code-sleuth

Trace code execution paths and cite file lines for root-cause analysis.

1|Updated Jun 7, 2024
One-click install
npx skills add https://github.com/jsco2t/dotfiles --skill code-sleuth
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-sleuth
Source: https://github.com/jsco2t/dotfiles/tree/main/.claude/skills/code-sleuth
Command: npx skills add https://github.com/jsco2t/dotfiles --skill code-sleuth

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Code Sleuth helps engineers uncover why code behaves as it does by building rigorous mental models, tracing interactions, and surfacing evidence from actual code. It prioritizes proof, chain-of-thought, and stepwise validation over guesswork, enabling precise root-cause analysis.

Core Features & Use Cases

  • Build a mental model of a codebase, mapping components, data flow, and contracts.
  • Trace execution paths end-to-end, recording decisions and side effects to justify conclusions.
  • Provide evidence-backed explanations for bugs or design questions, citing file paths, functions, and tests.

Quick Start

Use Code Sleuth to begin a structured investigation of a failing function by tracing its call chain and collecting code-level evidence.

Frequently Asked Questions about code-sleuth

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

FAQPage Schema
How do I find the root cause of a bug by tracing code execution paths?

Codebase investigation builds mental models mapping components, data flow, and contracts to explain unexpected behavior. It provides evidence-backed explanations for design questions by citing file paths, functions, and version history.

How do I debug unexpected behavior using git history and code analysis?

Trace execution paths end-to-end by recording decisions and side effects to justify conclusions with specific file and line citations. This satisfies requirements for evidence-backed conclusions and reproducible reasoning.

Can I use code analysis to understand component interactions and data flow?

Apply structured investigation to trace failing function call chains and collect code-level evidence. This stepwise validation prioritizes proof over guesswork to enable precise root-cause analysis.

What is the best way to investigate a codebase for reproducible reasoning?

Build a mental model of a codebase by mapping components, data flow, and contracts to investigate root causes. This approach prioritizes proof and stepwise validation over guesswork for precise analysis.

When should I not use evidence-based code investigation for debugging?

Apply structured investigation to trace failing function call chains and collect code-level evidence. This stepwise validation prioritizes proof over guesswork to enable precise root-cause analysis.