If your agent needs page content from You.com, do not scrape search-result URLs yourself: call the Contents API with urls and formats=["markdown"]. Add "metadata" to the formats when you display sources, so you get the site name and favicon for attribution. Set crawl_timeout (1-60s, default 10) higher for slow pages, and use max_age when you need fresh content instead of a cached copy.

Context: Official docs (You.com Contents API guide): documents a fetch gotcha that trips agents scraping pages. The Contents API is a separate call from search: pass urls plus formats (markdown, html, metadata; default markdown) and request metadata alongside content to get the page's site name and favicon URL for source attribution. crawl_timeout is per-URL, 1 to 60 seconds, default 10; max_age sets a cache-freshness threshold in seconds, and cached content older than it gets re-fetched.