Skills, prompts, AGENTS.md files and MCP servers that Codex reads. Skills are shared with Claude Code; the rest is Codex-native. 267,937 indexed, measured the same four ways as everything else here.
Public registry of code-knowledge graphs for AI agents with schema validation and drift detection. Runs locally from the @looptech-ai/understand-quickly-mcp npm package.
★not rated 48 1mo agoA
tokens not measured
originalApache-2.0
A lightweight arXiv literature digest skill for OpenClaw, with Zotero-driven interest profiling, research-map plus Zotero-semantic ranking, agent-enriched digest cards, and non-destructive feedback writeback.
Coordinate Codex with Claude Code in real time for collaborative software work. Use when the user wants Codex to plan or review while Claude Sonnet or Opus implements, when creative/frontend/product work should be delegated to Claude, when a Codex-to-Claude handoff loop is requested, or when phrases like "notify…
Complete guide for building beautiful apps with Expo Router. Covers fundamentals, styling, components, navigation, animations, patterns, and native tabs.
Walk a client through a Responsible AI assessment questionnaire and score their answers 0-5 across eight RAI dimensions (Governance, Privacy & Security, Safety, Veracity & Robustness, Controllability, Fairness, Explainability, Transparency). Use when a user asks to run a Responsible AI assessment, RAI scorecard…
A Chinese-language product-management adviser built around named product thinkers and decision frameworks. It guides discussions about requirements, roadmaps, competitor analysis, MVPs, and product decisions.
Personal assistant for the [YOUR LAST NAME] family. Use this skill when: Filling out forms (medical, insurance, school, government, travel) Making appointments or reservations Handling insurance claims or benefits questions Coordinating with service providers Travel planning and bookings Financial tasks (account info…
Build, review, and improve programmatic UIKit applications following proven patterns for Auto Layout, table views, collection views, navigation, animation, Core Data, networking, and architecture. Use when writing new UIKit features, refactoring existing view controllers, reviewing UIKit code quality, building custom…
Install, customize, launch, hot-switch, pause, inspect, verify, package, repair, recover, update, or restore a real reversible skin for the official Codex Desktop app on macOS or Windows. Use when a user asks to apply a Codex Dream Skin, turn a local or CDN image into a Codex theme, manage saved themes or the macOS…
A Chinese-language content creation workflow for OpenClaw, an automation platform. It covers seven stages from finding current topics through writing, quality review, and publishing.
Use when the user asks to write, draft, optimize, review, package, or generate Chinese invention patent materials, Chinese patents, invention patents, patent specifications, claims, abstracts, embodiments, abstract drawings, final Word/DOCX patent documents, editable equations, or complete patent application files…
Use when building, changing, or deploying an instancez backend - editing instancez.yaml (tables, RLS, auth, storage, rpc, functions), running the inz CLI, or debugging why a query is denied. instancez is a single-binary Supabase-compatible backend defined by one YAML file.
Use this skill when the user asks to create, edit, outline, storyboard, review, or improve presentations, slide decks, pitch decks, lecture decks, executive decks, or PPTX-oriented content.
Query the local tuneloop store of AI agent sessions with read-only SQL. Use when the user wants any metric, slice, or aggregation of their tuneloop data — cost, tokens, tool usage, or artifacts.
A tool for downloading subtitles from Bilibili, a Chinese video platform, splitting them into smaller text files, and preparing them for language-model summaries.
AI-powered multi-agent code review. Simulates a team of Principal Engineers reviewing code from different perspectives. Use when asked to review code, check a PR, analyze changes, or perform code review.
★not rated 48 5mo agoA44 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: