powerbi-boas-praticas

Pass

Audited by Gen Agent Trust Hub on Oct 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions involve analyzing user-provided Power BI technical artifacts (TMDL, DAX, or diagrams), which creates a standard ingestion point for external data.
  • Ingestion points: The auditing process relies on the user providing model definitions or measures for review as specified in SKILL.md.
  • Boundary markers: There are no explicit instructions to use delimiters or ignore instructions embedded within the provided models.
  • Capability inventory: The skill is limited to generating text-based recommendations and DAX/M code snippets; it does not possess tools for automated execution or network operations.
  • Sanitization: No specific sanitization or filtering of the input model data is prescribed.
  • [EXTERNAL_DOWNLOADS]: The skill refers to best practice rules hosted in the official Microsoft Analysis Services repository on GitHub.
  • Evidence: References to the microsoft/Analysis-Services GitHub repository in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 3, 2026, 06:57 PM