# k3-agentic-skills

A collection of curated AI agent skills for coding assistants, paired with a local semantic search engine over those skills powered by the Gemma embedding model, with no API keys or cloud dependency.

k3-agentic-skills packages dozens of curated skill files that document agent frameworks and workflows, such as Google's ADK, Gemini embeddings, and VPS-management patterns, for use in IDEs and agents that support skill-style instructions, including Antigravity, Cursor, Windsurf and Claude Code.

A companion local semantic search tool, mrl-indexer, builds a vector index over the skills and documentation using the EmbeddingGemma-300M model so an agent can find the right skill by natural-language query, entirely on the user's own machine. A separate guide covers running the Gemma 4 26B model locally for free inference.

Website: https://github.com/Fandry96/k3-agentic-skills