Agent skills for Codex CLI

Grill the user relentlessly about a plan, decision, or idea.

Create and edit images using OpenAI image generation API with GPT image models (gpt-image-1, gpt-image-1-mini, gpt-image-2 and later).

Build a presentation as markdown in two steps — an outline (deck-outline.md), then the full deck with speaker notes (deck.md) — plus an…

Create, refine, or retire agent skills. Use when the user has a repeatable workflow worth capturing as a skill, wants to improve or fix an…

Find open Singapore government tenders for any sector in real time using parallel TinyFish agents scraping multiple government tender…

Invoke only when the user explicitly asks to review code through the "single level of abstraction / layered error handling" lens — a…

Use to lock the single-mechanism arc before drafting — converts a set of correct experiments into one Molecular Cell story that runs…

Pattern for functions that accept either a single item or an array.

Draft an implementation plan (feature, fix, refactor) or findings writeup and save it under `.agents/scratch/plans/` — a single…

Integrate Siri voice interactions, Shortcuts, and intelligent suggestions into iOS/watchOS apps using Apple's SiriKit framework.

Pre-launch checklist for shipping a new website or web app.

Builds site selection and cannibalization analysis workflows in CARTO.

Suede Labs workflow for turning a website, PWA, dashboard, or marketplace into an iOS app.

Analyze existing XML sitemaps or generate new ones with industry templates.

Use this skill when the user wants to generate sitemap xml from a url.

Read and synthesize one or many academic papers into six connected sections: problem background, motivation, core idea…

Use this skill for ANY browser automation task — testing web pages, filling forms, taking screenshots, automating job applications…

Review changed code for reuse, quality, and efficiency, then fix any issues found.

Transforms articles into animated hand-drawn whiteboard diagram B-roll videos using AI video models like Seedance 2.0.

Reverse-engineer a provided video into a business-facing Chinese A3 landscape prompt PDF with extracted frame images, global setting…