Platforms
Codex Skills
This page uses OpenAI Codex material as a baseline and focuses on repository context, handoff loops, and AGENTS.md-style collaboration.
What you'll find here
OpenAI presents Codex as a product for getting real work done in repositories, not as a generic chat surface. For Codex skills, the key question is whether a workflow holds up in code review, issue handling, and day-to-day engineering tasks.
That makes this page a narrowing tool for Agent Skills that truly fit repository work. Start here to compare skills that explicitly line up with repository context, code changes, review loops, or AGENTS.md conventions before opening full details.
If a skill mentions coding in broad terms but never explains repository assumptions, output boundaries, or collaboration expectations, it is usually a weak Codex fit. Codex users need reusable workflows that survive real project work, not just elegant templates.
Official references
Related skills
Start with the skills that are closest to this topic and easiest to evaluate.


