The official Hugging Face skills repository that gives AI coding agents like Claude Code, Codex, and Cursor instant access to the entire ML ecosystem - from dataset creation to model training.

Ever wished your AI coding assistant actually understood machine learning workflows? This is Hugging Face’s answer - a curated collection of skills that turn generic coding agents into ML-savvy powerhouses. Instead of fumbling through documentation or writing boilerplate from scratch, your agent gets pre-packaged expertise for dataset creation, model training, evaluation, and deployment.

What makes this special is the universal compatibility. Whether you’re using Claude Code, OpenAI Codex, Google’s Gemini CLI, or Cursor, these skills work seamlessly across platforms using the standardized Agent Skills format. Each skill is a self-contained folder with instructions, scripts, and resources that your agent can immediately understand and execute. Think of it as giving your coding assistant a PhD in machine learning overnight.

With 9,500+ stars and backing from the Hugging Face team, this isn’t just another tool collection - it’s becoming the standard way to bridge AI agents and ML workflows. Installation is dead simple across all major platforms, and since it’s from Hugging Face, you know it stays current with the latest ML practices and models.

⭐ Stars: 9538
💻 Language: Python
🔗 Repository: huggingface/skills