krea
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied text and image references for image generation, creating a surface where malicious instructions in user data could potentially influence agent behavior.
- Ingestion points: User prompts and reference image data in SKILL.md.
- Boundary markers: No delimiters or instructions are used to separate user input from the execution context.
- Capability inventory: The skill performs network operations to gateway.pixazo.ai and writes credentials to ~/.pixazo/api-key.
- Sanitization: User input is passed directly to the API without filtering.
- [SAFE]: The skill uses the vendor's own infrastructure (gateway.pixazo.ai) for API communication.
- [SAFE]: API key management follows local security standards by using restricted file permissions (chmod 600) for storage in ~/.pixazo/api-key.
- [SAFE]: The referenced installation command (npx) points to the vendor's official repository.
Audit Metadata