What is the core principle of this debugging methodology?

The core principle is 'Measure, don't speculate.' It mandates adding instrumentation to produce specific data explaining a failure before making any code changes.

How does this skill handle complex asynchronous flows?

It utilizes OpenTelemetry spans and tracing patterns to track execution across different functions, making invisible bottlenecks and errors visible.

Can I use this for TypeScript projects using Result types?

Yes, the skill includes specific protocols for debugging Result-returning functions, including explicit logging for error paths and workflow step tracking.

Why does it suggest making only one change at a time?

Changing multiple variables at once makes it impossible to identify which specific fix solved the problem, leading to brittle code and unconfirmed root causes.

Does this skill work with unit testing?

Absolutely. It includes patterns for debugging failing tests systematically by logging mock interactions and internal state before assertions.

Debugging Methodology

Name: Debugging Methodology
Author: jagreehal

byjagreehal

0•

Analytics & Monitoring

Implements an evidence-based debugging protocol that prioritizes observability and systematic instrumentation over speculative fixes.

This skill provides a rigorous framework for troubleshooting software issues by enforcing a 'measure, don't speculate' philosophy. It guides Claude to instrument code with structured logging and OpenTelemetry spans at critical decision points before attempting any fixes, ensuring that every hypothesis is backed by hard data. By following a standardized protocol—from exact symptom documentation to one-change-at-a-time testing—it eliminates the common anti-pattern of random trial-and-error, making failures visible and resolutions repeatable in complex environments.

Key Features

01Systematic instrumentation patterns for TypeScript and Result types

02OpenTelemetry span integration for tracing complex async flows

03Standardized symptom documentation and context capture

04Evidence-based hypothesis formation framework

05Structured decision tree for troubleshooting navigation

060 GitHub stars

Use Cases

01Diagnosing intermittent race conditions or silent failures in production

02Troubleshooting complex asynchronous pipelines and multi-step workflows

03Debugging database connection leaks and external API integration failures

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jagreehal/jagreehal-claude-skills debugging-methodology

For use in Claude.ai and ChatGPT

Download Skill