Diagnoses hard bugs and performance regressions through a six-phase feedback-loop workflow.
npx skills add https://github.com/Yassimba/loom --skill diagnosing-bugs-yassimba
Please help me install this Agent Skill. Skill: diagnosing-bugs Source: https://github.com/Yassimba/loom/tree/main/skills/diagnosing-bugs Command: npx skills add https://github.com/Yassimba/loom --skill diagnosing-bugs-yassimba
High-intent search queries and answers about installing and using this skill.
Intermittent bugs require raising the reproduction rate rather than finding a clean repro. Loop the trigger 100 times, parallelize execution, add stress, narrow timing windows, or inject sleeps until the failure rate is high enough to debug against.
Performance regressions need measurement before fixing: establish a baseline with a timing harness, performance.now(), a profiler, or a query plan, then bisect between known states. Logs are usually the wrong tool for perf diagnosis.
Stop and say so explicitly instead of hypothesizing without a loop. Ask the user for environment access, a redacted captured artifact such as a HAR file or log dump, or permission to add temporary production instrumentation.
Writing the regression test first proves the test actually catches the bug by watching it fail before the fix and pass after. If no correct seam exists for the test, that absence itself signals an architecture problem worth flagging.
Yes, as a last resort it uses a human-in-the-loop bash template that prompts the user through manual steps and captures their observations as structured output. The agent then parses the captured values to continue the diagnosis loop.
jonkikyTrace reported issue symptoms to code paths and verify root cause hypotheses.
gittowerResolve small local maintenance tasks with test-first implementation and Codex review.
leocostaReproduce bugs deterministically and guide teams through structured four-phase triage workflows.
handsomelong922Facilitate collaborative brainstorming to convert ideas into validated design plans.
