Run OpenAI Codex in GitHub Actions with gh-aw

OpenAI Codex is OpenAI’s coding-focused agent runtime for repository work. gh-aw runs Codex inside GitHub Actions from a Markdown workflow and layers GitHub triggers, sandbox controls, and safe outputs on top so repository automation can stay event-driven and reviewable.

Setup

Set engine: codex and provide OPENAI_API_KEY. CODEX_API_KEY is also accepted and takes precedence when both are present. See OpenAI authentication for the supported configuration.

Initialize the repository

Run gh aw init --engine codex to configure the repository. The --engine codex flag skips Copilot-specific files (MCP server configuration, Copilot dispatcher skill) and writes only the files useful for any engine: .gitattributes, VS Code settings, and the custom agent file.

gh aw init --engine codex

---
on:
  schedule: daily

permissions:
  contents: read
  issues: read
  pull-requests: read

engine: codex

safe-outputs:
  create-issue:
    title-prefix: "[status] "
    labels: [report]
    close-older-issues: true
---

# Daily Repository Status

Analyze the repository and create a concise daily status report covering:
- Open issues and their priority
- Recent PR activity
- Upcoming work items

When to choose gh-aw vs. running Codex directly in Actions

Choose gh-aw when the workflow should stay in Markdown, reuse the same structure across engines, and enforce GitHub writes through safe outputs. Run Codex directly in Actions when the job needs a custom script layout and manual control over every integration point.