codex

Delegate coding and review workflows to the OpenAI Codex CLI in git repositories.

19|4|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill codex-carterwayneskhizeine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/carterwayneskhizeine/hermes-agent-windows-R/tree/main/skills/autonomous-ai-agents/codex
Command: npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill codex-carterwayneskhizeine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coding assistants can draft code, but coordinating iterative changes, reviewing diffs, and producing commit-ready output across files is still time-consuming and error-prone.

Core Features & Use Cases

  • Codex-driven coding in a real git repo: Runs OpenAI Codex CLI to implement features, refactor modules, and apply multi-file edits that are tied to a workspace’s git history.
  • One-shot execution and long-running background sessions: Supports short tasks that exit on completion and longer work that can be monitored, polled, and controlled.
  • PR review and parallel issue fixing: Enables review workflows by checking out PR branches and supports running multiple Codex instances in parallel using git worktrees for faster issue throughput.

Quick Start

Ask your agent to delegate a coding task to Codex by running a one-shot Codex exec command inside your git repository with a PTY enabled.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I automate pull request reviews and code refactoring in a git repository?

Automating pull request reviews and code refactoring requires delegating natural-language prompts to the OpenAI Codex CLI within a valid git workspace. The tool checks out PR branches, applies multi-file edits, and produces commit-ready repository changes tied to git history.

Can I run multiple coding agent instances in parallel for faster issue fixing?

Running multiple coding agent instances in parallel is supported using git worktrees. This allows you to manage multiple Codex CLI background sessions simultaneously, monitoring and polling interactive prompts to increase batch issue fixing throughput.

Do I need a PTY-enabled terminal to execute Codex CLI coding tasks?

A PTY-enabled terminal is required to execute Codex CLI coding tasks. The tool transforms natural-language prompts into repository changes by making interactive terminal calls, needing PTY support to handle prompts and optional background session management.

What is the best way to delegate feature development to an AI coding agent?

The best way to delegate feature development to an AI coding agent is using one-shot execution commands via the Codex CLI. This transforms your natural-language prompt directly into multi-file feature implementations within your git repository.

How does terminal automation handle long-running background sessions for coding workflows?

Terminal automation handles long-running coding workflows by supporting background session management with polling. You can monitor, poll, and control extended Codex CLI tasks, submitting responses for interactive prompts until the code changes are completed.

When should I not use a CLI coding agent for repository changes?

You should not use a CLI coding agent for repository changes without Codex CLI authentication or outside a valid git directory. The tool requires an interactive terminal execution environment to transform prompts into code safely.