Firecrawl Search

**Judgement:** A capable and focused skill for web search, scraping, and crawling via the Firecrawl API. Well-suited when you need fresh web data, including JS-heavy pages and multi-page site crawls. **Strengths & advantages** - Covers three major needs in one tool: search, single-page scrape, and full-site crawl. - Handles JS-heavy sites that simple HTTP fetchers often fail on. - Supports structured data extraction, enabling more direct downstream processing. - Simple, documented commands (`firecrawl_search`, `firecrawl_scrape`, `firecrawl_crawl`) and helper scripts. **Key risks & limitations** - **External dependency:** Requires a valid `FIRECRAWL_API_KEY`; availability, latency, and rate limits depend on the Firecrawl service. - **Legal/ToS considerations:** Web scraping and crawling must respect robots.txt, site terms of service, and jurisdictional regulations; misuse can create compliance risk. - **Data privacy:** Queries, URLs, and scraped content are sent to an external API; avoid sensitive or confidential targets unless you have appropriate agreements. - **Content quality & trust:** Results may include outdated, biased, or incorrect web content and should not be treated as authoritative without verification. **Recommended scenarios** - Need current information from the web that is not in local context or training data. - Need to ingest or index an entire documentation site or blog for later search and reasoning. - Need to scrape JS-rendered pages (e.g., modern SPAs) that do not render correctly via basic HTTP requests. - Need structured data extraction from repetitive web page templates for analytics or automation. **Less suitable when** - You must avoid sending data to third-party services for privacy or compliance reasons. - You only need simple HTTP fetches to a few known, static pages (a lighter-weight fetch skill may suffice). - Strict adherence to complex site-specific ToS or crawl policies is required and cannot be audited or configured at the Firecrawl level.