Web Browsing

Browse and summarize websites, extract content from URLs, search the web for information. Use when user asks to visit a website, get webpage content, search...

Security Scan

The skill's code and instructions are internally consistent with a web-browsing/summarization purpose and do not request unrelated credentials or install arbitrary remote code.

Purpose & Capability

Name/description match the provided files: SKILL.md describes browsing/searching/extraction and scripts/search_web.py implements web search (DuckDuckGo HTML) and URL fetching/HTML parsing. Required credentials or system access are not requested and would not be expected for this functionality.

Instruction Scope

SKILL.md tools list includes web_search, fetch_url, and extract_content. The packaged script provides search_web and fetch_url; there is no separate extract_content function but HTML parsing via BeautifulSoup is demonstrated in usage docs. SKILL.md and docs mention respecting robots.txt and rate limits, but the script does not check robots.txt or implement rate-limiting — those are left to the caller. The skill fetches arbitrary URLs (expected for browsing) so users should avoid sending private-login pages or secrets.

Install Mechanism

No install spec or remote downloads are present (instruction-only with a small local Python script). This is lower-risk than arbitrary installs. Note: the script depends on third-party Python packages (requests, bs4) but those are not installed by the skill itself — that's an operational requirement rather than a hidden installer.

Credentials

The skill declares no environment variables, no credentials, and does not read sensitive files. It only makes outbound HTTP(S) requests to user-supplied URLs and DuckDuckGo; those network calls are consistent with its stated purpose.

Persistence & Privilege

always is false and the skill does not request persistent system changes or modify other skills. It does not store tokens or alter agent-wide config.

Assessment

This skill appears to do what it says: search and fetch web pages. Before installing or using it, confirm your environment provides the Python packages 'requests' and 'beautifulsoup4'. Be cautious when supplying URLs that require authentication or contain private data — the skill will fetch and return page text. The implementation does not enforce robots.txt or rate-limiting, so avoid aggressive automated crawling; if you plan high-volume use, add delays and robots.txt checks. If you need extraction helpers named explicitly (extract_content) or stricter crawling rules, ask the developer to add those functions or to declare dependencies formally.

Web Browsing Skill

This skill enables browsing websites, extracting content from URLs, and searching the web for information.

When to Use

Use this skill when:

• User asks you to visit a website (e.g., "Check out example.com")
• User wants webpage content summarized
• User needs current information that requires internet search
• User provides a URL and wants it analyzed
• User asks to search for something online

How It Works

1. Direct URL Access

When given a specific URL:

User: "What's on https://example.com?"
→ Visit the page, extract main content, summarize key points

2. Web Search

When user asks to search:

User: "Find information about climate change"
→ Perform web search, present top results with summaries

3. Content Extraction

For specific data extraction:

User: "Get the latest news from techcrunch.com"
→ Navigate to site, extract relevant articles/headlines

Tools Available

• web_search: Search the web for information
• fetch_url: Visit and retrieve webpage content
• extract_content: Parse HTML and extract structured data

Best Practices

1. Be specific - Tell me what you want from the page (summary, specific data, latest news)
2. Provide URLs - If you have a specific page, share the URL directly
3. Clarify intent - Let me know if you need:
   • Quick summary
   • Detailed analysis
   • Specific data points
   • Latest updates

Examples

✅ "Visit https://news.ycombinator.com and summarize today's top stories"
✅ "Search for the latest React.js tutorial"
✅ "Check what's on Wikipedia's page about quantum computing"
✅ "Find pricing information from apple.com/iphone"
❌ Just say "browse the web" - be more specific!

Limitations

• Cannot interact with JavaScript-heavy sites (may miss dynamic content)
• Some sites block automated access
• Video/audio content cannot be played, only described if available
• Login-required pages won't work without credentials

---

Ready to browse! Just give me a URL or tell me what to search for. 🌐