power-bi-dax-development

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface that is considered safe due to established mitigations.
  • Ingestion points: Business requirements and user requests are gathered according to instructions in SKILL.md.
  • Boundary markers: The skill enforces Core Principles, such as Research First and Test Everything, which act as behavioral delimiters for the agent.
  • Capability inventory: Uses platform-provided tools such as measure_operations and calculation_group_operations.
  • Sanitization: The workflow includes mandatory validation steps using dax_query_operations to ensure correctness.
  • [COMMAND_EXECUTION]: The skill generates DAX formulas dynamically from provided templates.
  • Evidence: The agent is instructed to write and implement DAX measures based on user needs.
  • Context: This is standard behavior for a specialized development assistant and is performed within the constrained environment of a Power BI model.
  • [EXTERNAL_DOWNLOADS]: The skill references and searches official Microsoft documentation.
  • Evidence: Mentions tools for searching Microsoft Learn and code samples.
  • Context: These interactions are limited to well-known trusted services for documentation purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 04:21 AM