gemini
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text and prompts, along with external documentation, which could theoretically contain malicious instructions.
- Ingestion points: Processes user input via the
textandpromptfields in SKILL.md, and fetches remote documentation fromhttps://www.pixazo.ai/models/gemini.md. - Boundary markers: No specific delimiters or "ignore" instructions are used when interpolating input into the API request.
- Capability inventory: Performs network operations using
curl,requests(Python), andfetch(Node.js) to interact with the Pixazo API gateway. - Sanitization: No explicit sanitization of input data or API responses is performed before processing or display.
- [EXTERNAL_DOWNLOADS]: The skill fetches reference documentation from the vendor's official website at
https://www.pixazo.ai/models/gemini.mdto ensure correct API usage.
Audit Metadata