Agent skill · shipshitdev
qa-loop
Runs an interactive localhost QA session that accepts issues, screenshots, routes, and browser evidence; watches app, API, browser, and network errors; reproduces and fixes each defect; verifies the result; and creates one local commit per fix. Use when asked to start QA, QA a local app, fix localhost issues as they arrive, or work through screenshot feedback without leaving the current checkout.
What it needs
About 10k tokens when loaded.
What this skill does
Localhost QA Loop Run a persistent, user-directed QA session in the existing checkout. Keep the project's local environment available, reproduce each incoming issue against the running app, make the smallest correct fix, verify it, and commit it before taking the next issue. Contract Inputs: A Git repository with a locally runnable application Issues supplied incrementally as text, screenshots, recordings, routes, console output, network evidence, or expected-versus-actual behavior Existing local environment files and credentials required by the application Outputs: Running local app URLs and readiness state For each issue: reproduction evidence, root cause, fix, verification evidence, and local commit hash A continuously updated queue of user-reported and intercepted issues in pending, in progress, fixed, ignored, observed, or blocked states Creates/Modifies: Creates and pins one qa/YYYY-MM-DD branch in the current checkout Modifies only files required for the active issue Creates one conventional local commit after each verified fix May create ignored local logs, traces, or screenshots for diagnosis Persists redacted monitor checkpoints and queue records in the repository designated ignored scratch directory. Record the exact session-specific path before monitoring; never commit checkpoints or evidence. If no permitted persistent location exists, report restart coverage unavailable. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills shipshitdev/qa-loop--f9f755