d-inspect

Generate ranked root-cause hypotheses from bug symptoms and code paths.

Updated Jan 22, 2026
One-click install
npx skills add https://github.com/Hoang604/get-thing-done --skill d-inspect
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: d-inspect
Source: https://github.com/Hoang604/get-thing-done/tree/main/.claude/skills/d-inspect
Command: npx skills add https://github.com/Hoang604/get-thing-done --skill d-inspect

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps engineers diagnose root causes of bugs by analyzing symptoms and tracing code paths to generate ranked hypotheses.

Core Features & Use Cases

  • Read symptom descriptions
  • Inspect relevant code paths
  • Form multiple hypotheses ranked by confidence
  • Document reasoning for each hypothesis

Quick Start

Provide a symptom description at ./.gtd/debug/current/SYMPTOM.md and run the /d-symptom flow to load the symptom and trigger hypothesis generation.

Frequently Asked Questions about d-inspect

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

FAQPage Schema
How do I generate root-cause hypotheses from bug symptoms during debugging?

You provide a symptom description, and the tool traces relevant code paths to generate ranked root-cause hypotheses. It analyzes the symptoms, evaluates the code, and outputs a documented file containing hypotheses with confidence scores.

What is the best way to document debugging hypotheses for complex codebases?

The best way to document debugging hypotheses is to generate a structured markdown file that records ranked root-cause hypotheses and their reasoning. This ensures traceability of symptom analysis and code path traversal across your codebase.

How do I start a debugging session to analyze code paths from a symptom description?

Place your symptom description in a designated local markdown file and run the symptom loading flow. This triggers the inspection process, which reads the file, traverses the relevant code paths, and generates ranked hypotheses.

Can I trace code paths to form multiple hypotheses without additional dependencies?

Yes, you can trace code paths and form multiple hypotheses without any external dependencies. The tool operates independently to analyze symptoms, traverse the code, and output a markdown file documenting the ranked root-cause hypotheses.

Why does my debugging process need ranked confidence scores for root-cause hypotheses?

Ranked confidence scores are needed to prioritize debugging efforts effectively. By scoring each hypothesis based on symptom analysis and code path traversal, you focus on the most probable root causes first, reducing time spent on unlikely code paths.