What problem does it solve? Web scrapers fail in three predictable ways: they get blocked by anti-bot systems, break silently when sites change their markup, and produce dirty data that pollutes downstream pipelines. This Skill audits scraper code against all three failure modes before they cause production incidents. ## Core Features & Use Cases - Reliability Audit: Checks fetch-layer concerns like timeouts, exponential backoff, 429/Retry-After handling, and session reuse, plus parse-layer risks like fragile generated CSS selectors versus stable data-* anchors. - Data Quality Contracts: Enforces validated extraction schemas (e.g., Pydantic models), quarantine of parse failures with raw HTML snapshots, and canary checks that fail loudly when expected element counts drop to zero. - Blocking & Politeness Review: Evaluates rate limiting, robots.txt compliance, proxy rotation escalation, and whether an official API or JSON endpoint should replace scraping entirely. - Use Case: A scraper that worked for months suddenly returns empty product lists. Run this review to detect the blocked-but-200 response, identify the changed selectors, and add validation so future breakage alerts instead of silently writing nulls. ## Quick Start Review my Playwright scraper for the product listing pages and tell me where it will get blocked or break silently.
