AI Agent Hub
Back to plugins
🧰

dsh-feishu-mcp

Web Tools Updated 2026.09.11

Run the following command in DeepSeek Harness:

dsh plugin install zhengjy01/dsh-feishu-mcp

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install zhengjy01/dsh-feishu-mcp in DeepSeek Harness to install this plugin, available at https://github.com/zhengjy01/dsh-feishu-mcp .

About this plugin

Getting an AI agent to operate Feishu/Lark directly sounds natural, but the plumbing is where things usually break — reverse proxies, public IPs, hand-rolled SDK wrappers, each adding maintenance overhead. dsh-feishu-mcp takes a simpler path: it spawns the official Lark MCP server as a subprocess over stdio, auto-discovers every tool, and registers them on the DSH agent. No extra network components to babysit.

What the agent can do depends on the permissions you grant in the Feishu Open Platform console. IM messaging, Bitable CRUD, document read/write, calendar management, and drive file operations all surface as mcp__feishu__* tools. The plugin supports both tenant identity and user identity via userAccessToken, so private resources remain reachable. A web settings panel lets you fill in credentials, test the connection, and review the live tool list at a glance — no config file spelunking required.

Built for developers and teams who already collaborate on Feishu and run DeepSeek Harness for day-to-day development or ops. No MCP protocol expertise needed: install the plugin, configure your app credentials, and the agent can send messages, update spreadsheets, and pull documents, making Feishu a native part of your AI workflow.

Use Cases

  • Have the AI agent post messages and summarize threads in Feishu groups directly
  • Use natural language to let the agent read, write, and reorganize Bitable spreadsheets
  • Fold Feishu docs, calendar, and drive operations into your AI workflow to cut manual context-switching

Best For

  • Developers and product teams who collaborate on Feishu daily
  • Engineers running DeepSeek Harness for day-to-day dev or ops
  • Users who want the AI to handle Feishu chores and reduce repetitive manual work