model-checkpoint-evaluator

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled Python script named benchmark_calculator.py to process model checkpoints and calculate performance metrics.
  • [EXTERNAL_DOWNLOADS]: The skill references the huggingface_hub package as a dependency to facilitate interaction with the Hugging Face Hub for model deployment.
  • [CREDENTIALS_UNSAFE]: The skill is instructed to read an authentication token from a local file named hf_token.txt. This token is used to authenticate with the Hugging Face Hub, a well-known service, for legitimate repository management and model uploads.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 06:21 AM
Security Audit — agent-trust-hub — model-checkpoint-evaluator