Lark Integration
Advertencia de seguridad

Connect Lark (Feishu) messaging to OpenClaw via webhook bridge. Supports text, rich text (post), and image messages bidirectionally. Use when setting up Lark/Feishu as a messaging channel, receiving messages with images, sending replies back to Lark, reading Lark documents/wikis/bitables, or troubleshooting Lark integration issues. Covers both Lark International (larksuite.com) and China Feishu (feishu.cn).

Instalar
$clawhub install lark-integration

Lark Integration

Connect Lark (Feishu) to OpenClaw for bidirectional messaging with full rich content support.

Quick Start

# 1. Set credentials
echo "FEISHU_APP_ID=cli_xxx" >> ~/.openclaw/workspace/.env
mkdir -p ~/.openclaw/secrets
echo "your_app_secret" > ~/.openclaw/secrets/feishu_app_secret

# 2. Start bridge
cd skills/lark-integration/scripts
node bridge-webhook.mjs

# 3. Configure Lark webhook URL in developer console
# https://open.larksuite.com → Your App → Event Subscriptions
# URL: http://YOUR_SERVER_IP:3000/webhook

Architecture

Lark App ──webhook──► Bridge (port 3000) ──WebSocket──► OpenClaw Gateway
                           │                                   │
                           ◄────────── Reply ──────────────────┘

Supported Message Types

Type Direction Format
text ↔ Both Plain text
post → Receive Rich text with images, links
image → Receive Single image
Reply ← Send Text (cards via feishu-card skill)

Platform Detection

The bridge auto-detects platform from URLs: - *.larksuite.com → https://open.larksuite.com (International) - *.feishu.cn → https://open.feishu.cn (China)

Configuration

Environment Variables

Variable Required Description
FEISHU_APP_ID Yes App ID from Lark Developer Console
FEISHU_APP_SECRET_PATH No Path to secret file (default: ~/.openclaw/secrets/feishu_app_secret)
WEBHOOK_PORT No Webhook listen port (default: 3000)
FEISHU_THINKING_THRESHOLD_MS No Delay before "Thinking..." placeholder (default: 2500)
FEISHU_ENCRYPT_KEY No Encryption key if enabled in Lark
OPENCLAW_AGENT_ID No Agent to route messages to (default: main)

Lark App Permissions

Enable these scopes in Lark Developer Console → Permissions & Scopes:

Messaging: - im:message - Send and receive messages - im:message:send_as_bot - Send messages as bot - im:resource - Download message resources (images)

Documents (optional): - docx:document:readonly - Read documents - wiki:wiki:readonly - Read wiki spaces - sheets:spreadsheet:readonly - Read spreadsheets - bitable:bitable:readonly - Read bitables - drive:drive:readonly - Access drive files

Scripts

bridge-webhook.mjs

Main webhook bridge. Receives Lark events, forwards to OpenClaw, sends replies.

FEISHU_APP_ID=cli_xxx node scripts/bridge-webhook.mjs

setup-service.mjs

Install as systemd service for auto-start:

node scripts/setup-service.mjs
# Creates /etc/systemd/system/lark-bridge.service

Image Handling

Images in messages are: 1. Detected from post content or image message type 2. Downloaded via Lark API using message_id and image_key 3. Converted to base64 4. Sent to OpenClaw Gateway as attachments parameter

attachments: [{ mimeType: "image/png", content: "<base64>" }]

Group Chat Behavior

In group chats, the bridge responds when: - Bot is @mentioned - Message ends with ? or ？ - Message contains trigger words: help, please, why, how, what, 帮, 请, 分析, etc. - Message starts with bot name

Otherwise, messages are ignored to avoid noise.

Reading Documents

Use the feishu-doc skill to read Lark documents:

node skills/feishu-doc/index.js fetch "https://xxx.larksuite.com/docx/TOKEN"

Supported URL types: - /docx/ - New documents - /wiki/ - Wiki pages (auto-resolves to underlying doc) - /sheets/ - Spreadsheets - /base/ - Bitables (multi-dimensional tables)

Permission Note: Documents must be shared with the bot, or the bot must have tenant-wide read permission.

Troubleshooting

"forBidden" error when reading docs

  • Document not shared with bot → Add bot as collaborator
  • Missing scope → Enable docx:document:readonly in console

No messages received

  • Check webhook URL is accessible: curl http://YOUR_IP:3000/health
  • Verify webhook in Lark console shows "Verified"
  • Check bridge logs: journalctl -u lark-bridge -f

"must be string" error

  • Old bridge version → Update to use attachments for images

Images not received

  • Missing im:resource scope → Enable in Lark console
  • Token expired → Bridge auto-refreshes, restart if stuck

Service Management

# Check status
systemctl status lark-bridge

# View logs
journalctl -u lark-bridge -f

# Restart
systemctl restart lark-bridge

References

Detalles

Versión
v1.0.0
Descargas
2,919
Estrellas
2
ClawHub
Ver en ClawHub

Skills populares

diagram-generator
生成和编辑各种类型的图表（drawio、mermaid、excalidraw）。支持流程图、时序图、类图、ER图、思维导图、架构图、网络拓扑图等常见图表类型。能够根据自然语言描述创建新图表，也能读取并修改已有的 drawio/mermaid/excalidraw 文件。使用独立的 MCP server (mcp-diagram-generator) 生成图表文件，减少 token 消耗并保证输出一致性。支持自动配置管理，默认输出路径为项目目录下的 diagrams/{format}/，支持自定义路径和自动创建目录。
Find Skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Clawdbot Security Check
Perform a comprehensive read-only security audit of Clawdbot's own configuration. This is a knowledge-based skill that teaches Clawdbot to identify hardening opportunities across the system. Use when user asks to "run security check", "audit clawdbot", "check security hardening", or "what vulnerabilities does my Clawdbot have". This skill uses Clawdbot's internal capabilities and file system access to inspect configuration, detect misconfigurations, and recommend remediations. It is designed to be extensible - new checks can be added by updating this skill's knowledge.