The Agent Skills Directory

[SAFE]: No security issues detected. The skill consists of legitimate Python templates and educational documentation for a popular open-source deep learning framework. The code follows established best practices for machine learning development.
[EXTERNAL_DOWNLOADS]: The skill references and provides instructions for integrating with well-known and trusted technology services such as Weights & Biases, MLflow, Comet, Neptune, and TensorBoard. These are standard tools in the data science ecosystem and are documented neutrally for their intended purpose.
[COMMAND_EXECUTION]: The Python scripts are structured as boilerplate templates for model training and data handling. They do not contain any unsafe subprocess calls, shell execution, or privilege escalation attempts.
[CREDENTIALS_UNSAFE]: The documentation and scripts correctly use placeholders (e.g., "YOUR_API_KEY_HERE") for sensitive integration tokens, following safe configuration practices.
[PROMPT_INJECTION]: No prompt injection patterns or attempts to override agent behavior were found in the instructional content.

pytorch-lightning