← 返回 Skills 市场
1531
总下载
1
收藏
5
当前安装
1
版本数
在 OpenClaw 中安装
/install web-anti-crawl-search
功能描述
使用无头浏览器通过多个搜索引擎搜索内容。用于当用户需要搜索最新新闻、时事、股票信息或其他需要实时网络数据的内容时。支持百度、Bing、360、Sogou、微信、今日头条、谷歌等国内外搜索引擎。
使用说明 (SKILL.md)
Web Search Skill
使用无头浏览器(Playwright/Puppeteer)通过多个搜索引擎搜索内容。
搜索引擎列表
国内搜索引擎
- Baidu:
https://www.baidu.com/s?wd={keyword} - Bing CN:
https://cn.bing.com/search?q={keyword}&ensearch=0 - 360:
https://www.so.com/s?q={keyword} - Sogou:
https://sogou.com/web?query={keyword} - WeChat:
https://wx.sogou.com/weixin?type=2&query={keyword} - Toutiao:
https://so.toutiao.com/search?keyword={keyword} - Jisilu:
https://www.jisilu.cn/explore/?keyword={keyword}(基金/理财)
国际搜索引擎
- Google:
https://www.google.com/search?q={keyword} - Google HK:
https://www.google.com.hk/search?q={keyword} - DuckDuckGo:
https://duckduckgo.com/html/?q={keyword} - Yahoo:
https://search.yahoo.com/search?p={keyword} - Startpage:
https://www.startpage.com/sp/search?query={keyword} - Brave:
https://search.brave.com/search?q={keyword} - Ecosia:
https://www.ecosia.org/search?q={keyword} - Qwant:
https://www.qwant.com/?q={keyword} - WolframAlpha:
https://www.wolframalpha.com/input?i={keyword}
使用方法
运行搜索脚本:
cd /Users/wuwei/.openclaw/workspace/skills/web-search/scripts
node search.js \x3Cengine> \x3Ckeyword>
参数说明
engine: 搜索引擎名称(baidu, bing-cn, bing-int, 360, sogou, weixin, toutiao, jisilu, google, google-hk, duckduckgo, yahoo, startpage, brave, ecosia, qwant, wolframalpha)keyword: 搜索关键词(URL编码)
示例
# 搜索百度
node search.js baidu "A股今日走势"
# 搜索DuckDuckGo
node search.js duckduckgo "stock market news"
输出
脚本会打开无头浏览器,访问搜索页面,提取搜索结果标题和链接,并输出为markdown格式。
依赖
需要安装 Playwright:
npm install playwright
npx playwright install chromium
🆕 多搜索引擎对照搜索 (Multi-Search)
使用 multi-search.js 脚本可以同时搜索 Google 和 Bing,进行对照验证:
- ✅ 同时获取 Google 和 Bing 的搜索结果
- ✅ 智能去重合并,标记数据来源
- ✅ 如果 Google 访问失败,自动使用夸克AI搜索替代
- ✅ 支持 JSON 和 Markdown 两种输出格式
使用方法
cd /Users/wuwei/.openclaw/workspace/skills/web-search/scripts
node multi-search.js \x3Ckeyword> [options]
参数说明
| 参数 | 说明 |
|---|---|
keyword |
搜索关键词(必填) |
--google-only |
仅使用 Google 搜索 |
--bing-only |
仅使用 Bing 搜索 |
--quark-only |
仅使用夸克AI搜索 |
--max-results=N |
最多返回 N 条结果(默认 20) |
--format=md|json |
输出格式(默认 md) |
示例
# 基本用法 - 同时搜索 Google + Bing
node multi-search.js "A股今日热点"
# 输出为 JSON 格式
node multi-search.js "人工智能新闻" --format=json
# 获取更多结果
node multi-search.js "新能源汽车政策" --max-results=30
# 仅使用 Bing
node multi-search.js "美股行情" --bing-only
# 仅使用夸克AI(适合国内搜索)
node multi-search.js "国内财经新闻" --quark-only
输出示例(Markdown 格式)
============================================================
🔍 "A股今日热点" 搜索结果(共 15 条)
============================================================
1. A股三大指数集体上涨,科技股领涨
🔗 https://finance.sina.com.cn/...
📌 来源: Google + Bing
2. 新能源汽车板块异动,多只个股涨停
🔗 https://stock.eastmoney.com/...
📌 来源: Bing
3. 北向资金净流入超50亿元
🔗 https://finance.qq.com/...
📌 来源: Google
...
