debug-triage

Diagnose failures by reproducing issues and capturing decisive evidence.

Updated Feb 1, 2026

Search Tags:#verification #debug #evidence #reproduce #triage #root-cause #fault-isolation

Ready to run

npx skills add https://github.com/habrahgithub/node-backend-starter --skill debug-triage-habrahgithub

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: debug-triage
Source: https://github.com/habrahgithub/node-backend-starter/tree/main/.codex/skills/debug-triage
Command: npx skills add https://github.com/habrahgithub/node-backend-starter --skill debug-triage-habrahgithub

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diagnose failures with minimal context and reproducible evidence, helping teams triage issues across tests, services, builds, and deployments.

Core Features & Use Cases

Reproduce the issue with the smallest command or request path.
Capture decisive evidence: error type, failing step, and first relevant stack line.
Isolate likely fault domain: config, data, dependency, runtime, or code regression.
Propose the smallest safe fix and verify its effect on the problem.

Quick Start

Describe the failure scenario succinctly and request a targeted triage analysis.

Frequently Asked Questions about debug-triage

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

FAQPage Schema

How do I triage software failures when I have minimal context?▼

Failure triage with minimal context starts by reproducing the issue with the smallest command, capturing decisive evidence like the error type and first stack line, then isolating the fault domain to propose a safe fix.

What is the best way to reproduce an issue for effective fault isolation?▼

Reproduce the issue using the smallest command or request path possible to trigger the failure reliably, which helps isolate the fault domain accurately and capture decisive evidence for root-cause analysis.

How do I isolate the fault domain of a failing test or service deployment?▼

Isolate the fault domain by analyzing captured evidence such as error types and failing steps to determine if the root cause lies in configuration, data, dependencies, runtime environments, or code regressions.

Can I get safe remediation suggestions for build or deployment failures?▼

Yes, you can receive safe remediation suggestions that propose the smallest possible fix for the failing component and verify its effect across tests, services, and deployments to ensure stability.

What evidence should I capture to diagnose a production issue quickly?▼

Capture decisive evidence including the specific error type, the exact failing step in the workflow, and the first relevant stack trace line to diagnose production issues quickly and accurately.

Why does my triage workflow fail to find the root cause of a regression?▼

Triage workflows often fail to find a root cause when they lack reproducible evidence or skip fault domain isolation, making it necessary to reproduce the issue minimally and capture the first relevant stack line.