codex

Automate Codex CLI code analysis, refactoring, and editing with sandbox controls.

16|7|Updated Nov 20, 2025
One-click install
npx skills add https://github.com/jackspace/ClaudeSkillz --skill codex-jackspace
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/jackspace/ClaudeSkillz/tree/main/skills/codex
Command: npx skills add https://github.com/jackspace/ClaudeSkillz --skill codex-jackspace

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill provides comprehensive guidance on using the Codex CLI for automated code analysis, refactoring, and deterministic edits, including HPC/Slurm considerations and safe execution patterns.

Core Features & Use Cases

  • ✅ Deterministic code analysis and automated edits via codex exec
  • ✅ Stdin-based pipelines for blocked files on HPC
  • ✅ Resume sessions and profile-based execution
  • ✅ Cross-directory execution and workspace management

Quick Start

Install Codex CLI, connect to your project, and run a sample "analyze" or "refactor" command in a Harmless HPC context.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I automate code analysis and refactoring across my codebase?

Codex CLI automates deterministic code analysis and refactoring by executing analysis tasks and applying automated edits across your project. Install Codex CLI, connect to your codebase, and run codex exec commands with analysis or refactoring parameters to perform syntax review, architecture analysis, and multi-file refactoring in a controlled sandbox environment.

Can I use Codex CLI for automated code review and editing on HPC systems?

Yes, Codex CLI supports HPC and Slurm environments with sandbox controls and model selection constraints. It detects your environment automatically and provides stdin-based pipelines for blocked files, resume sessions, and profile-based configurations—use the --yolo flag to override sandbox restrictions on HPC clusters when needed.

What are the sandbox modes and safety constraints when running Codex CLI?

Codex CLI offers three sandbox modes: read-only (analysis only), workspace-write (safe edits to project files), and danger-full-access (unrestricted changes). HPC environments override sandbox mode with --yolo when authorized. Configure your execution with essential flags like -m (model), -c (config), -s (sandbox), and -C (constraints) to control risk and output behavior.

How do I handle large files and complex refactoring tasks with Codex CLI?

Codex CLI handles large file analysis and multi-file refactoring through command construction with flags like --full-auto for deterministic execution and --skip-git-repo-check for cross-directory projects. Use stderr redirection, execution validation, and resume support to manage long-running or complex tasks across your entire codebase.

Do I need to manually configure Codex CLI for each project or can I use profiles?

Codex CLI supports profile-based configurations that persist settings across projects, eliminating repetitive manual setup. Define profiles with your preferred models, sandbox levels, and command flags, then reuse them across different codebases and environments.

What happens when Codex CLI encounters errors or needs to resume an incomplete task?

Codex CLI validates execution and captures stderr for error handling and diagnostics. Resume sessions allow you to restart interrupted analysis or refactoring tasks without losing progress, supporting long-running operations and interrupted workflows in unstable HPC environments.