The Agent Skills Directory

[SAFE]: The skill provides documentation and implementation examples for fine-tuning large language models using established industry libraries.
[EXTERNAL_DOWNLOADS]: Fetches machine learning models and datasets from Hugging Face Hub and official GitHub repositories. These operations are standard for the machine learning development workflow.
[COMMAND_EXECUTION]: Instructs the user to perform package installations via pip and compilation of specific acceleration libraries from source. These are documented steps for setting up a fine-tuning environment.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external datasets for training purposes. While this is the primary intended functionality, users should ensure the integrity of datasets used for fine-tuning as the skill does not implement specific data sanitization logic before feeding data to the training loop.

peft