MCPInstall

wechat-channel

微信 (WeChat) 与 OpenClaw 的双向集成通道。基于 Wechaty + PadLocal 实现微信消息的接收和发送,支持私聊、群聊、@提及检测、图片/文件传输。当需要通过微信与 AI 助手交互、接收微信消息触发 AI 响应、或从 OpenClaw 发送消息到微信时使用此技能。

Installs
522
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/aaaaqwq/agi-super-team --skill wechat-channel
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads wechat-channel automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

微信 Channel 集成

  • Author: Daniel Li
  • Copyright © Daniel Li. All rights reserved.

将微信接入 OpenClaw,实现双向消息通道。

架构概述

┌─────────────┐     ┌──────────────────┐     ┌─────────────┐
│   微信用户   │ ←→  │  Wechaty Bridge  │ ←→  │  OpenClaw   │
│  (私聊/群聊) │     │  (PadLocal协议)   │     │   Gateway   │
└─────────────┘     └──────────────────┘     └─────────────┘
                           ↓
                    ┌──────────────────┐
                    │   消息格式转换    │
                    │   - 文本/图片/文件 │
                    │   - @提及检测     │
                    │   - 群聊/私聊路由  │
                    └──────────────────┘

核心组件

1. Wechaty Bridge (消息桥接服务)

独立运行的 Node.js 服务,负责:

  • 微信登录(扫码)
  • 消息收发
  • 联系人/群组管理
  • 与 OpenClaw Gateway 通信

2. OpenClaw Webhook 接收器

接收来自 Wechaty Bridge 的消息,转发给 AI Agent。

3. 消息发送 API

OpenClaw Agent 通过 HTTP API 发送消息到微信。

快速开始

前置条件

  • Node.js >= 18
  • PadLocal Token(付费服务,约 ¥200/月)
  • OpenClaw Gateway 运行中

1. 安装依赖

cd /home/aa/clawd/skills/wechat-channel
npm init -y
npm install wechaty wechaty-puppet-padlocal axios dotenv

2. 配置环境变量

cp .env.example .env
# 编辑 .env 填入配置

3. 启动服务

node scripts/wechat-bridge.js
# 扫描终端显示的二维码登录

配置说明

环境变量 (.env)

# PadLocal Token (必需)
# 获取方式: https://pad-local.com
PADLOCAL_TOKEN=YOUR_PADLOCAL_TOKEN

# OpenClaw Gateway 配置
OPENCLAW_GATEWAY_URL=http://127.0.0.1:18789
OPENCLAW_WEBHOOK_SECRET=your_webhook_secret

# 微信 Bot 配置
WECHAT_BOT_NAME=OpenClaw助手

# 安全配置
# 允许的用户微信ID (逗号分隔,留空允许所有)
ALLOWED_USERS=wxid_xxx,wxid_yyy
# 允许的群聊ID (逗号分隔,留空允许所有)
ALLOWED_GROUPS=xxx@chatroom,yyy@chatroom

# 群聊行为
# 是否需要@才响应群消息
REQUIRE_MENTION_IN_GROUP=true

# 日志级别
LOG_LEVEL=info

OpenClaw 配置 (openclaw.json)

{
  "channels": {
    "wechat": {
      "enabled": true,
      "webhookUrl": "http://localhost:3001/webhook",
      "webhookSecret": "your_webhook_secret",
      "dmPolicy": "allowlist",
      "allowFrom": ["wxid_xxx", "wxid_yyy"],
      "groups": {
        "xxx@chatroom": {
          "name": "工作群",
          "requireMention": true
        }
      }
    }
  }
}

消息格式

接收消息 (Webhook Payload)

{
  "type": "message",
  "channel": "wechat",
  "messageId": "msg_123456",
  "from": {
    "id": "wxid_sender",
    "name": "张三",
    "alias": "zhangsan"
  },
  "chat": {
    "id": "wxid_sender",
    "type": "private"
  },
  "text": "你好,帮我查一下天气",
  "timestamp": 1706745600000,
  "mentions": [],
  "replyTo": null
}

群聊消息

{
  "type": "message",
  "channel": "wechat",
  "messageId": "msg_789012",
  "from": {
    "id": "wxid_sender",
    "name": "张三"
  },
  "chat": {
    "id": "xxx@chatroom",
    "type": "group",
    "name": "工作群"
  },
  "text": "@OpenClaw助手 帮我总结一下今天的会议",
  "mentions": ["bot_wxid"],
  "isMentioned": true
}

发送消息 (API)

# 发送文本
curl -X POST http://localhost:3001/api/send \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_SECRET" \
  -d '{
    "to": "wxid_receiver",
    "type": "text",
    "content": "收到,正在处理..."
  }'

# 发送图片
curl -X POST http://localhost:3001/api/send \
  -H "Content-Type: application/json" \
  -d '{
    "to": "wxid_receiver",
    "type": "image",
    "url": "https://example.com/image.png"
  }'

