investigate

Guide structured root-cause debugging through phases for context, analysis, hypothesis, and verification.

5|Updated Apr 24, 2026

Search Tags:#workflow #debugging #learning #investigate #phases #root-cause

Ready to run

npx skills add https://github.com/timurgaleev/vibestack --skill investigate-timurgaleev

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/timurgaleev/vibestack/tree/main/skills/investigate
Command: npx skills add https://github.com/timurgaleev/vibestack --skill investigate-timurgaleev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic debugging with root cause investigation. Four phases: investigate, analyze, hypothesize, implement. Iron Law: no fixes without root cause.

Core Features & Use Cases

Guided, phase-based debugging workflow that prevents symptom-only fixes. Phase 1 gathers context and collects symptoms. Phase 2 analyzes patterns and checks prior learnings. Phase 3 forms and tests root cause hypotheses. Phase 4 implements a minimal, verifiable fix and writes regression tests. Phase 5 verifies the fix and records learnings. It also enforces safety rails like scope locks and learning capture.

Quick Start

Start a debug session by telling the agent to begin the investigate workflow and follow the guided phases.

Frequently Asked Questions about investigate

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

FAQPage Schema

What is root-cause debugging and how does it prevent symptom-only fixes?▼

Root-cause debugging systematically identifies and fixes underlying issues rather than surface symptoms. This structured workflow enforces an Iron Law: no fixes without root cause, guiding you through context gathering, pattern analysis, hypothesis testing, and verification.

How do I debug a stack trace using a structured investigation workflow?▼

To debug a stack trace, start a session and follow guided phases: gather context and collect symptoms, analyze patterns and prior learnings, form root cause hypotheses, implement a minimal verifiable fix, and verify while recording learnings.

What's the best way to analyze behavioral anomalies across my codebase?▼

The best way to analyze behavioral anomalies is using a repeatable protocol with scope control. You collect evidence, formulate hypotheses, and validate fixes, applying safety rails like scope locks to prevent unintended changes during investigation.

Can I use this debugging workflow for any error type or language?▼

Yes, this debugging workflow applies to errors, stack traces, and behavioral anomalies across codebases. It is language-agnostic and relies on gathering context, analyzing patterns, and testing hypotheses to resolve issues.

How do I write regression tests after identifying a root cause?▼

After identifying a root cause, the workflow implements a minimal, verifiable fix and writes regression tests. This occurs during the implementation phase, followed by verification and learning capture to prevent future recurrences.