dv-data

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution: The skill utilizes the dataverse CLI for record management, which involves executing shell commands to create, update, or delete data. This is a primary function of the skill for headless operations.
  • Dynamic Script Execution: The instructions guide the agent to generate and run Python scripts using the official Dataverse SDK. This process includes dynamic path manipulation to import local authentication modules.
  • Indirect Prompt Injection Surface: The skill includes patterns for importing data from external sources, such as CSV files. This represents an indirect injection surface where untrusted data could potentially influence agent actions. Ingestion points: CSV processing in SKILL.md and multi-table-fk-import.md. Boundary markers: None specified. Capability inventory: Shell command execution via CLI and record manipulation via SDK. Sanitization: Row data is mapped directly to dictionaries for API payloads.
  • Environment Access: The skill accesses environment variables like DATAVERSE_URL and utilizes a local auth.py script for session management, which are necessary for establishing authenticated connections to the service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:34 PM
Security Audit — agent-trust-hub — dv-data