Automates the QA grind after you ship to staging: drives a real browser through smoke tests, interaction flows, visual regression checks, and accessibility audits. It'll catch console errors, dead links, broken forms, layout shifts across three breakpoints, and WCAG violations before they hit production. The safety model is smart: defaults to read-only journeys and won't let you accidentally run a checkout flow against prod without explicit opt-in. Outputs a structured verdict (SHIP / SHIP WITH FIXES / DO NOT SHIP) with screenshots and Core Web Vitals. Works with any browser MCP. Honestly, this is the kind of thing you write yourself in bits and pieces over six months, so having it packaged saves you from maintaining yet another Playwright script.
npx -y skills add affaan-m/ecc --skill browser-qa --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
wshobson/agents
github/awesome-copilot