veterinarian-helper
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety filters were detected. The skill implements a 'gate check' mechanism to verify that the user is a veterinary professional before providing clinical information.
- [REMOTE_CODE_EXECUTION]: The skill consists entirely of Markdown templates and instructional text. No executable code (Python, JavaScript, or shell scripts) is included, and there are no mechanisms for dynamic code execution.
- [DATA_EXFILTRATION]: No network operations, credential harvesting, or access to sensitive local file paths were found. The skill does not transmit data externally.
- [EXTERNAL_DOWNLOADS]: No remote dependencies, package installations, or scripts from external URLs are referenced. References to veterinary manuals and guidelines are textual and for human verification only.
- [NO_CODE]: The skill is a 'no-code' implementation, relying strictly on structured prompt templates and guidelines for the agent's behavior.
Audit Metadata