做什么
使用 Docker 配置和配置 OpenClaw AI bot 网关与中文 IM 平台(Feishu, DingTalk, QQ, WeChat Work)
技能库 智客分类:运维与云 openclaw-china-docker
使用 Docker 配置和配置 OpenClaw AI bot 网关与中文 IM 平台(Feishu, DingTalk, QQ, WeChat Work)
官方网址:skills.sh
先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。
使用 Docker 配置和配置 OpenClaw AI bot 网关与中文 IM 平台(Feishu, DingTalk, QQ, WeChat Work)
官方 description 未单独写出 Use when。按规范,代理会在用户任务与这段 description 的关键词匹配时激活本技能。
按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:OpenClaw China Docker Skill、What is OpenClaw-China-Docker?、Installation、Prerequisites、Quick Start、Clone the repository。
文件分析:这是一份仅含 SKILL.md 的指令型技能,代理激活后整份正文进入上下文。 正文约 519 行,超过规范建议的 500 行,激活时仍会整份载入。
Deploy and configure OpenClaw AI bot gateway with Chinese IM platforms (Feishu, DingTalk, QQ, WeChat Work) using Docker
OpenClaw China Docker SkillWhat is OpenClaw-China-Docker?InstallationPrerequisitesQuick StartClone the repositoryCopy environment templateOR use minimal version for quick setupEdit environment variablesStart servicesView logsStop services
来源分类:skills.sh agent-skill
nameopenclaw-china-dockerdescription具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗
先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。
复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。
把 Agent Skill「openclaw-china-docker」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-67bc6b595847eadf-%E5%BC%80%E5%85%83%E4%B8%AD%E5%9B%BD%E9%81%93%E5%85%8B.html 请存为 .cursor/skills/openclaw-china-docker/SKILL.md 或 .claude/skills/openclaw-china-docker/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。
需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。
npx skills add 'https://github.com/reason-machines/hermes-skills' --list
npx skills add 'https://github.com/reason-machines/hermes-skills' --skill 'openclaw-china-docker'
CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。
name: openclaw-china-docker description: Deploy and configure OpenClaw AI bot gateway with Chinese IM platforms (Feishu, DingTalk, QQ, WeChat Work) using Docker triggers:
Skill by ara.so — Hermes Skills collection.
This skill helps you deploy and configure OpenClaw-China-Docker, a pre-configured Docker image that integrates OpenClaw AI bot gateway with mainstream Chinese IM platforms including Feishu (Lark), DingTalk, QQ Bot, and WeChat Work (WeCom).
OpenClaw-China-Docker is a production-ready Docker image that packages OpenClaw with pre-installed plugins for Chinese IM platforms. It provides:
.env files# Clone the repository
git clone https://github.com/justlovemaki/openclaw-china-docker.git
cd openclaw-china-docker
# Copy environment template
cp .env.example .env
# OR use minimal version for quick setup
cp .env.minimal .env
# Edit environment variables
nano .env
# Start services
docker compose up -d
# View logs
docker compose logs -f openclaw
# Stop services
docker compose down
docker pull justlikemaki/openclaw-docker-cn-im:latest
The project uses environment variables for configuration. Key sections:
# OpenAI compatible API
OPENCLAW_AI_PROVIDER=openai
OPENCLAW_OPENAI_API_KEY=${OPENAI_API_KEY}
OPENCLAW_OPENAI_BASE_URL=https://api.openai.com/v1
OPENCLAW_OPENAI_MODEL=gpt-4o
# Claude API
OPENCLAW_AI_PROVIDER=anthropic
OPENCLAW_ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
OPENCLAW_ANTHROPIC_MODEL=claude-3-5-sonnet-20241022
