The Agent Skills Directory

[SAFE] (SAFE): No malicious instructions, prompt injections, or obfuscation were found in the skill metadata (SKILL.md) or scripts.
[REMOTE_CODE_EXECUTION] (LOW): The script scripts/track_mlflow.py utilizes mlflow.pyfunc.load_model, which involves model deserialization. This is standard behavior for MLOps tools and is considered acceptable within the context of managing trusted machine learning models.
[DATA_EXPOSURE] (SAFE): No hardcoded secrets, sensitive file paths, or unauthorized network calls were detected. The script uses configurable MLflow tracking URIs and standard artifact logging protocols.
[COMMAND_EXECUTION] (SAFE): No arbitrary command execution, subprocess spawning, or shell access was identified.

mlops-engineer