how-to-guide

How to View OpenAI Skills Code: A Complete Guide to the Public Repository

February 16, 2026 openai/skills ↗

Yes, the OpenAI Skills repository is fully public on GitHub, allowing anyone to browse, clone, and explore the source code for skill installers and curated skill packages without authentication.

The OpenAI Skills repository provides a collection of modular, reusable automation components designed to extend AI capabilities. If you want to view OpenAI skills code to understand implementation details, customize existing skills, or contribute your own, you can access the entire codebase directly on GitHub under the openai/skills repository.

Where to Access the OpenAI Skills Repository

The complete source code for OpenAI Skills is hosted publicly at:


https://github.com/openai/skills

All files reside on the main branch, which serves as the default and current stable version. You can navigate the repository directly in your browser to inspect file contents, directory structures, and commit history without downloading anything locally.

Key Entry Points for Exploring the Codebase

Understanding where to start when you view OpenAI skills code helps you navigate the repository efficiently.

The README Documentation

The top-level README.md provides the primary entry point, explaining the repository's purpose, architecture overview, and basic usage instructions. This file contains high-level context before you dive into specific implementation files.

Skill Installer Scripts

The skill-installer tooling provides the infrastructure for managing skills. Located in skills/.system/skill-installer/, this directory contains scripts that handle listing, installing, and configuring skills. The SKILL.md file in this directory documents the installer's API and usage patterns.

Curated Skill Collection

The skills/.curated/ directory houses production-ready skill implementations that OpenAI maintains. Each subdirectory represents a self-contained skill package with its own scripts, configuration files, and documentation. Examples include image generation, Netlify deployment, and Notion integration skills.

Repository Structure and Organization

When you view OpenAI skills code, you'll notice a clear separation between system infrastructure and skill implementations.

The .system directory contains shared tooling and installation scripts that support the entire ecosystem. This infrastructure is separate from the actual skills, ensuring that the installer logic remains independent of any specific skill's functionality.

Individual skills follow a self-contained package structure. Each skill resides in its own directory with standardized subdirectories like scripts/ for executable code and agents/ for YAML configuration descriptors. This modularity allows you to examine, modify, or deploy individual skills without affecting the broader system.

How to Browse and Install Skills from Source

You can interact with the OpenAI Skills code directly using the provided Python scripts.

Listing Available Skills

To see every skill available in the repository, run the listing script from the repository root:


# list-skills.py – shows every skill that can be installed

# Run from the repository root

# ------------------------------------------------

#   $ python -m skills/.system/skill-installer/scripts/list-skills.py

# ------------------------------------------------

This script scans the repository structure and outputs all discoverable skills, including those in the curated collection and any custom skills you've added.

Installing Skills from GitHub

To install a specific skill from a remote GitHub location, use the installation script with the target URL:


# install-skill-from-github.py – fetches a skill definition and adds it to your local catalog

# ------------------------------------------------

$ python -m skills/.system/skill-installer/scripts/install-skill-from-github.py \
    https://github.com/openai/skills/tree/main/skills/.curated/imagegen

# ------------------------------------------------

This command downloads the image generation skill from the curated collection and registers it in your local environment.

Using Curated Skill Implementations

Each curated skill includes a YAML descriptor that defines how to invoke it. For example, the image generation skill uses the following configuration:


# Example OpenAI‑compatible YAML for the image‑gen skill

agents:
  openai:
    description: Generate images from text prompts
    steps:
      - name: generate
        run: |
          python -m skills/.curated/imagegen/scripts/image_gen.py "$PROMPT"

You can invoke this skill through any OpenAI‑compatible runtime that reads the openai.yaml descriptor located at skills/.curated/imagegen/agents/openai.yaml.

Essential Files to Review

When you view OpenAI skills code, these specific files provide the most insight into architecture and implementation:

These files demonstrate how the repository separates system infrastructure from skill implementations, and how each skill uses standardized descriptors to define its capabilities.

Summary

  • The OpenAI Skills repository is publicly available at https://github.com/openai/skills with all source code accessible on the main branch.
  • You can view OpenAI skills code directly in your browser or clone the repository locally using standard git commands.
  • The repository separates system infrastructure (installer tools in .system) from skill implementations (curated packages in .curated).
  • Key scripts like list-skills.py and install-skill-from-github.py provide programmatic access to browse and manage skills.
  • Each skill uses a standardized YAML descriptor (typically agents/openai.yaml) to define its interface and execution steps.

Frequently Asked Questions

Is the OpenAI Skills code open source?

Yes, the OpenAI Skills repository is fully open source and hosted publicly on GitHub. You can view OpenAI skills code without authentication, clone the repository, fork it for your own modifications, and submit contributions following the guidelines in contributing.md.

How do I clone the OpenAI Skills repository locally?

To clone the repository to your local machine, run the standard git command:

git clone https://github.com/openai/skills.git

This creates a local copy of the entire repository, including all curated skills, system installer scripts, and documentation, allowing you to browse and execute code offline.

What programming languages are used in OpenAI Skills?

The repository primarily uses Python for skill implementations and system tooling, as evidenced by scripts like image_gen.py, list-skills.py, and install-skill-from-github.py. Configuration files use YAML (particularly in agents/openai.yaml descriptors), and documentation is written in Markdown.

Can I contribute my own skills to the repository?

Yes, you can contribute new skills by following the guidelines outlined in contributing.md. The repository is designed to accept community contributions, with each skill packaged as a self-contained module containing scripts, YAML descriptors, and documentation. You submit contributions through standard GitHub pull requests after forking the repository.

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:
curl -s "https://instagit.com/install.md"
Works with
Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →