What problem does it solve? Hard bugs and performance regressions stall when developers jump straight to code-reading and guesswork without a reproducible signal. This Skill enforces a disciplined diagnosis loop — reproduce, minimise, hypothesise, instrument, fix, regression-test — so the root cause is found systematically instead of by luck. ## Core Features & Use Cases - Feedback-loop construction: Builds a fast, deterministic, agent-runnable red-capable command (failing test, curl script, headless browser, replay harness, bisection, fuzz loop) before any hypothesis work begins. - Structured hypothesis testing: Generates 3-5 ranked, falsifiable hypotheses, then instruments with tagged debug logs or profiling to test one variable at a time. - Regression locking and cleanup: Turns the minimised repro into a regression test at a correct seam, removes all debug instrumentation, and records the confirmed root cause in the commit message. - Use Case: A user reports that an export endpoint intermittently returns empty files. The Skill builds a loop that replays the failing request 100 times to raise the reproduction rate, minimises the input, tests ranked hypotheses with tagged logs, and lands a fix guarded by a regression test. ## Quick Start Ask the agent to diagnose the bug you are seeing, describing the exact symptom and how to trigger it, and let it build a reproduction loop first.
