investigate

Enforce root-cause investigation before applying software fixes.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/lyh2/gstack-deepseek --skill investigate-lyh2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/lyh2/gstack-deepseek/tree/main/investigate
Command: npx skills add https://github.com/lyh2/gstack-deepseek --skill investigate-lyh2

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematic debugging with a structured root-cause investigation workflow to prevent patchwork fixes and recurring outages.

Core Features & Use Cases

  • Four-phase approach: investigate, analyze, hypothesize, implement, to identify the real root cause before applying fixes.
  • Tooling integration and evidence gathering (grep, read, write, web search) to reproduce issues and validate hypotheses.
  • Scope control and governance via pretool freeze checks and plan-mode safeguards to minimize collateral changes.
  • Telemetry and contributor mode to improve the workflow over time and share learnings with the team.

Quick Start

Describe a bug you see, then run the root-cause investigation workflow to identify the underlying cause.

Frequently Asked Questions about investigate

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

FAQPage Schema
How do I find the root cause of a bug regression instead of applying patchwork fixes?

To find the root cause of a bug regression, you need a systematic debugging workflow that enforces evidence gathering and structured analysis before implementing fixes. This approach guides you from symptoms to evidence-driven remedies across four phases: investigate, analyze, hypothesize, and implement.

What is the best way to debug complex codebases for performance issues?

The best way to debug complex codebases for performance issues is using a structured root-cause investigation approach. It enforces tool-assisted analysis and safety checks to prevent unsafe changes, guiding you systematically from observed symptoms to evidence-driven remedies.

How do I start a systematic investigation workflow for software debugging?

To start a systematic investigation workflow for software debugging, describe the bug symptoms you observe. The workflow then enforces a four-phase process—investigate, analyze, hypothesize, implement—using tools like grep and web search to gather evidence before any fixes are applied.

Does root-cause debugging prevent premature or unsafe code changes?

Yes, root-cause debugging prevents premature or unsafe code changes by enforcing scope control and governance. It uses pretool freeze checks and plan-mode safeguards to minimize collateral changes, ensuring you validate hypotheses with evidence before modifying code.

Can I use telemetry to improve my debugging workflow over time?

Yes, you can use telemetry to improve your debugging workflow over time. The investigation process includes telemetry and contributor mode features designed to capture learnings from your root-cause analysis and share them with your team to prevent recurring outages.