Addresses review and issue comments on the open GitHub pull request for the current branch with the gh CLI.
By agent · Codex
Codex skills: 1008 worth installing
Skills in the open SKILL.md format that Codex reads. Each one has a plain-language summary and the command to install it.
Installing a skill in Codex
$ npx skills add <author>/<skill> --agent codexReplace the placeholder with the author and skill name from the skill's page. Start a new Codex session afterwards so the skill is picked up.
Skills for Codex
All skillsDebugs failing GitHub Actions checks on a pull request: inspects logs, summarises the failure and drafts a fix plan.
Creates and packages Codex-compatible animated pets and spritesheets from character art or visual references.
Creates and edits Jupyter notebooks for experiments, explorations and tutorials from bundled templates.
Reads, creates and updates issues, projects and team workflows in Linear.
Migrates instruction files, skills, agents and MCP config from other agents into Codex.
Deploys web projects to Netlify with the Netlify CLI, including preview and production deploys.
Captures conversations and decisions as structured Notion pages such as wiki entries, how-tos and FAQs.
Prepares meeting materials from Notion context: agendas, pre-reads and notes tailored to the attendees.
Researches across Notion pages and synthesises the findings into briefs, comparisons or reports with citations.
Turns a Notion spec into an implementation plan with tasks and progress tracking.
Looks up current official OpenAI documentation with citations, and helps choose a model for a use case.
Reads, creates and reviews PDF files where rendering and layout matter, checking pages visually.
Keeps a persistent browser or Electron session open for fast, iterative UI debugging.
Automates a real browser from the terminal with playwright-cli: navigation, forms, snapshots, screenshots and data extraction.
Deploys applications to Render by analysing the codebase and generating a render.yaml Blueprint.
Takes desktop or system screenshots of the full screen, a specific window or a pixel region.
Reviews code against security best practices for its language and framework, and suggests improvements.
Analyses a git repository to map who owns security-sensitive code and computes the bus factor.
Writes a threat model grounded in the repository: trust boundaries, assets, attacker capabilities and mitigations.
Inspects Sentry issues and events and summarises recent production errors through the Sentry CLI, read-only.
Generates text-to-speech narration and voiceovers with the OpenAI Audio API.
Transcribes audio files to text, with optional speaker diarisation.
Deploys applications and websites to Vercel, including preview deployments.
Questions
Codex and skills.
Do Claude skills work in Codex?
Often. A skill is a folder of instructions and scripts in the open SKILL.md format, which Codex reads. A skill that depends on tools only one agent provides will not carry over, so read the skill's source before you rely on it.
How do I install a skill in Codex?
Each skill page has an install command. Run it in a terminal, then start a new Codex session so the skill is picked up.
What is the difference between a skill and an instructions file such as AGENTS.md?
An instructions file is loaded in every session and describes the project. A skill is loaded only when a task calls for it, so it can be long and specific without costing context on every turn.