diagnose

Diagnose software bugs and performance regressions through structured reproduction and validation workflows.

Updated Apr 27, 2026
One-click install
npx skills add https://github.com/cchrisv/newtonSelector --skill diagnose-cchrisv
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: diagnose
Source: https://github.com/cchrisv/newtonSelector/tree/main/.agents/skills/diagnose
Command: npx skills add https://github.com/cchrisv/newtonSelector --skill diagnose-cchrisv

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill provides a disciplined process for investigating difficult bugs and performance regressions when symptoms are unclear, preventing guesswork and unverified fixes.

Core Features & Use Cases

  • Reproduction Loops: Builds reliable feedback loops through tests, scripts, traces, harnesses, or other deterministic signals before debugging.
  • Hypothesis-Driven Debugging: Guides ranked hypotheses, targeted instrumentation, fixes, and regression validation.
  • Use Case: When an application fails intermittently or becomes slower after a change, use this Skill to reproduce the issue, isolate the cause, and verify the solution.

Quick Start

Use the diagnose skill to investigate the reported bug, create a reproduction loop, identify likely causes, and validate the fix with regression testing.

Frequently Asked Questions about diagnose

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

FAQPage Schema
How do I systematically debug intermittent software failures?

Intermittent software failures require systematic debugging through structured reproduction loops, hypothesis-driven instrumentation, and regression validation to prevent unverified fixes. You build deterministic feedback loops using tests, scripts, and traces to reliably reproduce the issue before isolating the cause.

What's the best way to diagnose a performance regression after a code change?

Diagnosing a performance regression requires establishing deterministic reproduction loops to measure degraded application behavior, then applying hypothesis-driven analysis and targeted instrumentation to isolate the specific cause. You validate the performance fix through regression testing to ensure reliable resolution.

Why should I use hypothesis-driven debugging instead of guesswork?

Hypothesis-driven debugging replaces guesswork by ranking likely causes, applying targeted instrumentation, and validating fixes through regression tests. This structured workflow ensures difficult software bugs are resolved reliably through deterministic feedback loops rather than unverified assumptions.

How do I create a reliable reproduction loop for hard bugs?

A reliable reproduction loop for hard bugs is built using tests, scripts, traces, or harnesses that provide deterministic signals. This structured feedback loop must be established before debugging begins to ensure consistent validation of hypotheses and targeted instrumentation.

Can I use this debugging workflow for both broken behavior and performance issues?

Yes, this debugging workflow applies to both broken application behavior and degraded performance issues. It uses structured reproduction, analysis, and validation workflows to systematically resolve difficult software bugs and performance regressions through deterministic feedback loops.

When should I not use a structured bug diagnosis workflow?

A structured bug diagnosis workflow is unnecessary for trivial bugs with obvious causes, as the overhead of building reproduction loops and ranked hypotheses outweighs the benefit. It is designed specifically for difficult bugs and performance regressions with unclear symptoms requiring disciplined investigation.