investigate

Guide structured investigation of software issues to diagnose root causes.

1|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/gebl/anvil-skill-marketplace --skill investigate-gebl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/gebl/anvil-skill-marketplace/tree/main/generated/claude/skills/investigate
Command: npx skills add https://github.com/gebl/anvil-skill-marketplace --skill investigate-gebl

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

  • Structured investigation workflow: collect symptoms, read code, and reproduce steps to isolate root causes.
  • Evidence-based hypotheses: validate potential causes before implementing fixes.
  • Proactive guidance: when users report errors or abnormal behavior, recommends a disciplined debugging approach.

Quick Start

Start an investigation by collecting symptoms, then trace the code paths to form a root-cause hypothesis.

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 structured investigation work?

Root-cause debugging is a systematic workflow that collects symptoms, traces code paths, and validates evidence-based hypotheses to isolate the true source of software issues rather than applying superficial fixes.

How do I debug a flaky test or regression when the symptom is hard to reproduce?

Start by collecting all observable symptoms and stack traces, then trace the code paths involved to form a root-cause hypothesis. Reproduce the steps systematically to validate potential causes before implementing any fix.

What is the best way to analyze a stack trace to find the actual cause of an error?

The best way to analyze a stack trace is following a disciplined investigation workflow that enforces collecting evidence, reading code, and testing hypotheses against the trace data to ensure you address the real cause.

Can I use a structured debugging workflow for complex regressions and abnormal behavior?

Yes, a structured debugging workflow applies to errors, stack traces, flaky behavior, and regressions. It guides you through symptom collection, code tracing, and evidence-based hypothesis testing to resolve complex regressions.

Why should I not apply fixes without finding the root cause of an issue?

Applying fixes without finding the root cause often leads to recurring issues. The Iron Law of structured debugging enforces no fixes without root cause, ensuring that corrections address the actual underlying problem.

What are the phases of a systematic troubleshooting workflow for software issues?

The phases of a systematic troubleshooting workflow are investigate, analyze, hypothesize, and implement. This enforces a disciplined approach with defined checks and decision points to ensure fixes resolve the real cause.