# Use AIClient-2-API (recommended for unlimited tokens)
OPENCLAW_OPENAI_BASE_URL=http://your-aiclient-api:8000/v1
# Official plugin (recommended)
OPENCLAW_FEISHU_ENABLED=true
OPENCLAW_FEISHU_APP_ID=cli_xxx
OPENCLAW_FEISHU_APP_SECRET=xxx
OPENCLAW_FEISHU_VERIFICATION_TOKEN=xxx
OPENCLAW_FEISHU_ENCRYPT_KEY=xxx
# Legacy built-in version
OPENCLAW_FEISHU_OLD_ENABLED=true
OPENCLAW_FEISHU_OLD_APP_ID=cli_xxx
# ... (similar structure)
OPENCLAW_DINGTALK_ENABLED=true
OPENCLAW_DINGTALK_CLIENT_ID=dingxxx
OPENCLAW_DINGTALK_CLIENT_SECRET=xxx
OPENCLAW_DINGTALK_ROBOT_CODE=xxx
OPENCLAW_QQBOT_ENABLED=true
OPENCLAW_QQBOT_APPID=xxx
OPENCLAW_QQBOT_SECRET=xxx
OPENCLAW_QQBOT_INTENT=1 # Public messages
OPENCLAW_WECOM_ENABLED=true
OPENCLAW_WECOM_CORP_ID=wwxxx
OPENCLAW_WECOM_AGENT_ID=1000xxx
OPENCLAW_WECOM_SECRET=xxx
OPENCLAW_WECOM_TOKEN=xxx
OPENCLAW_WECOM_ENCODING_AES_KEY=xxx
# Enable Docker-in-Docker sandbox for code execution
OPENCLAW_SANDBOX_MODE=all # or 'non-main' or 'none'
Minimal docker-compose.yml:
version: '3.8'
services:
openclaw:
image: justlikemaki/openclaw-docker-cn-im:latest
container_name: openclaw-cn-im
restart: unless-stopped
env_file:
- .env
ports:
- "3000:3000"
volumes:
- ./workspace:/openclaw/workspace
- ./data:/openclaw/data
# Uncomment for Docker sandbox mode
# - /var/run/docker.sock:/var/run/docker.sock
environment:
- TZ=Asia/Shanghai
# Start services
docker compose up -d
# View logs (all services)
docker compose logs -f
# View logs (specific service)
docker compose logs -f openclaw
# Restart service
docker compose restart openclaw
# Stop services
docker compose down
# Stop and remove volumes
docker compose down -v
# Interactive shell
docker compose exec openclaw bash
# Execute single command
docker compose exec openclaw openclaw --version
# Pull latest changes
git pull origin main
# Pull latest image
docker compose pull
# Recreate containers
docker compose up -d --force-recreate
# Clean old images
docker image prune -f
# Use utility container for one-time operations
docker compose run --rm openclaw-utils bash
# Inside container:
cd /openclaw
npx openclaw plugin:install larksuite/openclaw-lark
# .env configuration
OPENCLAW_AI_PROVIDER=openai
OPENCLAW_OPENAI_API_KEY=${OPENAI_API_KEY}
OPENCLAW_OPENAI_MODEL=gpt-4o
# Enable multiple IM platforms
OPENCLAW_FEISHU_ENABLED=true
OPENCLAW_DINGTALK_ENABLED=true
OPENCLAW_QQBOT_ENABLED=true
OPENCLAW_WECOM_ENABLED=true
# Configure each platform's credentials...
OPENCLAW_FEISHU_APP_ID=cli_xxx
OPENCLAW_DINGTALK_CLIENT_ID=dingxxx
OPENCLAW_QQBOT_APPID=xxx
OPENCLAW_WECOM_CORP_ID=wwxxx
# Deploy
docker compose up -d
# Monitor all platforms
docker compose logs -f | grep -E 'feishu|dingtalk|qqbot|wecom'
# .env - Use AIClient for unlimited tokens
OPENCLAW_AI_PROVIDER=openai
OPENCLAW_OPENAI_BASE_URL=http://aiclient-api:8000/v1
OPENCLAW_OPENAI_API_KEY=dummy-key # AIClient doesn't require real key
OPENCLAW_OPENAI_MODEL=gpt-4o
# docker-compose.yml - Add AIClient service
services:
aiclient-api:
image: justlikemaki/aiclient-2-api:latest
ports:
- "8000:8000"
volumes:
- aiclient-data:/app/data
openclaw:
depends_on:
- aiclient-api
# ... rest of openclaw config
Initialize Agent Reach in conversation:
禁止使用web_search,web_fetch 工具, 必须使用 agent-reach 的工具来替代你自带的web_search,web_fetch ,并写入tools和记忆文档中
This switches from built-in web tools to Agent Reach for searching Twitter, Xiaohongshu, Weibo, Douyin, etc.
