querying-mlflow-metrics

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of a local script, scripts/fetch_metrics.py, to retrieve and aggregate metrics from MLflow tracking servers. The script is invoked via python with arguments for server configuration and metric selection.\n- [PROMPT_INJECTION]: The skill processes data from remote MLflow servers, which constitutes an indirect prompt injection surface. If a tracking server contains malicious data in trace names or assessment values, it could influence the agent's interpretation of the metrics.\n
  • Ingestion points: Data is fetched from the MLflow API at the URL provided in the --server argument, specifically targeting experiment metrics and trace attributes.\n
  • Boundary markers: No specific boundary markers or instruction-ignore tags are documented for handling external server responses.\n
  • Capability inventory: The skill executes a local Python script to perform network queries and display results.\n
  • Sanitization: The instructions do not describe any sanitization or validation of the retrieved server data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 02:51 PM
Security Audit — agent-trust-hub — querying-mlflow-metrics