web-search
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseWeb Search Skill
Web Search Skill
When to Use This Skill
何时使用此Skill
Use the web-search skill when you need:
- Current information - Events, news, or data after January 2025
- Latest documentation - Up-to-date framework/library docs (React 19, Next.js 15, etc.)
- Real-time data - Stock prices, weather, sports scores, etc.
- Fact verification - Check current status of projects, companies, or technologies
- Recent discussions - Community opinions, GitHub issues, Stack Overflow answers
- Product comparisons - Latest reviews and comparisons
- Troubleshooting - Search for specific error messages or solutions
Examples of when to use:
- User: "What are the new features in React 19?"
- User: "Search for the latest Next.js App Router documentation"
- User: "What's the current status of the Rust async project?"
- User: "Find recent discussions about Vue 3 performance"
当你需要以下信息时,请使用web-search skill:
- 当前信息 - 2025年1月之后的事件、新闻或数据
- 最新文档 - 最新的框架/库文档(React 19、Next.js 15等)
- 实时数据 - 股票价格、天气、体育比分等
- 事实核查 - 检查项目、公司或技术的当前状态
- 近期讨论 - 社区观点、GitHub问题、Stack Overflow答案
- 产品对比 - 最新评测和对比
- 故障排查 - 搜索特定错误信息或解决方案
使用示例场景:
- 用户:"React 19有哪些新功能?"
- 用户:"搜索最新的Next.js App Router文档"
- 用户:"Rust异步项目的当前状态是什么?"
- 用户:"查找关于Vue 3性能的近期讨论"
How It Works
工作原理
┌──────────┐ Bash ┌─────────┐ HTTP ┌──────────────┐ CDP ┌────────┐
│ Claude │───────────▶│ CLI.sh │───────────▶│ Bridge Server│──────────▶│ Chrome │
│ │ │ │ │ (localhost) │ Playwright│ │
└──────────┘ └─────────┘ └──────────────┘ └────────┘
│
▼
Google/Bing Search
Extract ResultsArchitecture:
- CLI Script - Simple bash interface for Claude
- Bridge Server - Express HTTP API (auto-started by Electron)
- Playwright Manager - Browser connection and session management
- Search Engine Layer - Google primary, Bing fallback
- Chrome Browser - Visible browser window (all operations transparent)
┌──────────┐ Bash ┌─────────┐ HTTP ┌──────────────┐ CDP ┌────────┐
│ Claude │───────────▶│ CLI.sh │───────────▶│ Bridge Server│──────────▶│ Chrome │
│ │ │ │ │ (localhost) │ Playwright│ │
└──────────┘ └─────────┘ └──────────────┘ └────────┘
│
▼
Google/Bing Search
Extract Results架构:
- CLI脚本 - 为Claude设计的简单bash接口
- 桥接服务器 - Express HTTP API(由Electron自动启动)
- Playwright管理器 - 浏览器连接和会话管理
- 搜索引擎层 - 优先使用Google,备用Bing
- Chrome浏览器 - 可视化浏览器窗口(所有操作透明可见)
Basic Usage
基础用法
Simple Search (Recommended)
简单搜索(推荐)
IMPORTANT: Always use the environment variable to locate the skill scripts. This ensures the skill works in both development and production environments.
$SKILLS_ROOTbash
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "search query" [max_results]For non-ASCII queries (Chinese/Japanese/etc.), prefer UTF-8 file input to avoid shell encoding issues on Windows:
bash
cat > /tmp/web-query.txt <<'TXT'
苹果 Siri AI 2026 发布计划
TXT
bash "$SKILLS_ROOT/web-search/scripts/search.sh" @/tmp/web-query.txt 10Examples:
bash
undefined重要提示: 请始终使用环境变量定位技能脚本。这能确保该Skill在开发和生产环境中都能正常工作。
$SKILLS_ROOTbash
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "搜索查询" [max_results]对于非ASCII查询(中文/日文等),建议使用UTF-8文件输入,以避免Windows系统上的Shell编码问题:
bash
cat > /tmp/web-query.txt <<'TXT'
苹果 Siri AI 2026 发布计划
TXT
bash "$SKILLS_ROOT/web-search/scripts/search.sh" @/tmp/web-query.txt 10示例:
bash
undefinedSearch with default 10 results
使用默认10条结果进行搜索
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "TypeScript 5.0 new features"
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "TypeScript 5.0 new features"
Limit to 5 results
限制为5条结果
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "React Server Components guide" 5
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "React Server Components guide" 5
Search for recent news
搜索近期新闻
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "AI news January 2026" 10
**Output Format:**
The script returns Markdown-formatted results:
```markdownbash "$SKILLS_ROOT/web-search/scripts/search.sh" "AI news January 2026" 10
**输出格式:**
脚本返回Markdown格式的结果:
```markdownSearch Results: TypeScript 5.0 new features
Search Results: TypeScript 5.0 new features
Query: TypeScript 5.0 new features
Results: 5
Time: 834ms
Query: TypeScript 5.0 new features
Results: 5
Time: 834ms
TypeScript 5.0 Release Notes
TypeScript 5.0 Release Notes
URL: [https://www.typescriptlang.org/docs/...]
TypeScript 5.0 introduces decorators, const type parameters...
URL: [https://www.typescriptlang.org/docs/...]
TypeScript 5.0 introduces decorators, const type parameters...
(More results...)
(更多结果...)
undefinedundefinedWorkflow Example
工作流示例
bash
undefinedbash
undefined1. Search for topic
1. 搜索主题
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js 14 features" 5
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js 14 features" 5
2. Analyze results and answer user
2. 分析结果并回复用户
3. Follow-up search if needed
3. 如有需要,进行后续搜索
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js Server Actions tutorial" 3
undefinedbash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js Server Actions tutorial" 3
undefinedAdvanced Usage
高级用法
Server Management
服务器管理
The Bridge Server is automatically managed by Electron. You typically don't need to start/stop it manually.
However, for manual control:
bash
undefined桥接服务器由Electron自动管理。通常你不需要手动启动/停止它。
不过,如需手动控制:
bash
undefinedStart server (if not already running)
启动服务器(如果尚未运行)
bash "$SKILLS_ROOT/web-search/scripts/start-server.sh"
bash "$SKILLS_ROOT/web-search/scripts/start-server.sh"
Stop server
停止服务器
bash "$SKILLS_ROOT/web-search/scripts/stop-server.sh"
bash "$SKILLS_ROOT/web-search/scripts/stop-server.sh"
Check health (start script will print endpoint status)
检查健康状态(启动脚本会打印端点状态)
bash "$SKILLS_ROOT/web-search/scripts/start-server.sh"
undefinedbash "$SKILLS_ROOT/web-search/scripts/start-server.sh"
undefinedDirect API Calls
直接API调用
For advanced use cases, you can call the HTTP API directly:
bash
undefined对于高级使用场景,你可以直接调用HTTP API:
bash
undefinedGet or create connection
获取或创建连接
CONNECTION_ID=$(curl -s -X POST http://127.0.0.1:8923/api/browser/connect
-H "Content-Type: application/json"
-d '{}' | grep -o '"connectionId":"[^"]*"' | cut -d'"' -f4)
-H "Content-Type: application/json"
-d '{}' | grep -o '"connectionId":"[^"]*"' | cut -d'"' -f4)
CONNECTION_ID=$(curl -s -X POST http://127.0.0.1:8923/api/browser/connect
-H "Content-Type: application/json"
-d '{}' | grep -o '"connectionId":"[^"]*"' | cut -d'"' -f4)
-H "Content-Type: application/json"
-d '{}' | grep -o '"connectionId":"[^"]*"' | cut -d'"' -f4)
Perform search
执行搜索
curl -X POST http://127.0.0.1:8923/api/search
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "query": "Playwright tutorial", "maxResults": 5 }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "query": "Playwright tutorial", "maxResults": 5 }"
curl -X POST http://127.0.0.1:8923/api/search
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "query": "Playwright tutorial", "maxResults": 5 }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "query": "Playwright tutorial", "maxResults": 5 }"
Navigate to specific URL
导航到特定URL
curl -X POST http://127.0.0.1:8923/api/page/navigate
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "url": "https://example.com\" }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "url": "https://example.com\" }"
curl -X POST http://127.0.0.1:8923/api/page/navigate
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "url": "https://example.com\" }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "url": "https://example.com\" }"
Take screenshot
截取屏幕截图
curl -X POST http://127.0.0.1:8923/api/page/screenshot
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "format": "png" }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "format": "png" }"
undefinedcurl -X POST http://127.0.0.1:8923/api/page/screenshot
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "format": "png" }"
-H "Content-Type: application/json"
-d "{ "connectionId": "$CONNECTION_ID", "format": "png" }"
undefinedBest Practices
最佳实践
1. Use Specific Queries
1. 使用具体的查询词
❌ Bad:
✅ Good:
bash scripts/search.sh "react"bash scripts/search.sh "React 19 new features and breaking changes"❌ 不佳:
✅ 良好:
bash scripts/search.sh "react"bash scripts/search.sh "React 19 new features and breaking changes"2. Limit Results Appropriately
2. 合理限制结果数量
- Quick lookup: 3-5 results
- Comprehensive research: 10 results
- Don't request more than needed (faster + less noise)
- 快速查询:3-5条结果
- 全面研究:10条结果
- 不要请求超出需要的数量(更快+更少干扰信息)
3. Check Server Status First
3. 先检查服务器状态
If search fails, verify the server is running:
bash
bash "$SKILLS_ROOT/web-search/scripts/start-server.sh" || echo "Server not running"如果搜索失败,请验证服务器是否在运行:
bash
bash "$SKILLS_ROOT/web-search/scripts/start-server.sh" || echo "服务器未运行"4. Reuse Connections
4. 复用连接
The CLI script automatically caches connections. Multiple searches in the same session will reuse the same browser connection for better performance.
CLI脚本会自动缓存连接。同一会话中的多次搜索会复用同一个浏览器连接,以提升性能。
5. Clean Output
5. 整理输出内容
Parse the Markdown output to extract key information for the user. Don't just dump all results - synthesize and summarize.
解析Markdown输出,为用户提取关键信息。不要直接输出所有结果——要进行综合和总结。
Common Patterns
常见模式
Pattern 1: Latest Documentation
模式1:最新文档查询
bash
undefinedbash
undefinedUser asks about latest framework features
用户询问最新框架特性
bash SKILLs/web-search/scripts/search.sh "Next.js 15 documentation" 5
bash SKILLs/web-search/scripts/search.sh "Next.js 15 documentation" 5
Parse results, find official docs, summarize features
解析结果,找到官方文档,总结特性
undefinedundefinedPattern 2: Troubleshooting
模式2:故障排查
bash
undefinedbash
undefinedUser reports an error
用户报告错误
bash SKILLs/web-search/scripts/search.sh "TypeError: Cannot read property of undefined React" 5
bash SKILLs/web-search/scripts/search.sh "TypeError: Cannot read property of undefined React" 5
Find Stack Overflow answers and GitHub issues, provide solution
查找Stack Overflow答案和GitHub问题,提供解决方案
undefinedundefinedPattern 3: Current Events
模式3:当前事件查询
bash
undefinedbash
undefinedUser asks about recent news
用户询问近期新闻
bash SKILLs/web-search/scripts/search.sh "AI developments January 2026" 10
bash SKILLs/web-search/scripts/search.sh "AI developments January 2026" 10
Summarize key news items from results
从结果中总结关键新闻事件
undefinedundefinedPattern 4: Comparison Research
模式4:对比研究
bash
undefinedbash
undefinedUser wants to compare technologies
用户想要对比技术
bash SKILLs/web-search/scripts/search.sh "Vue 3 vs React 18 performance 2026" 5
bash SKILLs/web-search/scripts/search.sh "Vue 3 vs React 18 performance 2026" 5
Synthesize comparison from multiple sources
综合多个来源的对比信息
undefinedundefinedPattern 5: API/Library Usage
模式5:API/库用法查询
bash
undefinedbash
undefinedUser needs specific API documentation
用户需要特定API文档
bash SKILLs/web-search/scripts/search.sh "Playwright page.evaluate examples" 5
bash SKILLs/web-search/scripts/search.sh "Playwright page.evaluate examples" 5
Extract code examples and usage patterns
提取代码示例和使用模式
undefinedundefinedError Handling
错误处理
Server Not Running
服务器未运行
Error:
✗ Bridge Server is not runningSolution:
- The server should auto-start with Electron
- If manual start needed:
bash SKILLs/web-search/scripts/start-server.sh - Check logs:
cat SKILLs/web-search/.server.log
错误信息:
✗ Bridge Server is not running解决方案:
- 服务器应随Electron自动启动
- 如需手动启动:
bash SKILLs/web-search/scripts/start-server.sh - 查看日志:
cat SKILLs/web-search/.server.log
Browser Launch Failed
浏览器启动失败
Error:
Failed to launch browserCause: Chrome not installed or not found
Solution:
- macOS: Install from https://www.google.com/chrome/
- Linux:
sudo apt install chromium-browser - Windows: Install from https://www.google.com/chrome/
错误信息:
Failed to launch browser原因: Chrome未安装或未找到
解决方案:
- macOS:从https://www.google.com/chrome/安装
- Linux:
sudo apt install chromium-browser - Windows:从https://www.google.com/chrome/安装
Connection Timeout
连接超时
Error: or
CDP port not readyConnection timeoutSolution:
bash
undefined错误信息: 或
CDP port not readyConnection timeout解决方案:
bash
undefinedStop server
停止服务器
bash SKILLs/web-search/scripts/stop-server.sh
bash SKILLs/web-search/scripts/stop-server.sh
Clear cache
清除缓存
rm SKILLs/web-search/.connection
rm SKILLs/web-search/.connection
Restart
重启
bash SKILLs/web-search/scripts/start-server.sh
undefinedbash SKILLs/web-search/scripts/start-server.sh
undefinedNo Search Results
无搜索结果
Error:
Found 0 resultsPossible causes:
- Query too specific or unusual
- Bing changed page structure (rare)
- Network issues
Solution:
- Try broader query
- Check internet connection
- Verify page loads manually at bing.com
错误信息:
Found 0 results可能原因:
- 查询词过于具体或特殊
- Bing页面结构变更(罕见)
- 网络问题
解决方案:
- 尝试更宽泛的查询词
- 检查网络连接
- 手动在bing.com验证页面是否能加载
Search Timeout
搜索超时
Error:
Search failed: timeoutSolution:
- Check internet connection
- Reduce max results
- Try again (might be temporary network issue)
错误信息:
Search failed: timeout解决方案:
- 检查网络连接
- 减少最大结果数量
- 重试(可能是临时网络问题)
Understanding Results
理解搜索结果
Result Structure
结果结构
Each search result contains:
markdown
undefined每个搜索结果包含:
markdown
undefined[Title of Result]
[结果标题]
URL: [https://example.com/page]
[Snippet/description from search results]
**Fields:**
- **Title** - Page/article title
- **URL** - Direct link (may include Bing tracking)
- **Snippet** - Preview text from the pageURL: [https://example.com/page]
[搜索结果中的片段/描述]
**字段说明:**
- **标题** - 页面/文章标题
- **URL** - 直接链接(可能包含Bing跟踪参数)
- **片段** - 页面的预览文本Parsing Results
解析结果
The search output is Markdown. Extract:
- Total results count
- Search duration
- Individual result titles and URLs
- Snippets for context
搜索输出为Markdown格式。提取以下内容:
- 结果总数
- 搜索耗时
- 单个结果的标题和URL
- 用于上下文的片段内容
Result Quality
结果质量
- Official docs - Usually appear in top 3 results
- Stack Overflow - Appears for technical questions
- Recent articles - Bing prioritizes recent content
- Chinese content - Bing works well in China, includes Chinese sources
- 官方文档 - 通常出现在前3条结果中
- Stack Overflow - 针对技术问题的结果
- 近期文章 - Bing优先展示近期内容
- 中文内容 - Bing在中国地区表现良好,包含中文来源
Performance Considerations
性能考量
Typical Latencies
典型延迟
- Server startup: ~2 seconds (one-time, auto-started)
- Browser launch: ~3 seconds (one-time, persists across searches)
- First search: ~2-3 seconds (includes browser connection setup)
- Subsequent searches: ~1 second (browser and connection reused)
- 服务器启动:约2秒(一次性,自动启动)
- 浏览器启动:约3秒(一次性,跨搜索持续运行)
- 首次搜索:约2-3秒(包含浏览器连接设置)
- 后续搜索:约1秒(复用浏览器和连接)
Optimization Tips
优化技巧
- Browser persistence - The browser stays alive across searches by default. Connections and pages are reused automatically. Set to close the browser after each search if needed.
