image-generation
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
gemini.pyandopenai_image.py) to perform image generation and editing tasks. - [DATA_EXFILTRATION]: The skill reads local files and sends them to Google and OpenAI APIs. This is standard functionality for image editing and product placement features, and the destinations are well-known technology services.
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes untrusted user prompts and reference images.
- Ingestion points: User prompts and reference image files provided through arguments to the Python scripts.
- Boundary markers: Absent. The skill instructions do not specify the use of delimiters or warnings to ignore instructions embedded in reference files.
- Capability inventory: The skill has the capability to read local files, write files to disk, and make network requests to external APIs.
- Sanitization: Absent. The scripts do not perform sanitization or validation of prompts or image content before transmission to the APIs.
Audit Metadata