Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

ReadGZH

v1.3.9

让 AI 读懂微信公众号。自研 7 阶段提取管线,99.89% 穿透反爬,Token 消耗降低 50–87%。支持 ChatGPT、Claude、Perplexity、Gemini 等平台无缝引用。| Let AI read and understand WeChat Official Accounts. Sel...

2· 611·5 current·6 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for sweesama/readgzh.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "ReadGZH" (sweesama/readgzh) from ClawHub.
Skill page: https://clawhub.ai/sweesama/readgzh
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Canonical install target

openclaw skills install sweesama/readgzh

ClawHub CLI

Package manager switcher

npx clawhub@latest install readgzh
Security Scan
Capability signals
Requires OAuth tokenRequires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
Name, README, SKILL.md and openapi.yaml consistently describe a WeChat article scraping/caching API (endpoints, MCP support, browser/CDP fallback). That functionality is coherent with the declared purpose. However, bundling a credentials.json with an apparent live API key is unexpected for a read-only/instruction-only skill and does not match the declared 'no required env vars' (incoherence).
Instruction Scope
SKILL.md instructions stay within the stated function: calling the readgzh.site API, configuring an MCP server, or using a browser CDP session. The instructions do not request reading local user files or other unrelated system secrets. They do encourage using an Authorization header or URL key parameter (which is part of the service design).
Install Mechanism
This is instruction-only (no install spec and no code). No downloads or archive extraction are specified, which is low-risk for installation mechanics.
!
Credentials
The package declares no required environment variables, but includes credentials.json containing a bearer-style API key (sk_live_...). Embedding a live-looking API key in distributed files is disproportionate and risky: it enables use of the service without separate auth, may expose a paid/support account, and contradicts the declared env/credential model.
Persistence & Privilege
Skill is not force-included (always:false) and does not request system-wide configuration changes. The MCP integration will auto-authenticate if configured — a normal convenience — but that increases the risk if a bundled credential is present.
Scan Findings in Context
[embedded_secret:credentials.json] unexpected: The package contains credentials.json with an apparent API key ("sk_live_..."). The skill declares no required env vars; bundling a secret is unexpected and may allow unauthorized use, billing abuse, or leakage of the key to others. Treat this as a sensitive credential disclosure.
[openapi_bearer_auth] expected: openapi.yaml defines BearerAuth and a query-parameter fallback for the API key. This is consistent with the skill's described API authentication, but the fallback encourages insecure URL key usage.
What to consider before installing
Before installing, consider that the skill will send URLs and article content to https://readgzh.site (their API/MCP server) and that the package includes an embedded API key (credentials.json). That embedded key is unexpected and risky: it could be abused, may be a shared/demo key, and could expose your activity or cause billing/abuse on the owner's account. Ask the publisher to remove the credentials file (or provide clear justification), rotate the exposed key, and instead require users to supply their own API key via a documented env var. If you don't trust readgzh.site, do not configure MCP or use the bundled key; use your own API key and review privacy/retention terms (shared cache means read articles may be stored and available to other users). If you proceed, monitor usage of any API key you supply and avoid passing sensitive or private article URLs you don't want stored on a third-party cache.

Like a lobster shell, security has layers — review code before you run it.

aivk97dz26ynpzknefqb7k114jvms82kqn5ai=1.3.0vk97dyq722cqyjtxrk89c2x9ty584vzabchinavk97dz26ynpzknefqb7k114jvms82kqn5china=1.3.0vk97dyq722cqyjtxrk89c2x9ty584vzablatestvk9787t4dm2n8t6zcjeeqmq7rqs8587mtlatest=1.3.7vk97dyq722cqyjtxrk89c2x9ty584vzabreadingvk97dz26ynpzknefqb7k114jvms82kqn5reading=1.3.0vk97dyq722cqyjtxrk89c2x9ty584vzabscrapingvk97dz26ynpzknefqb7k114jvms82kqn5scraping=1.3.0vk97dyq722cqyjtxrk89c2x9ty584vzabwechatvk97dz26ynpzknefqb7k114jvms82kqn5wechat=1.3.0vk97dyq722cqyjtxrk89c2x9ty584vzab
611downloads
2stars
6versions
Updated 1h ago
v1.3.9
MIT-0

