review-code

name: review-code description: Review and explore GitHub code repositories using Codex.

This skill specifies how to use Codex to explore code repositories.

Overview

You have access to the command-line tool codex.
codex is an AI coding agent that can read, modify, and execute code locally within a specified directory.

Core Rules

• codex is already installed and ready to use.
• You must run codex in exec mode for all operations.
  • Example:

    codex exec "Print to terminal a high-level overview of this repo"
    

• All work must occur inside ${CODEX_SANDBOX}, using these paths:
  • Repositories: ${CODEX_SANDBOX}/repos
  • Temporary files: ${CODEX_SANDBOX}/tmp
• Never execute outside of ${CODEX_SANDBOX} or modify system-level files.
• IMPORTANT: Confirm that ${CODEX_SANDBOX} exists prior to running Codex.

Example Workflow (Command Line)

cd ${CODEX_SANDBOX} # Navigate to Codex workspace codex exec whoami # Confirm identity cd ${CODEX_SANDBOX}/repos # Clone a repository to explore git clone https://github.com/uhsealevelcenter/IDEA IDEA cd ${CODEX_SANDBOX}/repos/IDEA # Enter repository and analyze codex exec "Print to terminal a high-level overview of this repo"

Usage Tips

• Invoke Codex directly from the shell using codex exec ... by default.
• Do not wrap Codex in Python subprocess unless requested or if direct shell execution is unavailable.
• For repository tasks, prefer: codex exec -C /path/to/repo --full-auto ""
• For work outside of repositories, use: codex exec --full-auto --skip-git-repo-check ""

Example Codex Commands

codex exec "Print to terminal the full code for the data processing module" codex exec "Summarize this repository structure and highlight the main entry points" codex exec "Generate docstrings for all public functions in this package" codex exec "Identify where to insert Codex integration hooks in this backend"

Example Code Repositories

https://github.com/uhsealevelcenter/IDEA # Intelligent Data Exploring Assistant https://github.com/uhsealevelcenter/Wyrtki-CSLIM # ENSO forecast model https://github.com/uhsealevelcenter/QCSoft # UHSLC GUI tool for Quality Control