gemini
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a standard integration for interacting with Google Gemini models. The tools defined are wrappers for API requests or CLI commands that facilitate AI-assisted tasks such as brainstorming and code review.
- [INDIRECT_PROMPT_INJECTION]: The skill includes functionality to process file content (using the
@filenamesyntax) and user-provided prompts. This represents a standard surface for indirect prompt injection common to all AI analysis skills; however, the presence of asandboxparameter for risky operations indicates a design that supports security best practices.
Audit Metadata