scenario-gemini-image
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill guides the user in constructing natural language prompts for the model_run tool, which are then processed by Google Gemini models. This setup creates an indirect prompt injection surface where user-supplied instructions could influence model outputs.
- Ingestion points: The prompt parameter within the model_run tool (SKILL.md).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between user data and model instructions.
- Capability inventory: The skill has the ability to execute model runs and download resulting assets (model_run, asset_download).
- Sanitization: There is no mention of input validation or sanitization for the prompt strings.
- [EXTERNAL_DOWNLOADS]: The documentation mentions an installation command (npx skills add scenario-labs/skills) for related tools. This points to the official repository of the skill author, scenario-labs, and is part of the intended setup process.
Audit Metadata