Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/openai/skills/llms.txt

Use this file to discover all available pages before exploring further.

OpenAI Skills are self-contained capability packages for Codex. Each skill is a folder containing a SKILL.md instruction file, optional scripts, reference documents, and assets — everything an AI agent needs to reliably perform a specific task, from deploying to Vercel to transcribing audio to navigating a Figma design.

Introduction

Learn what skills are, how they work, and when to use them.

Quickstart

Install your first skill and put it to use in minutes.

Create a Skill

Build a custom skill with the bundled scaffolding tools.

Skill Catalog

Browse all system and curated skills ready to install.

How Skills Work

Skills use a three-level loading system that keeps Codex’s context window efficient. Only what’s needed is ever loaded.
1

Metadata is always present

Every skill’s name and description from SKILL.md frontmatter are always in Codex’s context — roughly 100 words per skill. This is how Codex decides which skill to activate.
2

SKILL.md body loads on trigger

When Codex determines a skill is relevant, the full SKILL.md body — workflow steps, decision trees, script usage — is loaded. Keep it under 500 lines.
3

Bundled resources load on demand

Scripts in scripts/, reference docs in references/, and assets in assets/ are only read when Codex determines it needs them — keeping the context window lean.

Skill Tiers

System Skills

Pre-installed with every Codex instance. Includes skill-creator, skill-installer, imagegen, openai-docs, and plugin-creator.

Curated Skills

Maintained in skills/.curated — install by name using $skill-installer. Covers deployment, dev tools, design, media, and productivity.

Experimental Skills

Work-in-progress skills in skills/.experimental — install by specifying the folder path or a GitHub URL.

Build docs developers (and LLMs) love