analyze

Analyze repository code to produce ranked, evidence-based interpretations with file references.

24|2|Updated Mar 12, 2025
One-click install
npx skills add https://github.com/Orinks/AccessiWeather --skill analyze-orinks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze
Source: https://github.com/Orinks/AccessiWeather/tree/main/.codex/skills/analyze
Command: npx skills add https://github.com/Orinks/AccessiWeather --skill analyze-orinks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Read-only, evidence-based analysis of a codebase to reveal what the repository says about its behavior, structure, and dependencies, without proposing changes.

Core Features & Use Cases

  • Read-Only Synthesis: Generate a ranked synthesis with explicit confidence, concrete file references, and clear evidence-vs-inference boundaries.
  • Cross-file Tracing: Trace framework and module interactions to explain how components work together.
  • Evidence-first Reasoning: Separate observed facts from interpretation, with clearly labeled sources and rationale.
  • Pre-change Grounding: Provide grounded explanations before any proposed modifications, enabling safer decision-making.

Quick Start

Run a read-only analysis of the repository to produce a ranked, evidence-based synthesis.

Frequently Asked Questions about analyze

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

FAQPage Schema
How do I perform a read-only analysis of a codebase to understand its architecture?

Trace cross-file module interactions to map framework dependencies and component relationships, producing a ranked synthesis that explains how different parts of the codebase work together with explicit confidence levels.

What is evidence-based codebase analysis and when do I need it?

Evidence-based codebase analysis is a read-only method that separates observed repository facts from interpretation to explain structural behavior, needed when you require grounded architectural understanding before proposing any code modifications.

Can I trace cross-file module interactions without modifying the repository?

Yes, you can trace cross-file module interactions in a strict read-only mode, mapping framework dependencies and component relationships to explain how modules work together while ensuring the repository remains completely unchanged.

What's the best way to ground decisions before making code changes?

The best way to ground pre-change decisions is running a read-only repository analysis to obtain a ranked synthesis with concrete file references and clearly labeled sources, ensuring safer modifications through an explicit evidence-vs-inference boundary.

Does read-only codebase synthesis provide confidence levels for architectural interpretations?

Yes, read-only codebase synthesis provides a ranked set of architectural interpretations with explicit confidence levels and concrete file references, ensuring an evidence-vs-inference boundary so you can evaluate behavioral claims without implementing changes.