What problem does it solve? Web pages often fail to fetch due to 403/429 errors, Cloudflare interstitials, paywalls, or bot detection, leaving you without the content you need. This Skill recovers copies of those pages from third-party archive and rendering services instead of giving up or retrying the same blocked URL. ## Core Features & Use Cases - Fallback Ladder: Tries the Wayback Machine first, then archive.today domain rotation, then Jina Reader (when JINA_API_KEY is set), with an API-first pivot and real browser as last resorts. - Fake-Success Detection: Validates every recovered body using byte floors, interstitial title checks, and redirect-stub detection so dead Google Cache pages, AMP stubs, and rate-limit HTML are never mistaken for real content. - Provenance Discipline: Labels every result as a dated snapshot or live render so citations never present stale archives as current pages. - Use Case: A news article returns a paywall or Cloudflare "Just a moment..." page. Run the bundled script to pull the archived copy from Wayback or archive.today, complete with its snapshot timestamp for accurate citation. ## Quick Start Ask the agent to recover the blocked article at a given URL using the recover_page.py script and cite the result with its archive timestamp.