ReadGZH — 微信公众号文章 AI 阅读器

ReadGZH 是一款专为 AI 智能体设计的微信公众号内容解析工具。它通过服务端代理绕过微信的反爬虫机制,将复杂的公众号 HTML 转换为纯净、结构化的 Markdown 内容,大幅节省 Token 消耗。

⚠️ 使用前必读

情况限制解决方案
匿名用户(无 API Key)每天仅 10 次/IP,超出触发 429注册获取 API Key
带 API Key每日 50 credits(需在控制台领取)/ Pro 最高 2000/月-H "Authorization: Bearer sk_live_..."
缓存文章已转换过的文章再次读取完全免费,不限次数

429 错误原因:匿名请求超出 IP 频率限制(每天10次),不是服务故障, 注册 API Key 即可解决 → 免费注册 →


快速开始

直接对你的 AI 助手下令:

"帮我读一下这篇文章:[微信公众号链接]"


接入方式选择

方式一:MCP 协议(推荐)

OpenClaw 原生支持 MCP,配置后自动处理认证,无需手动注入 Header。

在 OpenClaw config 中添加:

{
  "mcpServers": {
    "readgzh": {
      "url": "https://api.readgzh.site/mcp-server"
    }
  }
}

重启 OpenClaw 后,ReadGZH 的 readgzh.read 等工具会自动被发现并使用。

可用 MCP Tools:

  • readgzh.read — 读取微信文章(URL)
  • readgzh.get — 通过 slug 获取已缓存文章
  • readgzh.search — 按关键词搜索已缓存文章
  • readgzh.list — 列出最近缓存的文章

方式二:REST API + curl

直接用 curl 调用(需手动加 Bearer Token):

# 基础调用(需 Bearer 认证)
curl "https://api.readgzh.site/rd?url=https://mp.weixin.qq.com/s/xxxxx" \
  -H "Authorization: Bearer <YOUR_API_KEY>"

# POST 方式(抓取并缓存)
curl -X POST "https://api.readgzh.site/rd" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -d '{"url": "https://mp.weixin.qq.com/s/xxxxx"}'

响应示例(POST):

{
  "success": true,
  "articleId": "abc123-...",
  "slug": "s/article-title",
  "cached": false,
  "data": {
    "title": "文章标题",
    "author": "作者名",
    "content": "文章纯文本内容...",
    "publishTime": "2025-01-01",
    "sourceUrl": "https://mp.weixin.qq.com/s/..."
  }
}

方式三:browser 读取(无需 API Key)

如果 OpenClaw 有 Chrome 会话(SS实战舱),可以直接用 CDP 打开微信文章 URL 读取 DOM,不需要 API:

browser(action=open, url="https://mp.weixin.qq.com/s/xxxxx")

API 端点一览

端点方式说明
GET /rd?url=...直接抓取并返回 HTML适合 AI 直接读取
POST /rd抓取并缓存返回 JSON(含 articleId + slug)
GET /rd?s=slug读取已缓存文章返回 HTML
GET /rd?id=...通过 ID 读取返回 HTML
GET /rd?s=...&mode=summaryAI 智能摘要(Pro 专属)返回结构化摘要 JSON
POST https://api.readgzh.site/mcp-serverMCP 协议JSON-RPC 2.0,自动认证

Credits 说明

类型消耗说明
简单文章(< 5 图)1 credit
复杂文章(≥ 5 图)2 credits
缓存文章读取免费不限次数
免费额度30 credits/天需在控制台点击「领取今日积分」
Pro 订阅最高 2000/月自动发放,无需领取

错误码

状态码含义
401未提供 API Key 或 Key 无效
402API Key 积分已用完,需充值
403功能需要 Pro 套餐(如 ?mode=summary
429匿名请求超出每天 10 次限制,注册 API Key 即可解决

开发者信息

Comments

Loading comments...