gemini-carousel
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses directive language designed to override default agent behaviors. Specifically, the headers 'CRITICAL: Auto-start on load' and the instruction 'Do not summarise' are attempts to force a specific execution path and suppress standard conversational summaries.
- [PROMPT_INJECTION]: The skill processes untrusted user input to generate design briefs and image prompts.
- Ingestion points: User-provided content is collected in Step 1.
- Boundary markers: The skill does not define clear delimiters or provide instructions to the model to ignore potential commands embedded within the user-provided text.
- Capability inventory: The skill generates complex prompts and layout instructions based on the input.
- Sanitization: There is no evidence of filtering or validation of the user-provided content before it is processed.
Audit Metadata