safe-fetch

Check if a URL is safe to fetch

Fetch a URL with prompt injection sanitization

Search the web with sanitized results

Use this agent when the user needs to research a topic using web sources. Searches, evaluates, fetches, and synthesizes — all through sanitized safe-fetch tools. <example> Context: User needs documentation for a library user: "Look up how to configure rate limiting in nginx" assistant: "I'll use the web-researcher agent to find and synthesize that information." <commentary> Multi-step research: search for docs, evaluate results, fetch best pages, synthesize answer. </commentary> </example> <example> Context: User wants to understand an API user: "Research the Stripe webhooks API and tell me how to verify signatures" assistant: "I'll use the web-researcher agent to research that." <commentary> Requires searching, finding official docs, fetching them, and extracting the specific section. </commentary> </example> <example> Context: User needs to compare approaches user: "Find best practices for Python async error handling" assistant: "I'll use the web-researcher agent to gather and compare recommendations." <commentary> Needs multiple sources fetched and compared to give a well-rounded answer. </commentary> </example> <example> Context: User asks about a specific error or issue user: "Search for solutions to the CORS preflight error with FastAPI" assistant: "I'll use the web-researcher agent to find solutions." <commentary> Research task requiring search, evaluation of Stack Overflow/docs/blogs, and synthesis. </commentary> </example>

This skill should be used when the user asks to "fetch a URL safely", "sanitize web content", "search the web securely", "check URL safety", "prevent prompt injection from web content", or discusses web fetching security. Also triggers when curl/wget is used to fetch web content.

3 hooks across 2 events