Skills for the Gemma and model/agent interactions
gemma-skills is a collection of skills for coding assistants that work with Gemma models and model-agent interactions. It supports building applications with Gemma and training or fine-tuning Gemma models through methods such as supervised fine-tuning, DPO, RLHF, and reward modeling. The catalogue entries are the project's skills for these workflows.
Skill Claude CodeCodex
Trigger this skill when building applications with Gemma or for general knowledge inquiries related to Gemma models (e.g. prompt structure, capabilities). Covers model selection, development workflows, and deployment best practices.
Skill Claude CodeCodex
Trigger this skill when the user wants to train, fine-tune, or adapt Gemma models (e.g. SFT, DPO, RLHF, Reward Modeling) on local hardware. Covers TRL, Unsloth, dataset preparation, validation, and GGUF/LiteRT conversion.