Verify running app behavior against spec done-when criteria with captured evidence.
npx skills add https://github.com/LeroyAdonis/kitfix-2.0 --skill check-leroyadonis
Please help me install this Agent Skill. Skill: check Source: https://github.com/LeroyAdonis/kitfix-2.0/tree/main/.agents/skills/check Command: npx skills add https://github.com/LeroyAdonis/kitfix-2.0 --skill check-leroyadonis
High-intent search queries and answers about installing and using this skill.
Run the check skill, which reads the done-when criteria from current-feature.md, starts the real app, exercises each criterion through actual interaction, and reports evidence-backed pass or fail verdicts per item.
A passing build only proves code compiles. Behavioral verification runs the actual app and confirms observable outcomes like downloads, route responses, and UI flows, capturing screenshots and output as proof.
It prefers Playwright for browser driving, screenshots, and console error checks when Playwright is already installed or declared in AGENTS.md. If not installed, it uses another real-browser evidence path and reports what was used.
No. It only observes and reports; it never edits source, commits, or merges. Failed criteria are handed back to the /implement step with specific notes on what to fix.
The criterion is reported as skipped with an explanation of why it could not be verified, such as requiring a deployed environment. It is never reported as a pass without evidence.
andregusman-raizBreak software specs into phased atomic tasks with dependencies and done criteria.
Busboom-FooValidate sprint tickets, tests, and process guidelines for closure readiness.
tldrawAutomate unit and integration tests for the tldraw SDK using Vitest.
cosmicstack-labsWrite stable end-to-end tests for web apps using Playwright and Cypress.
