What problem does it solve? After deploying to production, merging a risky PR, or upgrading dependencies, teams often lack immediate visibility into whether the release introduced regressions. This Skill continuously checks a deployed URL for broken pages, new console errors, failed API calls, and performance degradation so issues are caught before users report them. ## Core Features & Use Cases - Multi-Signal Health Checks: Watches HTTP status, console errors, network failures, Core Web Vitals (LCP/CLS/INP), key content elements, and API endpoint health against a baseline. - Flexible Watch Modes: Run a single quick check, a sustained watch at a set interval for a defined duration, or a diff mode comparing staging versus production. - Threshold-Based Alerting: Classifies findings as critical, warning, or info, and sends desktop notifications or Slack/Discord webhook alerts when critical thresholds are crossed. - Use Case: After pushing a dependency upgrade to production, run a sustained two-hour watch on your app URL to confirm LCP stays within 500ms of baseline and no new console errors appear. ## Quick Start Ask the AI to run canary-watch against your deployed URL, for example to monitor https://myapp.com every 5 minutes for 2 hours and alert on any critical regressions.
