WeChat Sender by dxx

通过OpenClaw微信通道主动发送文字、图片和文件消息，支持单发、定时和批量推送，需联系人已存在聊天记录。

Security Scan

The skill's code, docs, and runtime instructions match its stated purpose (sending WeChat text/images/files via a local OpenClaw Gateway); it requests no external credentials and contains no obfuscated or unexpected network calls, though the normal file-send functionality can read arbitrary local files and thus could be used to leak data if misused.

Purpose & Capability

Name/description describe sending messages via the OpenClaw weixin channel; the scripts implement a local HTTP client to POST to the documented local Gateway API and include contact management and authorization helpers — all coherent with the stated purpose.

Instruction Scope

SKILL.md and scripts only reference a local OpenClaw Gateway (127.0.0.1:18789) and local files. send_message.py will read any file path provided, base64-encode it, and include it in the outgoing payload — this is expected for file sending but also enables sending arbitrary local file contents to the Gateway. auth_helper enforces a local consent flow (writes .auth.json) but does not prevent programmatic use if the user or environment grants authorization.

Install Mechanism

No install spec; the skill is instruction/code-only and does not fetch remote artifacts or run installers. Files are plain Python scripts and docs. This minimizes install-time risk.

Credentials

The skill declares no environment variables or external credentials (proportionate). It does create and read local files (.auth.json and .contacts.json) in the project parent directory to track authorization and contacts — expected for this functionality but worth noting if you run the skill from a shared directory.

Persistence & Privilege

always is false and the skill does not request platform-wide privileges or modify other skills. It persists only its own local state files and does not attempt to enable itself or change system-wide settings.

Assessment

This skill appears to do what it claims: send WeChat messages via a local OpenClaw Gateway. Before installing, confirm you trust the OpenClaw Gateway at 127.0.0.1:18789 and the environment where the agent runs. Be aware that the file-send feature will read any local file path you provide and include its base64 content in the request — which is necessary for attachments but could be used to transmit sensitive local files if misused. Keep .auth.json and .contacts.json in a safe location, ensure you grant explicit authorization for bulk or automated sends, and avoid running the skill in environments where untrusted code or users could call it to exfiltrate data.

WeChat Sender Skill

通过OpenClaw的openclaw-weixin通道主动发送微信消息。

工作原理

本Skill利用OpenClaw Gateway与微信客户端的WebSocket连接，通过发送特定格式的消息包实现主动推送。

前置条件

1. OpenClaw Gateway已运行

   openclaw gateway status
   

2. 微信通道已配置

   • openclaw-weixin 插件已启用
   • 微信客户端已扫码登录

3. 目标联系人已存在聊天记录

   • 首次使用前需先收到对方消息或手动发送过消息

使用方法

发送文字消息

# 使用脚本发送
python3 scripts/send_message.py --to "联系人备注名" --text "消息内容"

# 示例
python3 scripts/send_message.py --to "张三" --text "你好，这是一条测试消息"

发送图片

python3 scripts/send_message.py --to "联系人备注名" --image "/path/to/image.jpg"

发送文件

python3 scripts/send_message.py --to "联系人备注名" --file "/path/to/document.pdf"

安全限制

限制项	说明
联系人白名单	只能发送给已存在的聊天对象
频率限制	个人号20条/分钟，超过会被限制
内容审核	敏感词会被拦截
用户确认	首次使用需用户显式授权

API参考

发送消息接口

POST http://127.0.0.1:18789/api/wechat/send
Content-Type: application/json

{
  "channel": "openclaw-weixin",
  "recipient": {
    "type": "user",  // user | group
    "name": "联系人备注名"
  },
  "message": {
    "type": "text",  // text | image | file
    "content": "消息内容"
  }
}

响应格式

{
  "success": true,
  "messageId": "msg-xxx",
  "timestamp": 1701234567890
}

错误处理

错误码	说明	解决方案
E001	未登录微信	检查微信通道状态
E002	联系人不存在	确认备注名正确，或先接收对方消息
E003	频率限制	等待60秒后重试
E004	消息内容违规	检查敏感词

脚本说明

• scripts/send_message.py - 发送消息主脚本
• scripts/contact_manager.py - 联系人管理
• scripts/auth_helper.py - 授权验证

相关文档

• references/wechat-api.md - 微信API详细说明
• references/rate-limits.md - 频率限制详情