diagnose

Diagnose software bugs and performance regressions through a structured 6-phase loop.

1|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/Herxinsasa/Skills-Collector --skill diagnose-herxinsasa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/Herxinsasa/Skills-Collector/tree/main/mattpocock-skills/skills/engineering/diagnose
Command: npx skills add https://github.com/Herxinsasa/Skills-Collector --skill diagnose-herxinsasa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Unstructured, ad-hoc debugging wastes hours on hard bugs, flaky issues, and performance regressions where random code inspection fails to identify root causes, leading to repeated fixes and unresolved technical debt.

Core Features & Use Cases

  • 6-Phase Disciplined Loop: Enforces a structured workflow from building a deterministic feedback loop through cleanup and post-mortem to avoid skipped critical steps.
  • Edge Case Support: Includes guidance for non-deterministic bug reproduction, scenarios with no available test seams, and human-in-the-loop debugging for UI-triggered issues.
  • Use Cases: Debugging production crashes, memory leaks, performance slowdowns, flaky test failures, and any complex bug requiring systematic root cause analysis.

Quick Start

Use the diagnose skill to execute the full structured 6-phase debugging workflow for the reported performance regression in the user authentication service.

Frequently Asked Questions about diagnose

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

FAQPage Schema
How do I find the root cause of a hard bug or performance regression?

Root cause analysis for hard bugs requires a disciplined 6-phase loop: building deterministic feedback, generating falsifiable hypotheses, targeted instrumentation, and regression test validation to ensure durable fixes.

What is the best way to debug non-deterministic flaky test failures?

Debugging flaky test failures requires building a deterministic pass/fail feedback loop and applying hypothesis testing. Edge case support handles non-deterministic reproduction and scenarios lacking available test seams.

How do I systematically debug memory leaks and production crashes?

Systematically debug memory leaks and production crashes by following a structured 6-phase workflow enforcing critical steps from building deterministic feedback loops through post-mortem analysis to prevent unresolved technical debt.

Does hypothesis testing work for UI-triggered human-in-the-loop debugging?

Yes, hypothesis testing supports human-in-the-loop debugging for UI-triggered issues. The workflow includes specific edge case guidance for scenarios requiring manual interaction to reproduce failures and validate fixes.

When should I use a structured debugging workflow instead of ad-hoc code inspection?

Use a structured debugging workflow when unstructured ad-hoc code inspection fails on hard bugs, flaky issues, and performance regressions, preventing repeated fixes and unresolved technical debt through disciplined regression testing.