diagnose

Diagnose hard bugs through a reproducible debugging loop with structured phases.

16|Updated Jun 6, 2026
One-click install
npx skills add https://github.com/Insik-Han/han-monorepo-template --skill diagnose-insik-han
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/Insik-Han/han-monorepo-template/tree/main/.agents/skills/diagnose
Command: npx skills add https://github.com/Insik-Han/han-monorepo-template --skill diagnose-insik-han

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user reports a bug, describes something broken, or mentions a performance regression.

Core Features & Use Cases

  • Build a fast, repeatable reproduction loop that surfaces the cause quickly.
  • Follow a phase-driven workflow: reproduce, hypothesise, instrument, implement a fix, and regression-test.
  • Applicable across unit/integration/e2e tests and performance scenarios to prevent regressions.

Quick Start

Run the diagnose loop on a failing scenario to reproduce, hypothesize, instrument, and validate the fix.

Frequently Asked Questions about diagnose

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is a disciplined debugging loop for hard bugs and performance regressions?

A disciplined debugging loop systematically resolves hard bugs by progressing through reproduction, minimization, hypothesis generation, instrumentation, fixing, and regression testing to pinpoint root causes.

How do I reproduce and diagnose a flaky bug across codebases?

Reproduce and diagnose flaky bugs by constructing a repeatable reproduction loop that exposes the failure signal, then generate hypotheses, instrument the code, apply a fix, and execute regression tests to confirm resolution.

Does this bug triage workflow work for performance regressions and e2e tests?

Yes, bug triage workflows applying a disciplined debug loop function across unit, integration, and e2e tests, as well as performance regression scenarios, by enforcing reproducible signals, instrumentation, and regression testing.

What's the best way to structure bug triage when a user reports something broken?

The best way to structure bug triage involves a phase-driven workflow: reproduce the failing scenario, minimize the failure, document hypotheses, instrument for signals, implement the fix, and execute regression tests to prevent recurrence.

When should I not use a hypothesis-driven debug loop for bug diagnosis?

Avoid a hypothesis-driven debug loop when reproducible signals are unattainable or the failing scenario cannot be minimized, as the workflow mandates structured phase transitions from reproduction to regression testing.