WEB_SEARCH_CLEANUP=1 - Limit results - Request only what you need (5-10 is usually enough)
- Batch searches - If multiple searches needed, do them consecutively to reuse connection
- Specific queries - More specific = faster and better results
- 浏览器持久化 - 默认情况下,浏览器会在多次搜索之间保持运行。连接和页面会自动复用。如果需要每次搜索后关闭浏览器,设置。
WEB_SEARCH_CLEANUP=1 - 限制结果数量 - 仅请求所需的数量(5-10条通常足够)
- 批量搜索 - 如果需要多次搜索,连续执行以复用连接
- 使用具体查询词 - 更具体的查询词=更快的速度和更好的结果
Security and Privacy
安全与隐私
Security Measures
安全措施
- Localhost only - Bridge Server binds to 127.0.0.1 (no external access)
- No network exposure - Not accessible from other machines
- Isolated browser - Uses separate Chrome profile, won't affect user's main browser
- Visible operations - All browser actions shown in visible window (transparent)
- No credentials - Skill never handles passwords or sensitive data
- 仅本地访问 - 桥接服务器绑定到127.0.0.1(无外部访问权限)
- 无网络暴露 - 无法从其他机器访问
- 隔离浏览器 - 使用独立的Chrome配置文件,不会影响用户的主浏览器
- 操作可见 - 所有浏览器操作在可视化窗口中显示(完全透明)
- 无凭据处理 - 该Skill从不处理密码或敏感数据
Privacy Considerations
隐私注意事项
- Search queries go through Google and/or Bing depending on availability
- Google/Bing may track searches (their standard privacy policies apply)
- No local storage of search history by the skill
- User can observe all browser activity in real-time
- 搜索查询会通过Google和/或Bing(取决于可用性)进行
- Google/Bing可能会跟踪搜索(遵循其标准隐私政策)
- 该Skill不会本地存储搜索历史
- 用户可以实时观察所有浏览器活动
Limitations
局限性
Current Limitations
当前局限性
- No CAPTCHA handling - If Google or Bing shows CAPTCHA, user must solve manually
- Engine availability varies by network/region - Auto mode falls back between Google and Bing
- English/Chinese focus - Optimized for English and Chinese results
- Basic extraction - Extracts titles and snippets, not full page content
- No authentication - Cannot search pages requiring login
- 无CAPTCHA处理 - 如果Google或Bing显示CAPTCHA,用户必须手动解决
- 搜索引擎可用性因网络/地区而异 - 自动模式会在Google和Bing之间切换
- 中英文优先 - 针对英文和中文结果优化
- 基础提取 - 仅提取标题和片段,不提取完整页面内容
- 无认证功能 - 无法搜索需要登录的页面
Not Suitable For
不适合的场景
- Searches requiring authentication
- Filling out forms or submitting data
- Actions requiring CAPTCHA solving (unless user manually solves)
- Mass scraping or automated bulk searches
- Accessing pages behind paywalls
- 需要认证的搜索
- 填写表单或提交数据
- 需要解决CAPTCHA的操作(除非用户手动解决)
- 大规模抓取或自动化批量搜索
- 访问付费墙后的页面
Troubleshooting Guide
故障排查指南
Quick Diagnostics
快速诊断
bash
undefinedbash
undefined1. Check server health
1. 检查服务器健康状态
2. Check server logs
2. 查看服务器日志
cat SKILLs/web-search/.server.log | tail -50
cat SKILLs/web-search/.server.log | tail -50
3. Test basic search
3. 测试基础搜索
bash SKILLs/web-search/scripts/search.sh "test" 1
bash SKILLs/web-search/scripts/search.sh "test" 1
4. Check Chrome installation
4. 检查Chrome安装情况
which google-chrome || which chromium || which chromium-browser
undefinedwhich google-chrome || which chromium || which chromium-browser
undefinedCommon Issues
常见问题
| Issue | Symptom | Solution |
|---|---|---|
| Server down | | Start server or restart Electron |
| Browser missing | | Install Chrome/Chromium |
