dv-solution

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Command Execution via CLI Tools: The skill relies on the Power Platform CLI (pac) to perform operations such as exporting, importing, and packing solutions. This is an expected pattern for administrative tasks in this ecosystem.
  • Indirect Prompt Injection Surface: The skill ingests data from Dataverse records (e.g., publishers, solutions, and forms) to inform its operations. This creates a potential surface where external data could attempt to influence the agent's logic.
  • Ingestion points: Data is retrieved using client.records.get in SKILL.md to identify existing publishers and verify deployment status.
  • Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore potentially malicious content embedded within the retrieved record fields.
  • Capability inventory: The skill possesses the capability to execute shell commands (pac) and perform network operations (urllib.request, DataverseClient).
  • Sanitization: The provided Python snippets do not implement explicit sanitization or validation of the values retrieved from the Dataverse environment before they are used in commands or logic.
  • Network Operations: The skill performs network requests to Dataverse API endpoints using the urllib.request library and a specialized Python client. These requests are used to perform administrative queries and operations within the authorized environment.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 06:14 PM