codex

Automate Codex CLI tasks with model selection and reasoning control.

1|Updated Aug 25, 2025
One-click install
npx skills add https://github.com/Alevsk/nix --skill codex-alevsk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/Alevsk/nix/tree/main/modules/ai/claude/skills/codex
Command: npx skills add https://github.com/Alevsk/nix --skill codex-alevsk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines the process of running Codex CLI tasks for code analysis, refactoring, and automated editing, eliminating manual setup and repetitive command construction.

Core Features & Use Cases

  • Model & reasoning control: Choose the Codex model and apply desired reasoning effort in a single prompt.
  • Safe, auditable execution: Enforce sandbox modes and persistent session handling with clear resume capabilities.
  • Guided command assembly: Build codex exec commands with standardized flags and directory context.

Quick Start

Start a Codex task by selecting a model and reasoning effort, then run a session:

  • codex exec -m gpt-5-codex --config model_reasoning_effort="high" --sandbox read-only --full-auto
  • To resume a session: echo "your prompt here" | codex exec --skip-git-repo-check resume --last 2>/dev/null

Frequently Asked Questions about codex

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

FAQPage Schema
How do I automate Codex CLI tasks for code analysis and refactoring?

You can automate Codex CLI tasks by using guided command assembly to select models, apply reasoning effort, and execute code analysis or refactoring within a single prompt.

Can I resume an interactive Codex CLI session after stopping it?

Yes, you can resume a session using the resume flag with the last command, allowing you to continue resumable workflows and persistent session handling directly from the CLI.

How do I enforce sandbox safety when running automated code edits?

You enforce sandbox safety by applying sandbox modes like read-only during command assembly, ensuring controlled execution and safe, auditable execution for automated code edits.

Does Codex CLI support model selection and reasoning effort control in one command?

Yes, Codex CLI supports model selection and reasoning effort control by passing model and reasoning configuration flags together within a single prompt execution command.

What is the best way to standardize flag handling in codex exec commands?

The best way to standardize flag handling in codex exec commands is through guided command assembly, which builds standardized flags and directory context automatically.

When should I use read-only sandbox mode for code analysis?

You should use read-only sandbox mode when performing code analysis that requires safe, auditable execution without modifying the underlying files or repository state.