What problem does it solve?
It helps you systematically diagnose hard bugs and performance regressions so you can reliably find the root cause instead of guessing or thrashing through code.
Core Features & Use Cases
- Build a tight feedback loop that turns the bug into a fast, deterministic pass/fail signal for bisection, hypothesis testing, and instrumentation.
- Reproduce and validate the exact failure so the investigation targets the same symptom the user reports.
- Generate falsifiable hypotheses and instrument deliberately using one-variable-at-a-time probes, then verify the fix with regression testing and cleanup.
Use case example: A service is throwing intermittent errors and latency has grown after a recent change; use this skill to construct a reproducible harness, rank hypotheses, instrument the system with tagged probes, implement the fix, and lock it in with a regression test.
Quick Start
Tell the AI: "diagnose this bug: it started after the last deploy, I see this error message and it happens in this scenario; first help me build a reproducible feedback loop and then guide me through hypotheses, instrumentation, and a regression test."