Back to directory

Agent Skills Repository

A curated bundle of sixteen installable skills for AI coding agents such as Claude Code, GitHub Copilot and Cursor, aimed at creative production tasks such as presentations, video editing and landing pages.

Agent Skills Repository

This repository packages sixteen skills as one install, including a film-and-presentation-production skill, an orchestrator that routes a request to the right specialist skill, a brand design system, a cinematic-presentation builder, a short-form video skill, a video editor, tools that turn a video into a parallax or sticky-hero landing page, motion-graphics compositions, an AI image skill and a diagramming skill. The install script also sets up supporting tools such as ffmpeg, Node.js, Python, a speech-transcription library and Python virtual environments, and symlinks the skills into the configuration folders of more than one agent host.

After installation, a user describes what they want in plain language, such as asking for a deck or a landing page built from a video, and the matching skill is triggered automatically; generated projects are saved into a fixed folder structure so they stay easy to find. It works on macOS, Linux and Windows, including through Git Bash or WSL.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.