Skills, prompts, AGENTS.md files and MCP servers that Codex reads. Skills are shared with Claude Code; the rest is Codex-native. 281,933 indexed, measured the same four ways as everything else here.
Evaluate Aquifer against your own workload. A shadow pilot on a branch, run for a few hours to a day against a replica or staging, measuring saturation, errors, latency, fairness and cost. Use it when someone asks whether Aquifer fits their system, wants to pilot it, or wants numbers before adopting it.
Decompose an evolved harness's score gain S(E) − S(B) on ONE evaluation set into Overfitting (O), Test-Time Scaling (T), and a residual upper bound on Generalizable Improvement (G). Build two controlled variants — a compute-matched baseline Bcc and a neutralized Eneutral — and re-score them on the same set.…
Find or install the anet command line and set up the local AgentNetwork (anet) node by following the setup guide of the installed anet. Use when the anet tools are missing, when the user asks to install, set up, update or repair anet or AgentNetwork, or when anet reports daemonnotrunning.
A research-paper reading guide for understanding one or more academic papers in depth. It provides explanations linked to the source, discusses equations and experiments, and connects methods to new projects.
Access atypica.ai's AI-powered business research capabilities through MCP protocol. Use when users need to conduct consumer research, market analysis, product testing, or generate insights about business questions. Supports creating research sessions, sending messages with AI execution, handling user interactions…
Associative memory for LLM agents: Graph RAG over a folder of markdown notes, in one SQLite file, with no server to run. Runs locally from the sqlite-graph-memory Python package. Needs 2 environment variables to run.
Create a new Fluence.Wpf consumer WPF app, a NavigationView MVVM shell, or a Windows Settings-inspired shell from the installed Fluence consumer templates.
A research assistant workflow that turns a research idea into literature searches, experiment plans, figures, paper summaries, and organised data. It asks you to confirm important choices instead of making them on its own.
Read the user's WHOOP recovery, HRV, sleep and strain through the local whoop MCP server. Use when the user asks about WHOOP data, recovery score, whether to train hard or rest today, sleep performance, strain, HRV trends, or a daily or weekly recovery brief. Starts with connection status, falls back to synthetic demo…
This skill should be used when the user asks to research, ascertain, or verify foreign or overseas laws — including "foreign law ascertainment", "ascertain the law of [country]", "overseas law research", "comparative law research", "what does [country] law say about…", "is [statute/case] still in force", point-in-time…
Maintained, agent-readable knowledge bases with calibrated search and provenance. Pro key adds XL. Runs locally from the agentwikis-mcp npm package. Needs 2 environment variables to run.
★not rated 15 yesterdayASocket: warn
tokens not measured
originalMIT
App Store Connect MCP server — manage iOS/macOS apps, TestFlight, in-app subscriptions, and App Store metadata from Cursor, Claude, or any Model Context Protocol client. Uses official App Store Connect API. Runs locally from the mcp-asc npm package. Needs 4 environment variables to run.
Create video compositions, animations, title cards, overlays, captions, voiceovers, audio-reactive visuals, and scene transitions in HyperFrames HTML. Use when asked to build any HTML-based video content, add captions or subtitles synced to audio, generate text-to-speech narration, create audio-reactive animation…
Convert Android XML layouts to Jetpack Compose. Use when asked to migrate XML layouts, convert views to composables, or help with Compose migration. Handles layouts, widgets, attributes, styles, and resource references.
A skill for sending proactive messages to WeChat users through the wxclawbot command-line tool. It can send text, images, videos, and files, even though the bot normally only replies after a user writes first.
Set visual and interaction direction for any UI surface (web, React, TUI, CLI, desktop, Qt, design-system tokens) before any UI code. Direction-first: generates 3-4 distinct directions via verbalized sampling, picks one via per-axis single-select, then derives palette, typography, spacing, motion budget. Loads when…
A project-initialisation skill that creates a lightweight structure of guides for AI coding agents. It maps project areas and adds sections such as contents, locations, commands, working methods, and cautions.
MCP server "dt-mcp" as configured in istefox/istefox-dt-mcp. Launched with https://github.com/istefox/istefox-dt-mcp/releases/download/v0.5.2/istefox-dt-mc. Needs 4 environment variables to run.
★not rated 14 5mo agoA
tokens not measured
originalMIT
Search for music tracks using natural language prompts or video content via the Harmix AI API. Use when the user asks to find music, search for songs, discover tracks, needs music recommendations based on mood/genre/tempo/scene, or wants to find music for a video, video soundtrack, or music that matches video content.
★not rated 14 7mo agoA72 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: