Which models are supported by this skill?

The skill supports various models through the Codex CLI, including gpt-5, gpt-5.1, and specialized variants like gpt-5.1-codex-max.

How do I enable full automation for programming?

Use the --full-auto flag or the -s workspace-write sandbox mode to allow Codex to create, modify, and delete files within your workspace.

What is the primary benefit of the Codex skill?

It allows for non-interactive, hands-off execution of complex coding tasks, meaning the AI can complete a full implementation plan without constant user approval prompts.

Is it safe to let Codex write to my files?

Yes, you can control safety using sandbox modes. The 'workspace-write' mode limits changes to your current project, while 'read-only' prevents any modifications.

Does this skill require any external installations?

Yes, it requires the Codex CLI to be installed and configured on your system (e.g., via npm or Homebrew) to function correctly.

Codex Automation Skill

Name: Codex Automation Skill
Author: Microck

byMicrock

•

生産性とワークフロー

Automates complex programming tasks using the Codex CLI for hands-off feature implementation and code refactoring.

概要

This skill enables Claude to delegate complex, time-consuming implementation plans to the Codex CLI via the codex exec command. It supports a fully autonomous, non-interactive execution mode, allowing for hands-off code generation, bug fixing, and system analysis. With support for various sandbox security levels—ranging from read-only analysis to full-system access—it allows developers to leverage advanced models like GPT-5.1 for production-grade coding tasks without needing to approve every incremental step during the workflow.

主な機能

Structured JSON output for progress tracking and performance metrics
89 GitHub stars
Support for advanced models like GPT-5 and GPT-5.1-codex
Autonomous decision-making based on codebase best practices
Non-interactive automation for hands-off task execution
Multiple sandbox modes including Read-Only and Workspace-Write

ユースケース

Automated security audits and dependency updates in a sandboxed environment
Implementing multi-file features or architectural plans designed by Claude
Performing large-scale code refactoring across entire workspaces autonomously

概要

主な機能

Structured JSON output for progress tracking and performance metrics
89 GitHub stars
Support for advanced models like GPT-5 and GPT-5.1-codex
Autonomous decision-making based on codebase best practices
Non-interactive automation for hands-off task execution
Multiple sandbox modes including Read-Only and Workspace-Write

ユースケース

Automated security audits and dependency updates in a sandboxed environment
Implementing multi-file features or architectural plans designed by Claude
Performing large-scale code refactoring across entire workspaces autonomously