Skills, prompts, AGENTS.md files and MCP servers that Codex reads. Skills are shared with Claude Code; the rest is Codex-native. 281,785 indexed, measured the same four ways as everything else here.
Provider-neutral, Windows-first Beta workflow for inspecting, preparing, and reviewing bounded frontend visual handoffs with optional external coding clients and a Codex-only path. Use only when the user explicitly invokes $frontend-visual-handoff; default to mode=inspect, require a separate second invocation with…
Use when analyzing, comparing, reviewing, or tearing down competitors, products, apps, mini programs, SaaS, platforms, devices, or feature areas from links, recordings, screenshots, files, reviews, or hands-on evidence; also use for positioning, parity, differentiation, battlecards, threats, and MVP priority decisions.
A writing workflow for creating and managing novels, including chapter files, character notes, outlines, and reader feedback. Git, a system for tracking file changes, is used to manage the project.
A production workflow for turning approved content into editable and exportable media such as technical diagrams, GIFs, animated web pages, videos, audio, and podcasts.
Turn raw client conversations, meeting transcripts, delivery notes, feedback, and business records into an evidence-backed business memory for solo service providers and small professional teams. Use when Codex needs to organize real operating material, extract verifiable facts and business events, update a client…
Local file operations on the user's machine: format conversion (image, PDF, Excel, DOCX, HTML, Markdown), file inspection (metadata, dimensions, page counts, sheet names), zip archive creation and extraction, and text extraction from PDF, DOCX, XLSX, HTML, Markdown, and image EXIF. Use when the user asks to convert…
Route Apple/iOS agent work between Argent MCP and XcodeBuildMCP (and when to combine both in one session). Use when choosing tooling for build, compile, test, run, install, launch, simulator control, UI automation, screenshots, QA, LLDB debug, logs, permissions, profiling, physical device, macOS, Android, Chromium…
Lets the agent work with GitHub: search code, read and create issues and pull requests, and manage repositories through the GitHub API. Runs locally from the @modelcontextprotocol/server-github npm package. Needs 1 environment variable to run.
A local search and reading service for papers on arXiv, a public website where researchers share scientific papers. It can find papers and parse their contents.
★not rated 24 1y agoA
tokens not measured
originalMIT
Multi-phase code-quality skill — primary frame for the evaluator role in Execute, the architecture axis in Plan, and correctness/readability in Refine. Multi-axis code review with severity vocabulary. Use when acting as the evaluator assessing a generator's output, and when reviewing any change before signalling…
A single search tool for past conversations held by local coding agents such as Claude Code and Codex. It can also help trace which agent tool or command produced a file.
Working-style skill that makes Opus (or Sonnet) agents work like a Fable agent. BINDING FROM THE MOMENT YOU LOAD THIS - visible top-level messages are short status summaries only (conclusion + next step, ≤2 lines); ALL deliberation goes in reasoning or a thoughts file via tool call, never visible text. Top-level agent…
Use when calling any OneTool pack tool via onetool/ot triggers or the MCP run tool — pack map with aliases, call syntax, kwarg-prefix and alias forgiveness, discovery, a greppable index of every command, recovery from disconnected servers, and large-result handling.
Direct access to Google Search Console, Google Analytics 4, and Bing Webmaster Tools data via OAuth/API key. Use whenever the user asks SEO, search performance, organic traffic, ranking, impressions, clicks, CTR, conversions, traffic source, landing page, or analytics questions and wants real numbers from their own…
Drive the user's real, logged-in browser from the shell via the reins CLI. Because it's their actual browser, every site is already authenticated — so you can scrape behind logins, read cookies/tokens/localStorage, watch and replay live API traffic, call a site's own API as the signed-in user, and…
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: