The best Codex skills in 2026: top picks, install commands and plugins

Codex has had skills since the end of 2025, and the open Agent Skills format means most of the good ones were written for several agents at once. That makes the list short and the choice easier: the skills worth installing in Codex are, by and large, the same ones developers already use elsewhere, plus a handful that only make sense with OpenAI's plugins.

This page ranks them, shows the Codex install command for each, and explains the parts of Codex's skill system that trip people up. If you use Claude as well, the companion list is the best Claude skills.

Repositories, star counts, licences and install commands were checked on GitHub and in OpenAI's documentation on 21 October 2026. Codex CLI was at version 0.163.

The top 10 Codex skills at a glance

Ranked by GitHub stars, which measure attention, not quality. Every one of these states Codex support in its README.

Two more that matter even though they are not single community repos: Anthropic's document skills (pdf, docx, xlsx, pptx, frontend-design) and OpenAI's own plugins marketplace. Both are covered below.

How Codex skills work

A Codex skill is a folder with a SKILL.md file, the same format Claude, Copilot and other agents use under the Agent Skills standard. The frontmatter needs only two fields:

---
name: release-notes
description: Write release notes from merged pull requests. Use when asked for a changelog or release notes.
---

1. List the pull requests merged since the last tag...

Optional folders sit next to it: scripts/ for code the skill runs, references/ for documents it reads, assets/ for templates, and agents/openai.yaml for Codex-specific settings.

Codex loads skills lazily. At startup it reads only each skill's name, description and path. The full SKILL.md is read when a task needs it. The startup list is capped at 2 % of the context window; with many skills installed, Codex first shortens descriptions and then drops skills. skills.max_context_tokens raises the cap, but the better fix is fewer, sharper skills.

Where skills live:

~/.codex/skills/ is still read but is deprecated; move skills to ~/.agents/skills/. Symlinked folders work, and two skills with the same name are both shown rather than merged.

Calling a skill. Codex picks one on its own when your task matches its description. To call one explicitly, type $ and the name ($skill-creator), or run /skills to browse. In ChatGPT the same skills are called with @.

How to install Codex skills

There are three routes, from quickest to most controlled.

1. $skill-installer inside Codex. Give it a curated skill name or a GitHub folder URL:

$skill-installer gh-fix-ci
$skill-installer install https://github.com/anthropics/skills/tree/main/skills/pdf

2. npx skills from a terminal. The cross-agent installer from vercel-labs/skills (~36k stars) installs for Codex and other agents at once:

npx skills add blader/humanizer --global --agent codex

3. Copy the folder. Put it in ~/.agents/skills/ for yourself, or commit it to .agents/skills/ in a repository so everyone who clones the repo gets it. This is the route we recommend for teams, because the skill is reviewed in a pull request like any other code.

Codex notices new and changed skills automatically. If one does not show up, restart.

The best Codex skills for coding

superpowers is the most used skill set on GitHub and it is now in OpenAI's official plugin marketplace. It makes the agent brainstorm before building, write a plan, write tests first and debug methodically instead of guessing. Install: run /plugins, search for superpowers, install, start a new session.

planning-with-files keeps task_plan.md, findings.md and progress.md on disk, with hooks that remind the agent to update them. Long tasks stop losing the thread when the context fills up. It ships as a native Codex plugin, or copy .agents/skills/planning-with-files and the .codex hook files into a repository (the repo's docs/codex.md has the steps).

caveman makes answers short and leaves code untouched, which cuts output tokens noticeably on long sessions. npx skills add JuliusBrussee/caveman -g, then $caveman.

OpenAI's own developer skills are worth a look for GitHub-centred teams: gh-fix-ci reads failing CI and proposes a fix, gh-address-comments works through review comments, and security-threat-model produces a threat model for a codebase. Install by name with $skill-installer. For a dedicated security pass, the codex-security plugin in the official marketplace goes further.

Pick one methodology collection, not three. superpowers and similar sets give competing instructions about how to plan and test, and the agent ends up following neither well.

The best Codex skills for design and front end

ui-ux-pro-max gives the agent a library of styles, palettes, font pairings and UX rules to pick from, instead of defaulting to the same purple gradient. Install:

npm install -g ui-ux-pro-max-cli
uipro init --ai codex

taste-skill is narrower and more opinionated: art direction that pushes interfaces away from the generic AI look. Install as a plugin:

codex plugin marketplace add Leonxlnx/taste-skill

then install taste-skill from /plugins.

Use one of these, not both. For connecting design files, the Figma plugin in OpenAI's marketplace lets Codex read frames and components directly.

The best Codex skills for documents and slides

Anthropic's document skills (pdf, docx, xlsx, pptx) are the most capable way to read and write Office files and PDFs with an agent, and they work in Codex. Install one by URL with $skill-installer, or copy the folder from anthropics/skills into ~/.agents/skills/. Note the licence: the document skills are source-available, not open source, and governed by Anthropic's terms. Fine for your own use; do not redistribute them.

ppt-master turns a document or a topic into a native, editable .pptx, not an image of slides. It needs Python 3.10+:

npx skills add hugohe3/ppt-master
pip install -r requirements.txt

hyperframes from HeyGen writes HTML and renders it to video, with 21 skills for scenes, captions and motion. It is in OpenAI's official marketplace, or npx skills add heygen-com/hyperframes.

