analyze

Analyze codebases to rank evidence-based explanations of behaviors and failures.

1|Updated May 7, 2026
One-click install
npx skills add https://github.com/zchee/json-repair-rs --skill analyze-zchee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze
Source: https://github.com/zchee/json-repair-rs/tree/main/.codex/skills/analyze
Command: npx skills add https://github.com/zchee/json-repair-rs --skill analyze-zchee

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps users understand complex codebases by providing grounded, read-only analysis that ranks possible explanations with confidence levels, aiding in informed decision-making.

Core Features & Use Cases

  • Deep Repository Analysis: Explains features, behaviors, or failures across multiple files based on code evidence.
  • Ranked Hypotheses: Prioritizes explanations by supporting evidence, helping users identify the most probable causes.
  • Use Case: When debugging a performance issue, this Skill can analyze the code to suggest the most likely bottleneck or cause based on the code structure and contracts.

Quick Start

Ask the analyze skill to review the repository and clarify its architecture or behaviors before making changes.

Frequently Asked Questions about analyze

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

FAQPage Schema
How do I analyze a codebase to find the cause of a bug?

To analyze a codebase and find a bug, this Skill examines multiple files, traces code paths, and evaluates evidence to provide ranked hypotheses of the most probable causes. It prioritizes explanations by supporting code evidence to aid debugging.

What is evidence-based codebase analysis for architecture review?

Evidence-based codebase analysis is a read-only process that clarifies repository architecture and behaviors by tracing code paths across multiple files. It provides grounded insights into structural decisions to support informed development planning.

Can I use this Skill to understand repository behaviors without modifying the code?

Yes, you can safely understand repository behaviors without modifying code because the Skill performs strictly read-only, non-intrusive analysis. It comprehends complex codebases by evaluating evidence without executing or altering the existing code structure.

How does this approach rank explanations for codebase failures?

This approach ranks explanations for codebase failures by evaluating supporting evidence across multiple files and tracing execution paths. It assigns confidence levels to each hypothesis, helping developers identify the most likely architectural or behavioral causes.

What is the best way to review unfamiliar repository architecture before making changes?

The best way to review unfamiliar repository architecture is to request a grounded, read-only analysis that traces code paths and explains structural decisions. This provides ranked insights into behaviors and contracts, ensuring informed decisions before modifications.

Are there limitations to analyzing code paths for debugging complex repositories?

A limitation of analyzing code paths for debugging is that the process is strictly read-only and does not execute the code, meaning dynamic runtime failures might not be fully captured. It relies on static structural evidence to rank behavioral hypotheses.