← 返回 Skills 市场

Wechat Automation

Name: Wechat Automation
Author: kaising-openclaw1

作者 kaising-openclaw1 · GitHub ↗ · v0.3.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install enterprise-wechat-bot

功能描述

企业微信自动化工具，支持通过Webhook发送文本和Markdown消息，群消息推送及定时任务管理。

使用说明 (SKILL.md)

WeChat Automation v0.2

企业微信自动化技能，支持 webhook 消息发送和自动化流程。

功能

✅ 企业微信 webhook 消息发送
✅ 群消息推送
✅ 定时任务支持
✅ 消息模板

安装

npx clawhub@latest install wechat-automation

配置

# 设置企业微信 Webhook Key
clawhub config set wechat.webhook_key YOUR_WEBHOOK_KEY

使用

发送消息

# 发送文本消息
clawhub wechat send --text "Hello World"

# 发送 Markdown 消息
clawhub wechat send --markdown "**重要通知**\
\
项目进度更新..."

# 发送到指定群
clawhub wechat send --text "消息内容" --chatid "CHAT_ID"

自动化示例

# 定时发送日报
clawhub wechat schedule --cron "0 18 * * *" --text "日报提醒"

API

sendMessage(text, options)

发送消息到企业微信群。

参数:

text (string): 消息内容
options (object): 可选配置
- markdown (boolean): 是否使用 Markdown 格式
- chatid (string): 目标群聊 ID
- mentioned_list (array): 需要@的用户列表

返回:

{
  "errcode": 0,
  "errmsg": "ok",
  "msgid": "msg_xxxxx"
}

示例

const wechat = require('wechat-automation');

// 发送通知
await wechat.sendMessage({
  text: '🎉 项目上线成功！',
  markdown: true
});

// 发送日报
await wechat.sendMessage({
  text: `
## 日报 2026-03-30

### 完成情况
- ✅ 功能 A
- ✅ 功能 B

### 明日计划
- [ ] 功能 C
  `,
  markdown: true
});

注意事项

需要企业微信管理员权限获取 Webhook Key
每个 Webhook 有发送频率限制（通常 20 条/分钟）
消息内容不能超过 2048 字节

更新日志

v0.2 (2026-03-30)

✅ 完善 API 文档
✅ 添加使用示例
✅ 支持 Markdown 格式

v0.1 (2026-03-29)

✅ 初始版本
✅ 基础 webhook 支持

安全使用建议

This skill largely implements a legitimate WeChat webhook sender, but there are noteworthy inconsistencies you should resolve before installing or giving it secrets: - Verify how the webhook key is provided: SKILL.md tells you to run 'clawhub config set wechat.webhook_key ...' but index.js does not read any config or environment variables — it only uses a webhookKey passed via CLI flag or programmatic option. Confirm that the clawhub platform will inject the stored key into the skill at runtime; if not, you must pass --webhook-key each invocation. - Don't paste your real Webhook Key, corpsecret, or other secrets into the UI until you've confirmed where the skill reads them. The docs show .env examples (WECHAT_WEBHOOK_KEY, CORPID, CORPSECRET) but those are not declared as required in the registry metadata. - Scheduling is advertised in documentation (cron examples) but no scheduling code is included. If you need scheduled sends, ask the author how scheduling is implemented (platform-level scheduler vs missing code). - If you plan to use the provided shell test script, ensure curl is available (test-webhook.sh uses curl). The main Node code does not require curl, but the meta file lists it as a required binary — another mismatch to clarify. - Recommended actions: run the local test script (test-webhook.js) with a throwaway/test webhook key first, inspect how your clawhub instance supplies config values, and confirm there are no hidden network endpoints other than qyapi.weixin.qq.com. If you need help verifying the integration behavior, request clarification from the author or review how clawhub injects 'clawhub config' entries into skill runtime environment.

功能分析

Type: OpenClaw Skill Name: enterprise-wechat-bot Version: 0.3.0 The skill bundle provides legitimate functionality for sending messages to Enterprise WeChat (WeCom) via webhooks. The implementation in `index.js` and `test-webhook.sh` correctly uses the official WeChat API endpoints, and the documentation (README.md, SKILL.md) accurately describes the setup and usage without any signs of malicious intent, data exfiltration, or prompt injection.

能力标签

requires-oauth-tokenrequires-sensitive-credentials

能力评估

ℹ Purpose & Capability

The name/description match the code: index.js and test scripts send messages to the official qyapi.weixin.qq.com webhook endpoint. However metadata and docs diverge: _meta.json lists curl as a required binary while the main implementation uses Node's https module; the docs advertise scheduling and clawhub config integration, but the shipped code only implements sendMessage and a simple CLI. These mismatches are plausible for a work-in-progress but are inconsistent with the stated capabilities.

⚠ Instruction Scope

SKILL.md instructs users to run 'clawhub wechat send' and to 'clawhub config set wechat.webhook_key ...' implying the skill will read stored config or support scheduling. The runtime code (index.js) does not read clawhub config or environment variables — it requires webhookKey be passed as --webhook-key (or via the options object when called programmatically). SKILL.md also mentions scheduling and cron usage, but there is no scheduling implementation in the provided code. This grants the agent unclear discretion (docs assume config injection that the code doesn't perform).

ℹ Install Mechanism

No install spec is provided (instruction-only install via clawhub), so nothing is pulled from arbitrary URLs. The repository includes code files (index.js, test scripts) and package.json, but no install script. The included test shell script uses curl — so if you use that script you need curl present (matching _meta.json), but the main Node implementation does not depend on external installers or downloaded archives.

⚠ Credentials

The docs and multiple files reference a Webhook Key and even list corpid/corpsecret/agent id in .env examples, yet the skill registry metadata declares no required env vars or primary credential. The code enforces presence of a webhookKey at runtime but does not read any environment variables or a config store itself. That mismatch means the skill asks the user (via docs) to store secrets but the runtime does not declare or automatically consume them — verify how the clawhub platform will inject config before providing any real webhook keys or other secrets.

✓ Persistence & Privilege

The skill does not request always:true, does not declare system-wide config changes, and has no install-time scripts. There is no evidence it attempts to persist beyond normal skill files or to alter other skills' configs.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install enterprise-wechat-bot
安装完成后，直接呼叫该 Skill 的名称或使用 /enterprise-wechat-bot 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v0.3.0

v0.3 优化版 - 企业微信 webhook 消息推送+Markdown+ 定时任务

v0.2.0

v0.2 企业微信自动化 - webhook 消息推送+Markdown+ 定时任务

元数据

Slug enterprise-wechat-bot

版本 0.3.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 2

常见问题