The best Codex skills for research, writing and marketing

last30days researches a topic across Reddit, X, YouTube, Hacker News and the web, limited to the last 30 days, and writes a brief with sources. It is a native Codex plugin, or npx skills add mvanhorn/last30days-skill -g. Because it fetches the open web, it is exactly the kind of skill that can pick up prompt injection; keep the default sandbox on.

humanizer strips the patterns that make text read as AI-written: inflated vocabulary, rule-of-three lists, hollow transitions. npx skills add blader/humanizer --global --agent codex.

marketingskills is a set of skills for conversion optimisation, copywriting, SEO and analytics. It ships as a plugin marketplace:

codex plugin marketplace add coreyhaines31/marketingskills

then install marketing-skills from /plugins. Pair it with MCP servers for the tools it talks about; our best MCP servers for marketing and SEO lists them.

Codex plugins vs skills

A skill teaches Codex one task. A plugin is a package that can hold several skills plus MCP servers, hooks (commands that run at points in the agent's lifecycle) and browser extensions, distributed through a marketplace.

The official marketplace is openai/plugins, with around 65 plugins including Figma, Notion, GitHub, Linear, Sentry, Stripe, Vercel, Supabase, PostHog, Datadog, superpowers and hyperframes. It is the default in /plugins. The older openai/skills repository is now marked deprecated and points there; its curated skills still install by name, but new work goes into plugins.

Codex also reads Claude-style plugin manifests (.claude-plugin/marketplace.json) and sets CLAUDE_PLUGIN_ROOT for their hooks, so many Claude Code plugins install in Codex without changes.

Do Claude skills work in Codex?

Mostly, unchanged. Both follow the Agent Skills standard, and .agents/skills/ is Codex's native location. What to watch for:

• Codex reads only name and description. Claude-specific frontmatter such as allowed-tools, model, disable-model-invocation and context is silently ignored. In particular, allowed-tools pre-approves nothing in Codex; approvals follow Codex's own sandbox.
• Invocation changes. A skill called with /name in Claude Code is $name in Codex.
• Tool names in the instructions. A skill that says "use the Read tool" or relies on ${CLAUDE_PLUGIN_ROOT} paths may need a line reworded.
• Scripts run in Codex's sandbox. In the default workspace-write mode the network is restricted, so skills that call APIs will ask for approval. Python and Node dependencies are yours to install.
• A missing description fails to load. Codex flattens name and description to one line; keep them short and specific.

If you are deciding between the two agents in the first place, see Claude Code vs Codex.

AGENTS.md vs skills

Both give Codex instructions, but they load differently.

• AGENTS.md is read at the start of every run: a global file in ~/.codex/, then one per directory from the repository root down to where you are, with nearer files taking precedence, capped at 32 KiB combined. Use it for facts that are always true: build and test commands, conventions, things never to do.
• Skills cost almost nothing until used. Use them for procedures that only matter sometimes: writing release notes, fixing CI, generating a deck.
• Plugins are for distribution: when a team should get the same skills, tools and hooks.

A useful test: if you would want the instruction in the agent's head on every single task, it belongs in AGENTS.md. Otherwise it is a skill.

Configuring and turning off skills

To disable a skill without deleting it, add this to ~/.codex/config.toml and restart:

[[skills.config]]
path = "/Users/you/.agents/skills/some-skill/SKILL.md"
enabled = false

To keep a skill but stop Codex from choosing it on its own, add agents/openai.yaml to the skill folder:

policy:
  allow_implicit_invocation: false

It then runs only when called with $. This is the right setting for skills that deploy, delete or send anything.

Are Codex skills safe?

A skill runs with the agent's permissions, and OpenAI publishes no specific guidance on vetting third-party skills. What Codex does give you:

• Sandbox modes: read-only, workspace-write (the default) and danger-full-access. Keep the default; --yolo disables both sandbox and approvals.
• Protected paths: in workspace-write, .git, .agents, .codex and .aws stay read-only, so an agent cannot rewrite its own skills or credentials.
• Destructive tools from MCP servers always require approval.
• Plugin hooks must be reviewed and trusted before they run; check them with /hooks, or turn hooks off with hooks = false under [features].

The checklist we use: read SKILL.md and every script before installing, prefer the official marketplace, pin a commit or release, set allow_implicit_invocation: false on anything that acts outside the repository, and keep skills that read the open web away from repositories with secrets. Our skills guide has the full review checklist; it applies to Codex unchanged.

Where to find more Codex skills

• openai/plugins: the official marketplace (~7.5k stars).
• VoltAgent/awesome-agent-skills: 1,000+ cross-agent skills including Codex (~36k stars, actively maintained).
• awesome-codex-skills: the Codex-specific list (~17k stars), but not updated since July 2026.
• skills.sh: the registry behind npx skills.

Skills for a whole team, across agents

The useful thing about the Agent Skills standard is that a team does not have to choose: the same reviewed skills can run in Codex, Claude Code and ChatGPT. The hard part is keeping one approved set, with the same MCP connections and the same rules, when people use different agents. Walma AI Hub gives every employee Claude, GPT and other models through one gateway in your own EU region, with shared skills and MCP connections, per-team model access, budgets and a full audit log. See how it works, or book a 20-minute walkthrough below.