What problem does it solve? Hard bugs and flaky performance regressions resist ad-hoc debugging: staring at code rarely finds the cause. This Skill enforces a disciplined six-phase diagnosis loop that builds a tight, deterministic pass/fail signal first, then reproduces, minimises, hypothesises, instruments, fixes, and cleans up. ## Core Features & Use Cases - Feedback-loop construction: Ten ranked strategies for building a red-capable reproduction command, from failing tests and curl scripts to bisection harnesses, differential loops, and human-in-the-loop bash scripts. - Structured hypothesis testing: Generates 3-5 ranked, falsifiable hypotheses before testing, then instruments with tagged debug logs or profiler baselines, changing one variable at a time. - Regression lockdown and cleanup: Writes the regression test before the fix at a correct seam, then verifies the original repro, removes all tagged instrumentation, and records the confirmed root cause. - Use Case: A user reports that an export endpoint intermittently returns stale data. The Skill guides building a deterministic failing test, minimising the repro, ranking hypotheses, instrumenting the suspected cache layer, and landing a fix with a regression test. ## Quick Start Ask the agent to diagnose the bug where the export endpoint intermittently returns stale data and have it build a failing reproduction loop first.
