skill-freshness-audit

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches documentation and code patterns from official sites like docs.databricks.com, mlflow.org, and learn.microsoft.com, as well as the databricks-solutions GitHub repository. These fetches are used solely for auditing and verification purposes and target well-known, trusted sources.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection (Category 8) by ingesting data from external URLs for processing. * Ingestion points: Document URLs in references/verification-sources.md and upstream paths from databricks-solutions/ai-dev-kit. * Boundary markers: Not present; the skill does not wrap external data in delimiters or provide ignore-instruction warnings. * Capability inventory: The skill uses WebFetch for network retrieval and the local scripts/scan_skill_freshness.py script for file system reads. * Sanitization: Absent; the skill performs no filtering or validation on the content retrieved from external sources before the agent processes it for drift detection.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 02:33 AM