troubleshoot

Diagnose and resolve active environmental, runtime, process, or tool failures.

8|Updated Feb 2, 2026
One-click install
npx skills add https://github.com/usurobor/cnos --skill troubleshoot-usurobor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: troubleshoot
Source: https://github.com/usurobor/cnos/tree/main/src/packages/cnos.eng/skills/eng/troubleshoot
Command: npx skills add https://github.com/usurobor/cnos --skill troubleshoot-usurobor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diagnose and resolve active environmental, runtime, process, or tool failures while work is blocked.

Core Features & Use Cases

  • Teaches a disciplined, hypothesis-driven approach to isolating root causes by preserving evidence, testing the cheapest hypotheses first, and recording results.
  • Provides a structured triage order (process state, kernel logs, tool outputs, resources, lifecycle, then application behavior) to minimize guesswork and risk.
  • Supports RCA handoff when escalation is necessary and ensures evidence is preserved for future analysis.

Quick Start

Describe the symptom and preserve evidence, then test the cheapest discriminating hypotheses in triage order.

Frequently Asked Questions about troubleshoot

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

FAQPage Schema
How do I isolate root causes during a live debugging session?

To isolate root causes during a live debugging session, apply a hypothesis-driven approach by preserving evidence, testing the cheapest discriminating hypotheses first, and recording results to minimize guesswork.

What is the best triage order for troubleshooting runtime failures?

The best triage order for troubleshooting runtime failures follows a structured sequence: process state, kernel logs, tool outputs, resources, lifecycle, and then application behavior.

How do I preserve evidence for incident management before escalating an outage?

Preserve evidence for incident management by capturing process states, kernel logs, and tool outputs before testing hypotheses, ensuring data remains available for RCA handoff during escalation.

Can I use hypothesis-driven troubleshooting for multi-component tool errors?

Yes, you can use hypothesis-driven troubleshooting for multi-component tool errors. It applies to active environmental, runtime, and process failures across single-task flows and multi-component work.

Why should I test the cheapest hypotheses first when diagnosing tool failures?

Testing the cheapest hypotheses first when diagnosing tool failures minimizes risk and guesswork, allowing you to efficiently isolate root causes without unnecessary environmental changes.