investigate

Guide structured four-phase debugging to identify root causes of software issues.

Updated Feb 27, 2026
One-click install
npx skills add https://github.com/zhifengzhang-sz/divine-book --skill investigate-zhifengzhang-sz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/zhifengzhang-sz/divine-book/tree/main/.claude/skills/gstack/investigate
Command: npx skills add https://github.com/zhifengzhang-sz/divine-book --skill investigate-zhifengzhang-sz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic, repeatable debugging that finds root causes before applying fixes, reducing firefighting and regressions.

Core Features & Use Cases

  • Four-phase debugging: investigate, analyze, hypothesize, implement.
  • Tooling orchestration: uses Bash, Read, Write, Edit, Grep, Glob, AskUserQuestion, WebSearch to gather symptoms, reproduce, and verify fixes.
  • Works across errors, crashes, performance issues, and unexpected behavior in local and remote environments.

Quick Start

Describe symptoms, collect logs, reproduce the bug, and trace to a root cause using the gstack workflow.

Frequently Asked Questions about investigate

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

FAQPage Schema
What is structured root-cause debugging for software?

Structured root-cause debugging is a repeatable investigation workflow that enforces collecting symptoms, reproducing the bug, hypothesizing, and implementing fixes to find the actual source of errors before applying changes.

How do I debug a software crash using a structured investigation workflow?

Debug a software crash by following a four-phase workflow: investigate symptoms, analyze logs and reproduce the bug, hypothesize the root cause, and implement fixes using Bash, Grep, Read, and Edit tools.

Can I use this structured debugging workflow for performance slowdowns and unexpected behavior?

Yes, this structured debugging workflow applies to errors, crashes, performance slowdowns, and unexpected behavior across local and remote environments by orchestrating Bash, Grep, Glob, Read, and WebSearch tools.

What's the best way to reproduce a bug before applying a software fix?

The best way to reproduce a bug is using a structured investigation approach that collects symptoms and logs first, then uses Bash and Grep to recreate the exact conditions, ensuring you find the root cause before fixing.

Does this root-cause debugging workflow require any specific dependencies or components?

No external dependencies or components are required. This root-cause debugging workflow orchestrates standard tools including Bash, Read, Write, Edit, Grep, Glob, AskUserQuestion, and WebSearch to perform the investigation.

When should I avoid a structured root-cause investigation workflow?

Avoid a structured root-cause investigation workflow when dealing with trivial syntax errors or obvious one-line fixes, as the four-phase process of collecting symptoms and hypothesizing adds overhead better suited for complex regressions.