investigate

Trace execution chains and collect evidence to diagnose software bugs.

Updated Apr 24, 2026
One-click install
npx skills add https://github.com/mulev/solo-dev --skill investigate-mulev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/mulev/solo-dev/tree/main/skills/investigate
Command: npx skills add https://github.com/mulev/solo-dev --skill investigate-mulev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic root cause analysis and fix design for software bugs and unexpected behavior. Evidence-only methodology with no assumptions — traces full execution chains, consults real documentation, presents findings with proof before proposing fixes. Two mandatory user gates: root cause confirmation and fix approval. Hands off to the configured planning skill for implementation.

Core Features & Use Cases

  • Evidence-based debugging: trace the full chain from entry to failure with citations.
  • Hard gates: require user confirmation before any plan or code changes.
  • Structured workflow: collect evidence, present root cause, then plan and execute.

Quick Start

Immediately invoke the investigate skill when a bug or unexpected behavior is reported, then await gates and requested evidence.

Frequently Asked Questions about investigate

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

FAQPage Schema
How do I perform root-cause analysis for a software bug without making assumptions?

Root-cause analysis without assumptions requires tracing the full execution chain from entry to failure, consulting real documentation, and presenting evidence-based findings with citations before proposing any fixes. This approach enforces a strict evidence-only methodology to prevent premature conclusions.

What is the best way to trace an execution chain to find the root cause of unexpected software behavior?

Tracing an execution chain to find a root cause involves systematically collecting precise evidence at each step from initial entry to the point of failure. The process requires reading actual documentation and verifying the complete trace before identifying the underlying fault.

How do I enforce approval gates before applying a fix during debugging?

Enforcing approval gates during debugging requires a structured workflow that pauses for user confirmation at two critical stages: after presenting the root cause and before proposing any code changes. This ensures no modifications occur without explicit sign-off.

Can I use an evidence-based debugging workflow for bugs of any size?

Evidence-based debugging workflows apply to bugs of all sizes, from minor unexpected behaviors to major system failures. The methodology scales by requiring full execution-chain tracing and precise evidence collection regardless of the issue's complexity.

When do I need a structured investigation workflow with hard gates for software debugging?

A structured investigation workflow with hard gates is needed when you must guarantee that root-cause findings are fully confirmed with proof before any planning or code changes begin. It prevents unauthorized fixes by requiring explicit user sign-off.

What happens after root cause confirmation in a gated debugging process?

After root cause confirmation in a gated debugging process, the workflow hands off to the configured planning skill to design and execute the implementation. The fix only proceeds once the second approval gate is explicitly passed.