Chinese Search

中文搜索增强，整合必应中国、搜狗微信等中文搜索引擎 (MVP v1.0)

Security Scan

The skill's code, instructions, and requirements are consistent with a simple web-search/scraping helper that uses curl to query Chinese search engines; it asks for no credentials, has no install steps, and its behavior matches its description.

Purpose & Capability

Name/description claim a Chinese search aggregator and the included files only require curl and show curl-based queries to Bing CN, Sogou WeChat, Baidu, Zhihu, etc. The requested capabilities (none beyond curl) are proportionate to the stated purpose.

Instruction Scope

SKILL.md and test-search.sh instruct the agent to run curl against public search engine endpoints and parse HTML; this is within the claimed scope but effectively encourages basic web scraping (including use of User-Agent, delays, and proxies). The instructions do not ask for unrelated files, credentials, or contact external endpoints beyond known search engines. Be aware scraping may trigger rate limits or violate site ToS.

Install Mechanism

No install spec is provided (instruction-only skill). This is low-risk: nothing is downloaded or written by an installer. The only runtime dependency declared is curl, which aligns with the scripts.

Credentials

The skill requests no environment variables, no credentials, and no config paths. That is appropriate for simple HTTP queries to public search engines.

Persistence & Privilege

always is false and model invocation is not disabled (normal). The skill does not request permanent system presence or modify other skills/config; no elevated privileges are requested.

Assessment

This skill is internally consistent and low-privilege: it uses curl to fetch HTML from public Chinese search engines and parse results. Before installing, consider: (1) scraping behavior — some sites block or rate-limit automated requests, so add delays and honor robots/TOS; (2) network access — Google-like endpoints may need a proxy in your environment; (3) parsing is brittle — HTML structure can change, so test locally with your typical queries; and (4) although no secrets are requested, be cautious if you later upgrade the skill to add aggregation/monitoring features that might require storage or API keys. If those features are added, re-check required env vars and install steps.

Chinese Search 中文搜索增强

MVP v1.0 - 整合 2 个核心中文搜索引擎，无需 API Key，支持高级搜索语法。

快速使用

基础搜索

# 百度搜索
curl -s "https://www.baidu.com/s?wd=关键词&rn=10"

# 必应中国
curl -s "https://cn.bing.com/search?q=关键词&count=10"

# 搜狗微信搜索
curl -s "https://weixin.sogou.com/weixin?type=2&query=关键词"

高级搜索语法

语法	说明	示例
site:	限定网站	site:zhihu.com AI 教程
filetype:	文件类型	filetype:pdf 研究报告
intitle:	标题包含	intitle:赚钱 技能
inurl:	URL 包含	inurl:blog 教程
""	精确匹配	"OpenClaw 技能开发"
-	排除关键词	AI -广告

搜索引擎列表

✅ v1.0 已实现

1. 必应中国 (Bing CN)

• URL: https://cn.bing.com/search?q={query}
• 特点：国际内容 + 中文
• 适用：技术文档、学术搜索、通用查询
• 状态：✅ 工作正常

2. 搜狗微信 (Sogou WeChat)

• URL: https://weixin.sogou.com/weixin?type=2&query={query}
• 特点：微信公众号文章
• 适用：自媒体内容、行业动态、营销素材
• 状态：✅ 工作正常
• 独特价值: 全网唯一可搜索微信公众号的技能

🚧 计划中 (v1.1-v2.0)

引擎	状态	说明
百度	🚧 优化中	需要 User-Agent/Cookie 处理
知乎	🚧 优化中	需要调整请求格式
360	📋 计划中	可能需要浏览器自动化
头条	📋 计划中	可能需要浏览器自动化
搜狗普通	📋 计划中	后续迭代
谷歌中国	📋 计划中	需要代理支持

实战案例

案例 1: 搜索中文技术教程

# 搜索 OpenClaw 技能开发教程
curl -s "https://www.baidu.com/s?wd=OpenClaw+ 技能开发 + 教程&rn=10"

# 搜索知乎相关讨论
curl -s "https://www.zhihu.com/search?type=content&q=OpenClaw+ 技能"

案例 2: 搜索行业报告

# 搜索 PDF 格式的行业报告
curl -s "https://www.baidu.com/s?wd=AI+ 行业报告+2026+filetype:pdf"

案例 3: 搜索微信公众号文章

# 搜索自媒体相关内容
curl -s "https://weixin.sogou.com/weixin?type=2&query=AI+ 自动化+ 赚钱"

集成到 OpenClaw

安装

npx clawhub@latest install chinese-search

使用

# 调用技能
/search 中文关键词

# 或直接用 curl
curl -s "wttr.in/中文搜索 + 关键词"

注意事项

1. 反爬虫: 部分搜索引擎有频率限制，建议添加延迟
2. 结果解析: 搜索结果需要解析 HTML 提取内容
3. 代理需求: 谷歌搜索需要代理才能访问
4. API 限制: 商业使用建议申请官方 API

扩展方向

1. 结果聚合: 整合多个引擎结果，去重排序
2. 摘要生成: 自动提取搜索结果摘要
3. 监控告警: 关键词监控，有新内容时通知
4. 批量搜索: 支持批量关键词搜索

---

作者: 小鸣 🦞
版本: v1.0
创建时间: 2026-03-29