What problem does it solve? Hard bugs and flaky performance regressions resist casual code reading and single-hypothesis guessing. This Skill enforces a disciplined diagnosis loop that builds a tight, red-capable reproduction signal before any hypothesis is tested, preventing wasted hours of unfounded theorizing. ## Core Features & Use Cases - Feedback Loop Construction: Ten ranked strategies for building a pass/fail signal, from failing tests and curl scripts to property-based fuzzing, bisection harnesses, and human-in-the-loop bash scripts. - Structured Six-Phase Process: Build a loop, reproduce and minimize, generate 3-5 falsifiable ranked hypotheses, instrument with tagged debug logs, fix with a regression test, then clean up and post-mortem. - Non-Deterministic Bug Handling: Techniques to raise reproduction rates for flaky bugs through looping, parallelization, stress, and timing control. - Use Case: A user reports an intermittent export failure in production. The Skill guides the agent to build a replay harness from a captured request, minimize the failing scenario, rank hypotheses, instrument with tagged logs, and land a fix with a regression test. ## Quick Start Ask the agent to diagnose the bug where the export button throws an error, and have it follow the diagnosis loop starting with a reproducible failing test.
