Power BI Custom Visuals Development

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill documentation facilitates the processing of external data via the DataView API in Power BI visuals (src/visual.ts). This creates a surface for Indirect Prompt Injection, where malicious content in Power BI data fields could attempt to influence agent behavior during visual debugging or development.
  • Ingestion points: options.dataViews is processed in the update method of visual.ts and in various reference files.
  • Boundary markers: The provided examples do not explicitly implement prompt delimiters or instructions to ignore embedded commands in the ingested data.
  • Capability inventory: The skill environment is configured with Bash, Write, Edit, Glob, Grep, and Read tools.
  • Sanitization: The instructions demonstrate high security awareness by explicitly warning against using innerHTML or D3.html() with user data and recommending safe alternatives like textContent.
  • [EXTERNAL_DOWNLOADS]: The skill directs users to install the powerbi-visuals-tools package globally via npm and references official Microsoft documentation. These are well-known and trusted resources for Power BI development.
  • [COMMAND_EXECUTION]: The skill involves the execution of standard toolchain commands using the pbiviz CLI (e.g., pbiviz new, pbiviz start, pbiviz package) and the installation of local development SSL certificates. These operations are routine for the intended development workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 05:31 AM
Security Audit — agent-trust-hub — Power BI Custom Visuals Development