Dingtalk Connection

xiaomingx/moltbot-connector-feishu-dingtalk

This bridges DingTalk's outgoing webhooks to Clawdbot Gateway so you can run AI agents through DingTalk conversations. You get a local Python server that sits between DingTalk's cloud and your gateway, handling the webhook callbacks and websocket connection. It includes a macOS launchd setup script for auto-starting the bridge service, which is handy if you're running this long-term. The architecture is straightforward: DingTalk user messages hit their cloud, webhook to your bridge.py, then forward to the gateway. You'll need to expose the webhook endpoint publicly and configure your DingTalk bot's callback URL. Most useful if you're already invested in the DingTalk ecosystem and want conversational AI there.

skills.sh installs: 521
GitHub stars: 7
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add xiaomingx/moltbot-connector-feishu-dingtalk --skill dingtalk-connection --agent claude-code

Installs into .claude/skills of the current project.

Files

DingTalk Bridge

Bridge DingTalk bot messages to Clawdbot Gateway over a local WebSocket connection.

Architecture

DingTalk user -> DingTalk cloud -> webhook (bridge.py) -> Clawdbot Gateway -> AI agent

Setup

1. Create DingTalk bot

Create a bot that can send outgoing webhooks and set the callback URL to your public endpoint.

2. Run bridge

cd <skill-dir>/dingtalk-connection
uv sync
DINGTALK_SIGNING_SECRET=your_secret uv run python bridge.py

3. Auto-start (macOS)

uv run python setup_service.py
launchctl load ~/Library/LaunchAgents/com.clawdbot.dingtalk-bridge.plist

Environment variables

Variable	Required	Default
`DINGTALK_PORT`	—	`3210`
`DINGTALK_PATH`	—	`/dingtalk`
`DINGTALK_SIGNING_SECRET`	—	—
`DINGTALK_BOT_ID`	—	—
`DINGTALK_BOT_NAME`	—	—
`CLAWDBOT_CONFIG_PATH`	—	`~/.clawdbot/clawdbot.json`
`CLAWDBOT_AGENT_ID`	—	`main`
`DINGTALK_THINKING_THRESHOLD_MS`	—	`2500`