ads-photoshoot
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user inputs, such as product descriptions and image URLs, which are interpolated directly into prompts for the image generation model.
- Ingestion points: User-provided product description, product image path, and product URL in SKILL.md.
- Boundary markers: None identified; the templates use direct interpolation of variables like [product description].
- Capability inventory: Executes /banana generate and writes files to the local ./product-photos/ directory.
- Sanitization: The skill does not define any validation or escaping mechanisms for the user-supplied text before it is included in the generative prompt.
- [COMMAND_EXECUTION]: The skill relies on executing external CLI tools (/banana setup, /banana generate, /ads generate) to configure API access and perform the core image generation logic.
- [EXTERNAL_DOWNLOADS]: The skill allows the use of external URLs for product images and requires the installation of the banana-claude and nanobanana-mcp toolsets (vendor-related resources) which are external dependencies.
Audit Metadata