Can I use specific versions of GPT-5?

Yes, the skill allows you to specify models such as gpt-5, gpt-5.1, or gpt-5.1-codex using the -m or --model flags.

What is the primary benefit of the Codex skill?

It enables 'codex exec' mode, which allows Claude to perform complex tasks autonomously using the Codex CLI and GPT-5 models without needing approval for every individual action.

How do I enable file editing with this skill?

You should use the Workspace-Write mode by adding the --full-auto or -s workspace-write flags to the codex command. This is the recommended mode for most programming tasks.

What are the sandbox modes available?

There are three modes: Read-Only (default) for analysis, Workspace-Write for editing files within your project, and Danger-Full-Access for system-level operations and network access.

Codex Automation Skill

Name: Codex Automation Skill
Author: Activer007

byActiver007

Productivity & Workflow

Enables non-interactive, hands-off automation for executing complex development tasks using Codex and GPT-5 models.

About

The codex-skill integrates the Codex CLI into Claude's environment, allowing for fully autonomous execution of technical plans and feature implementations. It enables Claude to leverage advanced models like GPT-5 and GPT-5.1 to perform codebase-wide refactoring, bug fixes, and feature development without requiring constant user approval. With granular sandbox modes—ranging from secure read-only analysis to full system access—it provides a flexible framework for streamlining end-to-end development workflows and increasing developer velocity.

Key Features

0 GitHub stars
Support for advanced models including GPT-5, GPT-5.1, and GPT-5.1-codex
Structured JSON output for integration with other tools and automated reporting
Configurable sandbox policies (Read-Only, Workspace-Write, and Danger-Full-Access)
Autonomous execution mode for hands-off task completion without approval prompts
Comprehensive CLI support for model selection, profiles, and custom directories

Use Cases

Performing codebase-wide migrations or refactors autonomously
Executing an entire feature implementation plan designed by Claude in one pass
Running automated security audits and performance benchmarks with detailed summaries

About

Key Features

0 GitHub stars
Support for advanced models including GPT-5, GPT-5.1, and GPT-5.1-codex
Structured JSON output for integration with other tools and automated reporting
Configurable sandbox policies (Read-Only, Workspace-Write, and Danger-Full-Access)
Autonomous execution mode for hands-off task completion without approval prompts
Comprehensive CLI support for model selection, profiles, and custom directories

Use Cases

Performing codebase-wide migrations or refactors autonomously
Executing an entire feature implementation plan designed by Claude in one pass
Running automated security audits and performance benchmarks with detailed summaries