Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
faizkhairi/claude-code-blueprint
Skill Claude Code
Scan git diffs for project-specific anti-patterns. Triggers on: 'scan diff', 'check diff', 'anti-pattern check', 'pattern scan', 'review changes'.
Skill Claude Code
Load the latest model checkpoint, run evaluation on the test set, and generate a metrics report with confusion matrix. Use this after training to assess model performance or to re-evaluate a specific checkpoint.
wpacademy/wordpress-dev-skills
Skill Claude CodeCodex
Develop WordPress themes following official WordPress coding standards, theme review requirements, accessibility standards (WCAG 2.1 AA), and modern web development best practices. Use this skill whenever the user wants to create, scaffold, or develop a WordPress theme — including block (FSE) themes, classic themes…
Skill Claude CodeCodex
Use when the user wants a specific weather model, needs ensemble uncertainty ranges, requests seasonal outlooks, or asks for long-term climate projections.
Skill Claude CodeCodex
Use when the user asks Codex, Claude Code, ChatGPT-style coding agents, or another AI agent to build, inspect, run, debug, or revise COMSOL Multiphysics / COMSOL Desktop models. Choose the simplest real COMSOL control path for the task: saved .mph inspection, local COMSOL documentation, direct COMSOL executables…
Skill Claude Code
Query Lighter DEX market data on Lighter and Robinhood Lighter (orderbooks, trades, candles, funding rates) and execute trades (limit and market orders, modify, cancel, leverage, margin, withdraw, spot and perp transfers) on perpetual and spot markets. Use when the user mentions Lighter, zkLighter, perps, perpetual…
Skill Claude CodeCodex
An open-source security gateway that sits between an AI application and an LLM service. It can remove or clean sensitive information from requests and responses before they are passed through.
Skill Claude CodeCodex
A Chinese-language guide for checking and recovering X accounts affected by rate limits, warnings, or other enforcement actions. X is the social network formerly known as Twitter.
jiayushi1-ux/script-to-shot-engine
Skill Codex
A guide for turning existing character, scene, weapon, and prop artwork into action or confrontation video prompts. It organizes fights, chases, negotiations, and similar scenes as clear chains of cause and effect.
Skill Claude CodeCodex
Translate English PDF documents (especially academic papers) into Chinese using PDFMathTranslate-next (pdf2zh), producing a monolingual Chinese PDF and a bilingual side-by-side PDF with formulas, figures, and layout preserved. Automatically installs the pdf2zhnext environment via uv and reuses the model/API the user…
Skill Claude CodeCodex
Use this skill when building, enriching, auditing, or uploading B2B paid ads audiences to Google Customer Match, Meta/Facebook Custom Audiences, or LinkedIn Matched Audiences. Triggers on phrases like '/deepline-ads-audience', '/deepline-ads-audiences', 'upload this audience', 'create custom audiences', 'personal…
Skill Claude CodeCodex
End-to-end short-form video cloning + editing pipeline. Downloads influencer video, detects scenes (talking-head vs B-roll), generates HeyGen avatar with voice clone, overlays avatar onto talking-head segments with varied zoom levels, word-syncs audio to match original caption timing, preserves B-roll and sound…
Skill Claude CodeCodex
Operate Conway Cloud from AI agents, with Automaton-first defaults and HTTP fallback.
Skill Codex
A review workflow for checking and adapting content before publishing it on Douyin, Xiaohongshu, Weixin Video Accounts, or Kuaishou.
Skill Claude CodeCodex
Sigillo is a self-hostable open-source alternative to Doppler. Use when working with sigillo run, sigillo setup, sigillo login, managing secrets, projects, or environments. Also load when integrating Sigillo into CI, Cloudflare Workers, Docker, Vercel, or any other deployment target.
Skill Codex
A business and financial analysis skill with multiple data sources and several analysis modes. It supports reports, competitor comparisons, industry models, macroeconomic research, and scenario analysis.
NVIDIA-AI-IOT/jetson-bsp-skills
Skill Claude CodeCodex
Bootstrap a custom carrier board by forking carrier files and scaffolding a DT overlay from the reference devkit. Use after jetson-init-source; not for module-level or kernel-DTB changes.
liangdabiao/deepseek-v4-flash-vision-video-rag
Skill Claude CodeCodex
A video-understanding and question-answering tool based on a DeepSeek vision model. It can examine videos such as recordings, surveillance footage, games, films, or animations and refer to events with timestamps.
Skill Claude CodeCodex
Write Harbor task verifiers using Reward Kit. Use when creating or editing a task's tests/ directory, adding grading criteria, setting up LLM/agent judges, or designing verifiers that produce a reward score.
quantskills/skill-quant-factor-skill-factory
Skill Codex
Use when converting OHLCV alpha ideas into QuantSkills organization factor Skills, batch-generating non-duplicate framework-neutral quant factor Skill folders, validating them on cached real market data such as AkShare A-share and Yahoo US data, and writing factor evaluation reports.
Skill Claude Code
Natural-language pinrule rule input — handles two paths. (A) Single rule add/modify/remove from user's plain description. (B) Scenario rule pack generation — when user names a domain ("I mainly do research / writing / legal / data analysis"), synthesize 5-7 tailored rules by combining their existing local rule files…
Skill Claude CodeCodex
Boundary-mapping mode for invasive changes, new interfaces, data contracts, and risky tradeoffs.
Skill Claude CodeCodex
A routing guide for controlling application interfaces through the project’s DCC-CUA system. DCC applications are digital-content tools such as Maya, Blender, and Houdini.
Skill Claude Code
Read from and write to the shared Hive Mind corpus (the hv CLI) — append-only institutional memory shared across agents (Hermes, Claude Code, …) and machines. Use to record decisions/outcomes/constraints and to check what the hive already knows before acting — especially on errors, uncertainty, or a named…
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: