The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation provides installation commands for standard, well-known machine learning libraries including torch, transformers, datasets, evaluate, and accelerate via the uv package manager.
[CREDENTIALS_UNSAFE]: The documentation references the use of a HUGGINGFACE_TOKEN for authentication but correctly uses a placeholder string ('your_token_here') rather than hardcoding any sensitive credentials.
[REMOTE_CODE_EXECUTION]: No suspicious remote script execution patterns (such as 'curl | bash') or dynamic execution of untrusted input were found. Model loading and inference are performed using standard library APIs from the well-known Hugging Face Hub.
[PROMPT_INJECTION]: The skill instructions do not contain any patterns attempting to override system prompts, bypass safety filters, or extract underlying model instructions.
[DATA_EXFILTRATION]: No patterns indicative of sensitive file access or unauthorized data transmission to external domains were detected.

transformers