vd-plumbing-vs-behavior

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains purely instructional content focused on software testing methodologies and the distinction between component verification and end-to-end behavior testing.
  • [SAFE]: No suspicious commands, network activities, or data exfiltration patterns were detected. The mentions of stdio and MCP callTool are illustrative of testing concepts, not executable instructions for the agent to perform malicious actions.
  • [SAFE]: The skill does not include any external dependencies, package installations, or remote script execution. The reference to the vanta CLI tool is documented as an example for running integration tests within a development workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 06:06 PM
Security Audit — agent-trust-hub — vd-plumbing-vs-behavior