| Port conflict | | Stop conflicting process on port 8923 |
| Stale connection | | Remove |
| Network issue | | Check internet connection |
| 问题 | 症状 | 解决方案 |
|---|---|---|
| 服务器宕机 | | 启动服务器或重启Electron |
| 浏览器缺失 | | 安装Chrome/Chromium |
| 端口冲突 | | 停止端口8923上的冲突进程 |
| 连接失效 | | 删除 |
| 网络问题 | | 检查网络连接 |
Reset Everything
完全重置
If all else fails, full reset:
bash
cd SKILLs/web-search如果所有方法都无效,执行完全重置:
bash
cd SKILLs/web-searchStop server
停止服务器
bash scripts/stop-server.sh
bash scripts/stop-server.sh
Clean cache and state
清除缓存和状态
rm -f .connection .server.pid .server.log
rm -f .connection .server.pid .server.log
Rebuild
重新构建
npm run build
npm run build
Restart
重启
bash scripts/start-server.sh
bash scripts/start-server.sh
Test
测试
bash scripts/search.sh "test" 1
undefinedbash scripts/search.sh "test" 1
undefinedExamples for Claude
面向Claude的示例
Example 1: User Asks About Latest Framework
示例1:用户询问最新框架特性
User: "What are the new features in Next.js 15?"
Claude's approach:
bash
undefined用户: "Next.js 15有哪些新功能?"
Claude的处理方式:
bash
undefinedSearch for Next.js 15 features
搜索Next.js 15特性
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js 15 new features" 5
**Then:** Parse results, identify official Next.js blog/docs, summarize key features for user.bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Next.js 15 new features" 5
**后续步骤:** 解析结果,识别Next.js官方博客/文档,为用户总结关键特性。Example 2: Troubleshooting Error
示例2:故障排查
User: "I'm getting 'Cannot find module' error in TypeScript"
Claude's approach:
bash
undefined用户: "我在TypeScript中遇到了'Cannot find module'错误"
Claude的处理方式:
bash
undefinedSearch for the specific error
搜索特定错误的解决方案
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "TypeScript Cannot find module error solution" 5
**Then:** Extract solutions from Stack Overflow results, provide step-by-step fix.bash "$SKILLS_ROOT/web-search/scripts/search.sh" "TypeScript Cannot find module error solution" 5
**后续步骤:** 从Stack Overflow结果中提取解决方案,提供分步修复指南。Example 3: Current Events
示例3:当前事件查询
User: "What happened in AI this month?"
Claude's approach:
bash
undefined用户: "本月AI领域有什么动态?"
Claude的处理方式:
bash
undefinedSearch for recent AI news
搜索近期AI新闻
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "AI news January 2026" 10
**Then:** Synthesize news from multiple sources, provide summary of key events.bash "$SKILLS_ROOT/web-search/scripts/search.sh" "AI news January 2026" 10
**后续步骤:** 综合多个来源的新闻,为用户提供关键事件总结。Example 4: Documentation Lookup
示例4:文档查询
User: "How do I use React Server Components?"
Claude's approach:
bash
undefined用户: "如何使用React Server Components?"
Claude的处理方式:
bash
undefinedSearch for RSC documentation and tutorials
搜索RSC文档和教程
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "React Server Components guide tutorial" 5
**Then:** Find official React docs and good tutorials, explain with examples.bash "$SKILLS_ROOT/web-search/scripts/search.sh" "React Server Components guide tutorial" 5
**后续步骤:** 找到官方React文档和优质教程,结合示例进行讲解。Example 5: Comparison Research
示例5:对比研究
User: "Should I use Vite or webpack in 2026?"
