered table schema design using DeepSeek
- Approval instance query, alerts, and statistics
- Supports both Feishu and Lark environments
- WebSocket-based Feishu bot for mobile access
- Batch write with auto field-name matching and type conversion

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Feishu Business Operations
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Install globally via npm install -g feishu-scenarios-mcp or run directly with npx feishu-scenarios-mcp. Configure a .env file with your Feishu app credentials (FEISHU_APP_ID, FEISHU_APP_SECRET, and optionally FEISHU_BITABLE_APP_TOKEN, FEISHU_ENV, FEISHU_USER_ID, DEEPSEEK_API_KEY). For Claude Desktop, add the server to your claude_desktop_config.json with the same environment variables.

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "feishu business operations": {
            "feishu-scenarios": {
                "command": "npx",
                "args": [
                    "-y",
                    "feishu-scenarios-mcp"
                ],
                "env": {
                    "FEISHU_APP_ID": "<\u4f60\u7684\u98de\u4e66App ID>",
                    "FEISHU_APP_SECRET": "<\u4f60\u7684\u98de\u4e66App Secret>",
                    "FEISHU_BITABLE_APP_TOKEN": "<\u4f60\u7684\u591a\u7ef4\u8868\u683cApp Token>",
                    "FEISHU_ENV": "feishu",
                    "FEISHU_USER_ID": "<\u4f60\u7684\u98de\u4e66Open ID (\u53ef\u9009)>",
                    "DEEPSEEK_API_KEY": "<\u4f60\u7684DeepSeek Key (\u53ef\u9009)>"
                }
            }
        }
    }
}

McpServers

{
    "feishu-scenarios": {
        "command": "npx",
        "args": [
            "-y",
            "feishu-scenarios-mcp"
        ],
        "env": {
            "FEISHU_APP_ID": "<\u4f60\u7684\u98de\u4e66App ID>",
            "FEISHU_APP_SECRET": "<\u4f60\u7684\u98de\u4e66App Secret>",
            "FEISHU_BITABLE_APP_TOKEN": "<\u4f60\u7684\u591a\u7ef4\u8868\u683cApp Token>",
            "FEISHU_ENV": "feishu",
            "FEISHU_USER_ID": "<\u4f60\u7684\u98de\u4e66Open ID (\u53ef\u9009)>",
            "DEEPSEEK_API_KEY": "<\u4f60\u7684DeepSeek Key (\u53ef\u9009)>"
        }
    }
}

feishu-scenarios-mcp

AI-native business operations MCP server for Feishu/Lark. 8 scenario-based tools covering multidimensional tables (Bitable) and approvals — search, query, write, create, AI schema design, approval tracking, alerts, and statistics.

Installation

npm install -g feishu-scenarios-mcp

Or run directly:

npx feishu-scenarios-mcp

Configuration

Create a .env file with your Feishu app credentials:

FEISHU_APP_ID=cli_xxxxxxxxxxxx
FEISHU_APP_SECRET=xxxxxxxxxxxxxxxxxxxxxxxx
FEISHU_BITABLE_APP_TOKEN=xxxxxxxxxxxx
FEISHU_ENV=feishu

| Variable | Required | Description |
|----------|----------|-------------|
| FEISHU_APP_ID | Yes | Feishu app ID from Developer Console |
| FEISHU_APP_SECRET | Yes | Feishu app secret |
| FEISHU_BITABLE_APP_TOKEN | For bitable | Bitable app token |
| FEISHU_ENV | No | feishu (default) or lark |
| FEISHU_USER_ID | For approval | Your Feishu user open_id |
| DEEPSEEK_API_KEY | For schema AI | DeepSeek API key for AI table design |

Claude Desktop Setup

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "feishu-scenarios": {
      "command": "npx",
      "args": ["-y", "feishu-scenarios-mcp"],
      "env": {
        "FEISHU_APP_ID": "cli_xxxxxxxxxxxx",
        "FEISHU_APP_SECRET": "xxxxxxxxxxxxxxxx",
        "FEISHU_BITABLE_APP_TOKEN": "xxxxxxxxxxxx",
        "FEISHU_ENV": "feishu"
      }
    }
  }
}

Tools

Bitable (多维表格)

| Tool | Description |
|------|-------------|
| bitable_search | Search tables by keyword. Returns table ID, name, field preview, record count. |
| bitable_query | Query records with optional filters. |
| bitable_upsert | Batch write records. Auto field-name matching and type conversion. |
| bitable_create | Create a new table with field definitions. |
| bitable_schema_ai | AI-designed table schema from business description. Requires DEEPSEEK_API_KEY. |

Approval (审批)

| Tool | Description |
|------|-------------|
| approval_list | Query approval instances by status, type, applicant, time range. |
| approval_alerts | Find bottlenecks and overtime approvals. Default threshold 48h. |
| approval_stats | Approval statistics: by status, type, month, top applicants. |

Usage Examples

Search tables:
> "搜一下包含'审批'的表格"

Query records:
> "查一下积压跟踪表里 pending 状态的记录"

Write records:
> "往跟踪表写一条记录,负责人张三,状态待处理"

AI design table:
> "帮我设计一张跟踪审批积压的表"

Approval alerts:
> "谁审批最慢"

Approval statistics:
> "本月审批通过率"

Feishu Bot (Mobile)

The package also includes a WebSocket-based Feishu bot for mobile access:

npx feishu-scenarios-mcp-bot

Setup guide: docs/bot-setup.md

Required Feishu Permissions

- approval:instance — read approval instances
- bitable:app — access bitable
- im:message — bot messaging
- im:message:send_as_bot — send bot messages

License

MIT

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.