What problem does it solve? Headless scrapers fail on paywalled or login-gated sites because they lack the user's session cookies and login state. This Skill captures content from those sites by driving the user's real, already-authenticated Chrome browser, so requests carry the user's cookies, login state, and local IP. ## Core Features & Use Cases - Authenticated capture via bb-browser: Drives the user's logged-in Chrome to navigate, render, and extract article text from paywalled pages. - Strict escalation gating: Runs only when headless rungs failed due to auth/paywall, when sites.yaml marks the domain as authenticated, or when the user explicitly requests a logged-in session; otherwise it redirects to the cheaper scrape-article path. - Consistent output schema: Saves markdown with the same frontmatter and directory layout as scrape-article, adding extractor, rung, and authenticated fields. - Use Case: A user needs an article from Haaretz or Calcalist+ that is subscriber-only. After confirming they are logged in to Chrome, this Skill captures the full article text and saves it to the local captures directory. ## Quick Start Capture this subscriber-only article from Haaretz using my logged-in Chrome session and save it as markdown.
