
openai / skills
✓ Official★ 23,200A skill package that teaches your agent 42 capabilities — every one documented and browsable below, no GitHub required · by openai.
Each skill below is one capability this package teaches your agent. Install the whole package, or open a skill to install just that one.
Build, review, refactor, or architect ASP.NET Core web applications using current official guidance for .NET web development. Use when working on Blazor Web…
20 files — installable on its own
Build, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications that combine an MCP server and widget UI. Use when Codex needs to design tools,…
14 files — installable on its own
Build a composable CLI for Codex from API docs, an OpenAPI spec, existing curl examples, an SDK, a web app, an admin tool, or a local script. Use when the user…
6 files — installable on its own
Deploy applications and infrastructure to Cloudflare using Workers, Pages, and platform services. Covers 40+ Cloudflare products organized by use case: compute (Workers, Pages, Durable Objects, Workflows), storage (KV, D1, R2, Queues), AI/ML (Workers AI, Vectorize, Agents SDK), networking (Tunnel, Spectrum), and security (WAF, Bot Management, Turnstile) Includes decision trees to route users to the right product based on their need (run code, store data, add AI, networking, security, or...
50 files — installable on its own
Fetch design context, screenshots, and assets from Figma, then translate designs into production code. Provides three core commands: get_design_context for structured node representation, get_screenshot for visual reference, and asset download via localhost endpoints Requires a mandatory workflow: fetch design context, retrieve screenshot, download assets, then implement using project conventions and design tokens Outputs React + Tailwind as a design representation; implementation must reuse...
11 files — installable on its own
Connects Figma design components to code components using Code Connect mapping tools. Use when user says "code connect", "connect this component to code", "map…
12 files — installable on its own
Generates custom design system rules for the user's codebase. Use when user says "create design system rules", "generate rules for my project", "set up design…
12 files — installable on its own
Create a new blank Figma file. Use when the user wants to create a new Figma design or FigJam file, or when you need a new file before calling use_figma.…
9 files — installable on its own
Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to…
9 files — installable on its own
Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, set…
27 files — installable on its own
Translate Figma designs into production-ready code with pixel-perfect visual fidelity. Structured seven-step workflow: extract node IDs from Figma URLs, fetch design context and screenshots, download assets, translate to project conventions, and validate against Figma specs Integrates with Figma MCP server (remote or desktop) to access layout properties, typography, colors, design tokens, and component structure Emphasizes design system reuse, design token mapping, and 1:1 visual parity;...
8 files — installable on its own
**MANDATORY prerequisite** — you MUST invoke this skill BEFORE every `use_figma` tool call. NEVER call `use_figma` directly without loading this skill first.…
33 files — installable on its own
Locate and address review comments on GitHub PRs using gh CLI with user-guided workflows. Fetches all comments and review threads from the open PR for the current branch, displaying them with summaries of required fixes Requires gh CLI authentication with elevated permissions (workflow/repo scopes); prompts re-authentication if rate limits or auth errors occur mid-run Presents numbered comment list to user for selection, then applies fixes to chosen comments Handles sandboxing constraints by...
9 files — installable on its own
Debug and fix failing GitHub Actions checks in pull requests using gh CLI inspection and log analysis. Inspects PR checks via GitHub CLI, fetches GitHub Actions logs, and summarizes failure context with actionable log snippets Requires prior gh authentication with repo and workflow scopes; verifies status before proceeding Drafts a fix plan (using create-plan skill if available, otherwise inline) and implements only after explicit user approval Treats external CI providers (Buildkite, etc.)...
9 files — installable on its own
Create, repair, validate, preview, and package Codex-compatible animated pets and pet spritesheets from character art, screenshots, generated images, or visual…
17 files — installable on its own
Generate or edit images via OpenAI's API with a bundled CLI for deterministic, reproducible runs. Supports three workflows: generate new images, edit existing images (inpainting, masking, background replacement, object removal), and batch runs across multiple prompts or variants Defaults to gpt-image-1.5 and requires OPENAI_API_KEY for live API calls; uses the bundled scripts/image_gen.py CLI for all operations Includes a structured decision tree and prompt augmentation template to classify...
15 files — installable on its own
Create and scaffold Jupyter notebooks for experiments and tutorials with bundled templates. Two notebook kinds: experiment for exploratory analysis and hypothesis-driven work, tutorial for instructional step-by-step content Helper script new_notebook.py generates clean notebooks from templates, avoiding manual JSON authoring Workflow emphasizes small, focused code cells paired with markdown explanations, with reference guides for experiment patterns, tutorial structure, and safe editing of...
16 files — installable on its own
Manage Linear issues, projects, and team workflows with structured MCP integration. Provides 20+ tools across issue management, project planning, documentation search, and team collaboration Supports eight practical workflows including sprint planning, bug triage, workload balancing, release planning, and retrospectives Requires Linear MCP server setup with OAuth authentication; includes Windows/WSL configuration guidance Executes operations in logical batches: read first for context, then...
7 files — installable on its own
Migrate supported instruction files, skills, agents, and MCP config into Codex project and global files.
24 files — installable on its own
Deploy web projects to Netlify with automatic authentication, site linking, and framework detection. Handles full deployment workflow: authentication verification, site linking or creation, dependency installation, and preview or production deploys Supports both browser-based OAuth and API token authentication; guides users through login if needed Automatically detects project configuration from netlify.toml or prompts for build command and publish directory based on framework Includes error...
11 files — installable on its own
Capture conversations and decisions into structured, linkable Notion pages for wikis, how-tos, FAQs, and decision logs. Supports six content types: decisions, how-to guides, FAQs, wiki entries, learning notes, and documentation pages, each with dedicated database templates and schemas. Includes a five-step workflow: define capture purpose, locate the target database, extract and structure content, create or update the Notion page, and link from related hub pages. Provides reference guides...
23 files — installable on its own
Prepare meeting materials with Notion context and Codex research; use when gathering context, drafting agendas/pre-reads, and tailoring materials to attendees.
24 files — installable on its own
Research Notion content and synthesize into structured briefs, reports, or comparisons with citations. Search and fetch Notion pages using targeted queries, then organize findings by theme with inline source citations and a references section Choose from four output formats (quick brief, research summary, comparison, comprehensive report) based on scope and user goal Create and update Notion pages using built-in templates; link sources directly and track changes as new information arrives...
28 files — installable on its own
Convert Notion specs into linked implementation plans, tasks, and progress tracking. Automates the workflow from spec discovery through task creation: search and fetch specs, parse requirements, generate implementation plans, and create tracked tasks in Notion. Includes templates for quick vs. full implementation plans, task creation patterns, and progress update cadences to match project scope. Provides reference guides for requirement extraction, task sizing (1–2 day chunks), dependency...
24 files — installable on its own
Access current OpenAI developer documentation with MCP tools and model-selection guidance. Prioritizes the OpenAI Developer Docs MCP server for all queries; falls back to web search only on official OpenAI domains if MCP returns no results Includes reference files for model selection, GPT-5.4 upgrade planning, and prompting guidance; always verifies recommendations against live docs before answering Covers seven core OpenAI products: Apps SDK, Responses API, Chat Completions, Codex, gpt-oss,...
14 files — installable on its own
PDF reading, creation, and validation with visual rendering and programmatic generation. Render PDF pages to PNG for visual inspection of layout, spacing, and typography before delivery using Poppler ( pdftoppm ) Generate PDFs programmatically with reportlab for reliable formatting; extract text and metadata with pdfplumber or pypdf Enforce quality standards: no clipped text, overlapping elements, broken tables, or rendering artifacts; ASCII hyphens only, human-readable citations Use...
6 files — installable on its own
Terminal-driven browser automation with element snapshots and interactive UI workflows. Operates via playwright-cli wrapper script (requires npx ); supports headless and headed modes for visual debugging Core workflow: open page, snapshot for stable element references, interact using refs, re-snapshot after navigation or DOM changes Includes form filling, clicking, typing, multi-tab management, screenshot/PDF capture, and trace recording for flow debugging Element refs (e.g., e3 , e15 )...
13 files — installable on its own
Persistent browser and Electron debugging with live session handles across code iterations. Maintains the same Playwright handles ( browser , context , page ) across multiple js_repl cells, eliminating restart overhead during iterative UI testing and QA. Supports desktop web (explicit viewport or native-window mode), mobile web emulation, and Electron app debugging from a single persistent session. Includes structured QA workflow: write a coverage inventory upfront, run functional tests with...
8 files — installable on its own
Create and scaffold plugin directories for Codex with a required `.codex-plugin/plugin.json`, optional plugin folders/files, and baseline placeholders you can…
8 files — installable on its own
Deploy applications to Render with Git-backed services or prebuilt Docker images via Blueprint or direct creation. Supports two deployment methods: Blueprint (Infrastructure-as-Code via render.yaml) for multi-service apps, and Direct Creation (MCP tools) for single services Analyzes codebases to determine runtime, build/start commands, environment variables, and required datastores (databases, Redis, cron jobs) Generates render.yaml configurations with automatic service type selection (web,...
24 files — installable on its own
OS-level desktop and window screenshot capture with flexible save locations and region selection. Supports full-screen, app/window, active-window, and pixel-region captures across macOS, Linux, and Windows Automatically routes to OS default screenshot location, user-specified paths, or temp directory based on context Includes platform-specific helpers (Python for macOS/Linux, PowerShell for Windows) with preflight permission checks on macOS to reduce repeated sandbox prompts Prioritizes...
14 files — installable on its own
Language and framework-specific security reviews with actionable vulnerability detection and fix guidance. Supports Python, JavaScript/TypeScript, and Go with framework-specific best practices loaded from a references directory Operates in three modes: secure-by-default code generation, passive vulnerability detection during development, and full security audit reports with severity prioritization Generates detailed markdown reports with line-number references, impact statements, and numeric...
