investigate

Guide engineers through a four-phase debugging workflow to produce root-cause hypotheses and verification reports.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/tak-bro/claude-code --skill investigate-tak-bro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/tak-bro/claude-code/tree/main/skills/investigate
Command: npx skills add https://github.com/tak-bro/claude-code --skill investigate-tak-bro

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic debugging that prevents masking symptoms by forcing root-cause investigation before fixes.

Core Features & Use Cases

  • Structured four-phase debugging workflow: investigate, pattern analysis, hypothesis testing, and implementation.
  • Phase-driven scope locking, pattern recognition, and evidence-driven verification to improve issue resolution.
  • Use Case: When a bug spans multiple modules, follow the phases to isolate the root cause and produce a verification report.

Quick Start

Begin by collecting symptoms and reproducing the failure to initiate the root-cause investigation.

Frequently Asked Questions about investigate

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

FAQPage Schema
What is a structured root-cause debugging workflow?

Root-cause debugging is a systematic workflow that forces you to investigate underlying issues before applying fixes. It uses a four-phase process covering investigation, pattern analysis, hypothesis testing, and implementation to prevent masking symptoms.

How do I debug a bug that spans multiple software modules?

To debug multi-module bugs, follow a structured workflow to investigate symptoms, reproduce the failure, and lock scope. This process isolates the root cause through pattern recognition and evidence-driven hypothesis testing before implementation.

How do I verify a software fix without masking the original symptoms?

Verify fixes by generating a root-cause hypothesis and a verification report during the implementation phase. This evidence-driven approach ensures the root cause is tested and resolved, preventing symptom masking.

What is the best way to test a root-cause hypothesis for a software bug?

The best way to test a root-cause hypothesis is through a disciplined debugging workflow. After investigating symptoms and analyzing patterns, you test the hypothesis against evidence and produce a verification report to validate the fix.

When do I need pattern analysis for software debugging?

Pattern analysis is needed during the second phase of root-cause debugging. After collecting symptoms and reproducing the failure, you analyze patterns to lock the investigation scope and form a root-cause hypothesis.

Does root-cause debugging work without reproducing the failure?

Reproducing the failure is required to initiate the investigation phase. Root-cause debugging begins by collecting symptoms and reproducing the issue to establish a baseline for pattern analysis and hypothesis testing.