# .env
OPENCLAW_SANDBOX_MODE=all # Run all code in isolated containers
# docker-compose.yml - MUST mount Docker socket
services:
openclaw:
volumes:
- /var/run/docker.sock:/var/run/docker.sock
# In chat, OpenClaw can now safely execute:
"""
Create a Python script that analyzes data
"""
# Code runs in isolated container, not on host
# Mount custom config instead of using env vars
# docker-compose.yml
services:
openclaw:
volumes:
- ./openclaw.json:/openclaw/openclaw.json
{
"ai": {
"provider": "openai",
"openai": {
"apiKey": "${OPENAI_API_KEY}",
"baseURL": "https://api.openai.com/v1",
"model": "gpt-4o"
}
},
"channels": {
"feishu": {
"enabled": true,
"appId": "${FEISHU_APP_ID}",
"appSecret": "${FEISHU_APP_SECRET}"
}
}
}
# docker-compose.yml - Persist important data
services:
openclaw:
volumes:
- ./workspace:/openclaw/workspace # Work files
- ./data:/openclaw/data # Database & state
- ./logs:/openclaw/logs # Log files
# Backup data
tar -czf openclaw-backup-$(date +%Y%m%d).tar.gz workspace/ data/
# Restore data
tar -xzf openclaw-backup-20260517.tar.gz
# Check logs
docker compose logs openclaw
# Common causes:
# 1. Missing required env vars
grep "OPENCLAW_.*_ENABLED=true" .env
# Ensure corresponding credentials are set
# 2. Port conflicts
netstat -tlnp | grep 3000
# Change port in docker-compose.yml if needed
# Verify plugin installation
docker compose exec openclaw bash
npx openclaw plugin:list | grep lark
# Reinstall if needed
npx openclaw plugin:install larksuite/openclaw-lark
# Check webhook URL accessibility
curl -X POST http://your-server:3000/api/feishu/webhook
# Test API connectivity
docker compose exec openclaw bash
curl -H "Authorization: Bearer ${OPENCLAW_OPENAI_API_KEY}" \
${OPENCLAW_OPENAI_BASE_URL}/models
# Check model name
echo $OPENCLAW_OPENAI_MODEL
# Must match available models from API
# Ensure Docker socket is mounted
docker compose exec openclaw ls -la /var/run/docker.sock
# Should show: srw-rw---- 1 root docker
# Verify sandbox mode
docker compose exec openclaw env | grep SANDBOX
# Should show: OPENCLAW_SANDBOX_MODE=all
# Check Docker accessibility
docker compose exec openclaw docker ps
# Should list containers, not error
# Fix ownership
sudo chown -R 1000:1000 workspace/ data/ logs/
# Or run container with host user
# docker-compose.yml
services:
openclaw:
user: "${UID}:${GID}"
# Check resource usage
docker stats openclaw
# Limit resources in docker-compose.yml
services:
openclaw:
deploy:
resources:
limits:
cpus: '2'
memory: 4G
reservations:
memory: 2G
# List installed plugins
docker compose exec openclaw npx openclaw plugin:list
# Remove conflicting plugin
docker compose exec openclaw npx openclaw plugin:uninstall plugin-name
# Clear plugin cache
docker compose exec openclaw rm -rf /openclaw/.openclaw/plugins/.cache
docker compose restart openclaw
Key variables (see .env.example for complete list):
OPENCLAW_AI_PROVIDER: openai, anthropic, google, etc.OPENCLAW_SANDBOX_MODE: all, non-main, noneOPENCLAW_LOG_LEVEL: debug, info, warn, errorOPENCLAW_SERVER_PORT: Default 3000OPENCLAW_SERVER_HOST: Default 0.0.0.0# Development
cp .env.minimal .env.dev
docker compose -f docker-compose.yml -f docker-compose.dev.yml up
# Production
cp .env.example .env.prod
# Configure all platforms
docker compose -f docker-compose.yml -f docker-compose.prod.yml up -d
# docker-compose.yml
services:
openclaw:
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:3000/api/health"]
interval: 30s
timeout: 10s
retries: 3
start_period: 40s
docs/quick-start.mddocs/configuration.mddocs/aiclient-2-api.mddocs/advanced.mddocs/faq.mddocs/developer-notes.md.env files instead of hardcoding credentialsdocker compose logs -ftar -czfgit pull before upgrading运维与云
用于容器管理,图像操作,调试的基本道克命令和工作流程.
运维与云
从开放的代理技能生态系统中发现并安装技能. 使用时:(1)用户问"我如何做X",X可能拥有现有技能,(2)用户说"为X找到技能"或"是否为X有技能",(3)用户问"你能否做X",X是专门能力,(4)用户想扩展代理能力,(5)用户想搜索工具,模板,或工作流程,(6)用户提到他们希望…
运维与云
构建,部署,评价,优化,微调,管理微软创始代理,模型,资源最终结束. 用法:铸造、azd ai代理、azd 提供/部署、托管代理脚手架/开发/运行/部署/拖放、即时代理创建、创建代理、更新代理、添加代理工具、援引代理、代理.yaml、评估代理、批量活化、连续活化、持续监测、代理C…
运维与云
执行已有.azure/部署计划.md和基础设施文件的ALREADY-PREPARED应用程序的Azure部署。 当用户向 CREATE 请求新应用程序时, 请不要使用此技能 。 这种技能运行了 azd上行, azd 部署, trarraform 应用, 和 az 部署命令带有内置…