15 files — installable on its own
Analyze git repositories to build a security ownership topology (people-to-file), compute bus factor and sensitive-code ownership, and export CSV/JSON for…
11 files — installable on its own
Repository-grounded threat modeling that maps trust boundaries, assets, and abuse paths to concrete code evidence. Enumerates entry points, data flows, and trust boundaries anchored to actual repository structure and configuration Derives realistic attacker goals tied to specific assets (credentials, PII, integrity-critical state, compute resources) rather than generic checklists Prioritizes threats using likelihood and impact reasoning, with explicit assumptions about deployment,...
7 files — installable on its own
Read-only Sentry API queries for production error inspection and health summaries. Requires SENTRY_AUTH_TOKEN with read-only scopes ( project:read , event:read , org:read ); supports optional org, project, and base URL environment variables Core commands: list unresolved issues, fetch issue details, retrieve events for an issue, and inspect individual event data with automatic PII redaction Defaults to 24-hour time range, production environment, and 20-result limit (max 50); supports...
7 files — installable on its own
Comprehensive guide for creating modular skills that extend Codex's specialized capabilities. Skills are self-contained folders with a required SKILL.md file and optional bundled resources (scripts, references, assets) that provide domain expertise, workflows, and tool integrations Follow progressive disclosure: metadata always in context, SKILL.md body loaded on trigger, bundled resources loaded as needed to minimize context bloat Six-step creation process: understand concrete examples,...
10 files — installable on its own
Install Codex skills from curated lists or custom GitHub repositories into your local skills directory. List available skills from the default curated collection or experimental/custom paths; installed skills are annotated in the output Install individual skills by name from the curated list, or install multiple skills from any GitHub repo (public or private) using repo paths or URLs Supports git sparse checkout as a fallback for authentication-protected repos; uses existing git credentials...
11 files — installable on its own
Text-to-speech generation for narration, voiceovers, IVR prompts, and accessibility reads via OpenAI Audio API. Supports single clips and batch processing; defaults to gpt-4o-mini-tts-2025-12-15 with built-in voices (cedar, marin, and others) Includes instruction augmentation for voice affect, tone, pacing, emotion, and emphasis; instructions supported only on GPT-4o mini TTS models Enforces 4096-character input limit per request and 50 requests/minute rate cap; splits longer text into...
20 files — installable on its own
Transcribe audio files to text with optional speaker diarization and known-speaker hints. Supports fast text transcription via gpt-4o-mini-transcribe and speaker-labeled diarization via gpt-4o-transcribe-diarize Accepts multiple audio formats and optional known-speaker references (up to 4 speakers) to improve diarization accuracy Outputs as plain text, JSON, or diarized JSON with configurable output directories to prevent overwrites Requires OPENAI_API_KEY environment variable; uses bundled...
11 files — installable on its own
Deploy applications to Vercel as preview or production environments. Supports both Vercel CLI and fallback script-based deployment methods, with automatic framework detection and packaging Always deploys as preview by default unless user explicitly requests production deployment Returns preview URL immediately and claim URL for fallback deployments to manage the deployment Requires 10-minute timeout for build completion; escalates network permissions only if sandbox blocks outbound...
9 files — installable on its own
Bootstrap, develop, and design modern WinUI 3 desktop applications with C# and the Windows App SDK using official Microsoft guidance, WinUI Gallery patterns,…
24 files — installable on its own
Use only when the user explicitly asks to stage, commit, push, and open a GitHub pull request in one flow using the GitHub CLI (`gh`).
7 files — installable on its own
[!IMPORTANT] This repository is deprecated. For current Codex skill and plugin examples, use the OpenAI Plugins repository. If you want to add your own skills to Codex, follow the Build plugins guide, which includes instructions for creating a skill-only plugin.
Agent Skills
Agent Skills are folders of instructions, scripts, and resources that AI agents can discover and use to perform at specific tasks. Write once, use everywhere.
Codex uses skills to help package capabilities that teams and individuals can use to complete specific tasks in a repeatable way. This repository catalogs skills for use and distribution with Codex.
Learn more:
Installing a skill
Skills in .system are automatically installed in the latest version of Codex.
To install curated or experimental skills, you can use the $skill-installer inside Codex.
Curated skills can be installed by name (defaults to skills/.curated):
$skill-installer gh-address-commentsFor experimental skills, specify the skill folder. For example:
$skill-installer install the create-plan skill from the .experimental folderOr provide the GitHub directory URL:
$skill-installer install https://github.com/openai/skills/tree/main/skills/.experimental/create-planAfter installing a skill, restart Codex to pick up new skills.
License
The license of an individual skill can be found directly inside the skill's directory inside the LICENSE.txt file.
Install the whole package (42 skills):
npx skills add https://github.com/openai/skillsOr install a single skill:
npx skills add https://github.com/openai/skills --skill <name>Pick the skill name from the Skills tab — each entry there installs independently.