huggingface-best

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses the Hugging Face authentication token stored at ~/.cache/huggingface/token to authenticate API requests. This is standard behavior for Hugging Face integration and the token is sent only to official huggingface.co API endpoints.
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to huggingface.co to fetch benchmark data, leaderboard rankings, and model metadata. These are directed at a well-known, trusted service for the skill's primary purpose of model comparison.
  • [COMMAND_EXECUTION]: The skill uses curl, jq, and the hf CLI to interact with the Hugging Face Hub. These commands are used for data retrieval and metadata parsing and do not involve arbitrary command injection or unauthorized system changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 06:27 AM