# 发送文件
curl -X POST http://localhost:3001/api/send \
  -d '{
    "to": "wxid_receiver",
    "type": "file",
    "path": "/path/to/file.pdf",
    "filename": "report.pdf"
  }'

安全策略

私聊策略 (dmPolicy)

策略说明
open允许所有人私聊(危险)
allowlist仅允许 allowFrom 列表中的用户
pairing需要配对审批

群聊策略

配置说明
requireMention: true必须@机器人才响应
allowFrom群内允许触发的用户列表

使用场景

1. 个人助手

用户: 帮我查一下明天北京的天气
Bot: 明天北京天气:晴,温度 -5°C ~ 5°C,建议穿羽绒服。

2. 群聊助手

用户: @OpenClaw助手 总结一下刚才的讨论
Bot: 刚才讨论的要点:
1. 项目进度需要加快
2. 下周三前完成设计稿
3. 周五进行代码评审

3. 自动化通知

// 从 OpenClaw Agent 发送通知
await sendWechatMessage({
  to: 'xxx@chatroom',
  text: '⚠️ 服务器 CPU 使用率超过 90%,请检查!'
});

故障排查

登录问题

问题: 扫码后无法登录 解决:

  1. 检查 PadLocal Token 是否有效
  2. 确认微信账号未被限制
  3. 尝试重新获取 Token

消息收发问题

问题: 消息发送失败 解决:

  1. 检查网络连接
  2. 确认目标用户/群组 ID 正确
  3. 查看日志中的错误信息

连接断开

问题: 服务运行一段时间后断开 解决:

  1. 使用 PM2 管理进程,自动重启
  2. 检查 PadLocal 服务状态
  3. 实现心跳检测和重连机制

限制说明

PadLocal 限制

  • 需要付费 Token(约 ¥200/月)
  • 单 Token 只能登录一个微信号
  • 可能受微信风控影响

微信平台限制

  • 发送频率限制(建议间隔 1-2 秒)
  • 群聊人数限制
  • 文件大小限制(约 100MB)
  • 不支持小程序消息

功能限制

  • 不支持语音消息转文字(需额外集成)
  • 不支持视频号内容
  • 红包、转账等敏感功能不可用

相关文件

  • scripts/wechat-bridge.js - 主服务代码
  • scripts/message-handler.js - 消息处理逻辑
  • .env.example - 环境变量模板
  • references/wechaty-api.md - Wechaty API 参考

TODO

  • 获取 PadLocal Token
  • 配置 OpenClaw Webhook 接收
  • 测试私聊消息收发
  • 测试群聊 @提及
  • 配置安全策略
  • 部署为系统服务
  • 实现断线重连
  • 添加消息队列(高并发场景)

FAQ

What does the wechat-channel skill do?

微信 (WeChat) 与 OpenClaw 的双向集成通道。基于 Wechaty + PadLocal 实现微信消息的接收和发送,支持私聊、群聊、@提及检测、图片/文件传输。当需要通过微信与 AI 助手交互、接收微信消息触发 AI 响应、或从 OpenClaw 发送消息到微信时使用此技能。

How do I install wechat-channel?

Run npx skills add https://github.com/aaaaqwq/agi-super-team --skill wechat-channel in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/wechat-channel/ (personal) or .claude/skills/wechat-channel/ (project).

Which AI agents can use wechat-channel?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is wechat-channel safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains wechat-channel?

It is published by aaaaqwq in aaaaqwq/agi-super-team. It has 522 recorded installs.

More from aaaaqwq/agi-super-team

View all →

multi-search-engine

aaaaqwq/agi-super-team

Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and WolframAlpha knowledge queries. No API keys required.

Backend & APIs2.7K installs

web-scraping-automation

aaaaqwq/agi-super-team

自动化爬取网站数据和 API 接口。当用户需要抓取网页内容、调用 API、解析数据或创建爬虫脚本时使用此技能。

project-management

aaaaqwq/agi-super-team

Open the skill to read its SKILL.md instructions.

468 installs

wecom-automation

aaaaqwq/agi-super-team

Open the skill to read its SKILL.md instructions.

feishu-automation

aaaaqwq/agi-super-team

Open the skill to read its SKILL.md instructions.

memory-hygiene

aaaaqwq/agi-super-team

Open the skill to read its SKILL.md instructions.

363 installs

Popular other skills

View category →

hyperframes-cli

heygen-com/hyperframes

The HyperFrames CLI by job - scaffold, check, look at frames, edit the timeline, preview, render locally or in the cloud, publish, project history - plus reading its JSON and recovering from build or render failures.

1.3M installs

triage

mattpocock/skills

Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.

1M installs

ask-matt

mattpocock/skills

Ask which skill or flow fits your situation. A router over the skills in this repo.

778.4K installs

to-questionnaire

mattpocock/skills

Turn a decision you can't fully answer into a questionnaire for someone else to fill in.

508.9K installs

wait-what

mattpocock/skills

Stop. That last message did not land: re-pitch it.

459.7K installs

loop-me

mattpocock/skills

Grill me about specs for the workflows I want to build, within this workspace.

447.9K installs