Claude's approach:
bash
undefined用户: "2026年我应该使用Vite还是webpack?"
Claude的处理方式:
bash
undefinedSearch for recent comparisons
搜索近期对比内容
bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Vite vs webpack 2026 comparison" 5
**Then:** Analyze multiple perspectives, provide balanced recommendation.bash "$SKILLS_ROOT/web-search/scripts/search.sh" "Vite vs webpack 2026 comparison" 5
**后续步骤:** 分析多个视角,提供平衡的推荐建议。Tips for Effective Use
有效使用技巧
- Be specific in queries - Include version numbers, dates, or specific aspects
- Parse results carefully - Don't just copy-paste, synthesize information
- Verify with multiple sources - Cross-check important information
- Cite sources - Tell user which sources you're using
- Explain limitations - If search doesn't find good results, tell user
- Use follow-up searches - One search might not be enough, do multiple if needed
- Check result dates - Prefer recent articles for current info
- Look for official sources - Prioritize official docs and authoritative sources
- 查询词要具体 - 包含版本号、日期或特定方面
- 仔细解析结果 - 不要直接复制粘贴,要综合信息
- 多源验证 - 对重要信息进行交叉核对
- 注明来源 - 告诉用户你使用的来源
- 说明局限性 - 如果搜索未找到优质结果,告知用户
- 使用后续搜索 - 一次搜索可能不够,如有需要进行多次搜索
- 检查结果日期 - 优先选择近期文章获取当前信息
- 寻找官方来源 - 优先考虑官方文档和权威来源
Technical Details
技术细节
Technologies Used
使用的技术
- Playwright Core - Browser automation framework
- Chrome DevTools Protocol - Low-level browser control
- Express.js - HTTP API server
- Google + Bing Search - Multi-engine fallback search strategy
- Bash Scripts - Simple CLI interface
- Playwright Core - 浏览器自动化框架
- Chrome DevTools Protocol - 底层浏览器控制
- Express.js - HTTP API服务器
- Google + Bing Search - 多搜索引擎备用策略
- Bash脚本 - 简单的CLI接口
System Requirements
系统要求
- Node.js 18+
- Google Chrome or Chromium installed
- Internet connection for searches
- ~100MB RAM for Bridge Server
- ~200MB RAM for Chrome instance
- Node.js 18+
- 已安装Google Chrome或Chromium
- 用于搜索的互联网连接
- 约100MB内存用于桥接服务器
- 约200MB内存用于Chrome实例
File Locations
文件位置
- Server:
SKILLs/web-search/dist/server/index.js - Logs:
SKILLs/web-search/.server.log - PID:
SKILLs/web-search/.server.pid - Connection cache:
SKILLs/web-search/.connection
- 服务器:
SKILLs/web-search/dist/server/index.js - 日志:
SKILLs/web-search/.server.log - PID:
SKILLs/web-search/.server.pid - 连接缓存:
SKILLs/web-search/.connection
Additional Resources
额外资源
- Full documentation:
SKILLs/web-search/README.md - Usage examples:
SKILLs/web-search/examples/basic-search.md - API reference: See README.md for complete API documentation
- Troubleshooting: See examples/basic-search.md
- 完整文档:
SKILLs/web-search/README.md - 使用示例:
SKILLs/web-search/examples/basic-search.md - API参考: 查看README.md获取完整API文档
- 故障排查: 查看examples/basic-search.md
Support
支持
For issues:
- Check for errors
.server.log - Run basic test:
node SKILLs/web-search/scripts/test-basic.js - Verify Chrome installation
- Check internet connection
- Review troubleshooting section above
Remember: This skill provides real-time access to current information. Use it whenever users need information beyond your knowledge cutoff or when accuracy of current data is important.
如遇问题:
- 查看中的错误信息
.server.log - 运行基础测试:
node SKILLs/web-search/scripts/test-basic.js - 验证Chrome安装情况
- 检查互联网连接
- 查看上述故障排查部分
注意: 此Skill提供对当前信息的实时访问。当用户需要超出你的知识截止日期的信息,或当前数据的准确性很重要时,请使用它。