evaluate-ml-pipeline
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill orchestrates its workflow by calling a custom command-line interface (
python -m skore_skills). These commands are used to manage project status, interface with API metadata, and handle site generation. - [DYNAMIC_EXECUTION]: The skill implements a results materialization process by generating a Python script (
materialize.py) from a local template and executing it. This script handles theskore.evaluatecall and persists the resulting reports to the project's workspace. - [DATA_EXPOSURE]: The skill facilitates the transmission of machine learning reports to remote backends (Skore Hub or MLflow). This includes managing environment-specific configurations like tracking URIs and Hub workspace identifiers, which is consistent with the skill's stated purpose of evaluation persistence.
Audit Metadata