dv-query
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- Data Ingestion (Indirect Prompt Injection): The skill reads records and metadata from Dataverse and ERP systems, creating an interface where untrusted data could be processed. This is a common property of data-querying skills and is managed through the use of structured query languages like OData and FetchXML.
- Command Execution: The skill involves executing the
dataverseCLI to perform operations, which is the expected method for headless environment interactions. - External Dependencies: The skill documentation references the installation of standard data science libraries and the official Microsoft Dataverse SDK from public registries.
Audit Metadata