错误处理机制
- Google 访问失败:自动切换到夸克AI搜索
- 所有引擎失败:返回空结果并输出错误信息
- 部分引擎失败:返回成功引擎的结果,并标记失败的引擎
特性对比
| 特性 | Bing | 夸克AI | |
|---|---|---|---|
| 国内访问 | ⚠️ 不稳定 | ✅ 稳定 | ✅ 稳定 |
| 中文搜索 | ✅ 好 | ✅ 好 | ✅ 优秀 |
| 实时新闻 | ✅ 快 | ✅ 快 | ✅ 快 |
| 英文搜索 | ✅ 优秀 | ✅ 好 | ⚠️ 一般 |
推荐使用场景
- 默认模式(Google + Bing):获取最全面的结果
- 仅 Bing:国内访问稳定,中文搜索优秀
- 仅 夸克AI:国内搜索首选,中文理解好
安全使用建议
This skill appears to do what it claims (headless-browser scraping across many search engines). Before installing or running it, consider:
- Stealth/anti-crawl: The code uses a stealth plugin and user-agent spoofing to evade anti-bot measures. That is plausible for crawling but may violate search engines' terms of service and could get your IP blocked. Only use in environments where you accept that risk.
- Package/supply-chain risk: You must run npm install and npx playwright install which will download many packages and browser binaries. Review package-lock.json and run in an isolated environment (container/VM) if you are concerned about supply-chain or untrusted code.
- Local path in SKILL.md: The README uses a hardcoded local path (/Users/wuwei/...), update to your own path before running to avoid confusion.
- Privacy/network: The scripts visit external sites and output results to console — they do not appear to exfiltrate data to hidden endpoints, but if you paste sensitive queries into the tool those search terms will be sent to third-party search engines.
Recommended actions: inspect the package-lock.json, run npm install only in an isolated environment, consider using official search APIs (Bing/Google) if you need stable/legal access, and be mindful of legal/ToS implications for scraping/anti-crawl evasion.
功能分析
Type: OpenClaw Skill
Name: web-anti-crawl-search
Version: 1.0.0
The skill bundle provides legitimate web searching capabilities using Playwright and stealth plugins to aggregate results from multiple search engines (Google, Bing, Baidu, etc.). The scripts (search.js and multi-search.js) contain standard web scraping logic for DOM extraction and result deduplication, with no evidence of data exfiltration, malicious execution, or harmful prompt injection instructions.
能力评估
Purpose & Capability
Name/description (headless browser multi-engine search) align with the included files: search.js and multi-search.js implement Playwright-based scraping of many search engines and news sites. Required binaries/env vars are none, which is consistent because the code depends on npm packages instead.
Instruction Scope
SKILL.md and the scripts instruct the agent/user to launch headless browsers, visit many public search/news URLs, extract titles/links, and output results. The instructions do not ask to read unrelated local files or environment variables. Two points to note: SKILL.md contains a hardcoded user-local path (cd /Users/wuwei/...), which is odd but not harmful, and the runtime code intentionally tries to evade anti-bot detection (stealth plugin + user-agent spoofing), which is scope creep away from a benign 'query an API' approach and could violate target sites' terms of service.
Install Mechanism
There is no automatic install spec in the registry entry — the README asks the user to run npm install and npx playwright install. The repository includes package.json and package-lock.json referencing common npm packages (playwright, playwright-extra, puppeteer-extra-plugin-stealth). This is a normal but higher-risk pattern than an instruction-only skill because it requires pulling many third-party packages and browser binaries (supply-chain risk). No downloads from arbitrary personal servers are present.
Credentials
The skill declares no required environment variables, credentials, or config paths and the code does not read any secrets or unrelated env vars. The lack of requested credentials is proportionate to its purpose.
Persistence & Privilege
always:false and user-invocable:true (default) — the skill does not request persistent/global privileges or modify other skills. It does not attempt to enable itself autonomously beyond the platform default.
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 web-anti-crawl-search - 安装完成后,直接呼叫该 Skill 的名称或使用
/web-anti-crawl-search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
v1.0.0
Initial release of the "web-anti-crawl-search" skill:
- Enables real-time web searching via multiple search engines using a headless browser.
- Supports both domestic (Baidu, Bing CN, 360, Sogou, WeChat, Toutiao, Jisilu) and international (Google, Google HK, DuckDuckGo, Yahoo, Startpage, Brave, Ecosia, Qwant, WolframAlpha) search engines.
- Provides scripts for both single-engine searches and multi-engine (Google + Bing + QuarkAI fallback) comparative searches.
- Outputs search results in Markdown or JSON format.
- Handles engine failures gracefully with automatic fallback and error marking.
- Includes detailed usage instructions and sample commands for end users.
元数据
常见问题
web-search 是什么?
使用无头浏览器通过多个搜索引擎搜索内容。用于当用户需要搜索最新新闻、时事、股票信息或其他需要实时网络数据的内容时。支持百度、Bing、360、Sogou、微信、今日头条、谷歌等国内外搜索引擎。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1531 次。
如何安装 web-search?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install web-anti-crawl-search」即可一键安装,无需额外配置。
web-search 是免费的吗?
是的,web-search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
web-search 支持哪些平台?
web-search 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 web-search?
由 wei.wu(@dlutwuwei)开发并维护,当前版本 v1.0.0。
推荐 Skills