diagnose

Guide systematic debug loops to diagnose bugs and performance regressions.

1|Updated Jan 17, 2026

Search Tags:#root cause analysis #bug diagnosis #performance regression #debugging loop #hypothesis formation

Ready to run

npx skills add https://github.com/alexandersumer/config --skill diagnose-alexandersumer

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/alexandersumer/config/tree/main/.agents/skills/diagnose
Command: npx skills add https://github.com/alexandersumer/config --skill diagnose-alexandersumer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps diagnose difficult bugs and performance regressions, guiding you through a disciplined debug loop to identify and fix the root cause efficiently.

Core Features & Use Cases

Disciplined Debug Loop: Offers structured steps to diagnose bugs, from hypothesis creation to verification.
Proof Policy: Encourages reuse of proven methods while introducing new checks when necessary.
Performance Regression Analysis: Provides tools to measure and bisect performance issues.
Use Case: When faced with an elusive bug or a performance regression, use this Skill to create a minimal reproducible example, form hypotheses, and systematically eliminate potential causes.

Quick Start

Use the diagnose skill to create a proof policy for your latest bug and follow the structured process to identify the root cause.

Frequently Asked Questions about diagnose

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

FAQPage Schema

How do I diagnose a complex bug and find the root cause?▼

To diagnose a complex bug and find the root cause, follow a disciplined debug loop that involves forming a hypothesis, creating a minimal reproducible example, and verifying potential causes systematically to eliminate variables.

What is the best way to troubleshoot a performance regression?▼

The best way to troubleshoot a performance regression is to use a structured approach that measures and bisects the issue, applying a proof policy to reuse proven methods while introducing new checks only when necessary.

How do I create a minimal reproducible example for an elusive bug?▼

Creating a minimal reproducible example for an elusive bug requires narrowing down the scope through hypothesis formation, systematically testing conditions, and broadening your search only when prior assumptions are proven incorrect.

Can I use a structured debug loop for existing codebases?▼

Yes, you can use a structured debug loop for existing codebases, as the approach applies to both new and legacy code by establishing a proof policy that systematically verifies hypotheses without relying on guesswork.

What is a proof policy in bug diagnosis?▼

A proof policy in bug diagnosis is a strategy that encourages reusing proven verification methods to validate hypotheses, introducing entirely new checks only when the existing proof methods fail to isolate the root cause.

When should I broaden my search during a debugging loop?▼

You should broaden your search during a debugging loop only when necessary, specifically after your current proof policy has exhausted its verification methods and failed to confirm the initial hypothesis for the complex bug.