Fallback web fetcher using headless browser with stealth mode. Use ONLY after web_fetch fails with 403, captcha, access denied, empty content, or "enable JavaScript" errors. Never use preemptively — always try web_fetch first.
Fallback tool when web_fetch fails due to bot detection or JavaScript requirements.
1. User requests URL content
2. Try web_fetch first (ALWAYS)
3. If web_fetch succeeds → done
4. If web_fetch fails with blocking indicators → use puppeteer-fetch:fetch_page
Use puppeteer-fetch when web_fetch returns:
puppeteer-fetch:fetch_page
├── url (required): Full URL to fetch
├── timeout (optional): ms, default 30000
└── waitFor (optional): CSS selector to wait for before extracting
Returns markdown (not HTML).