image-generation
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
scripts/generate_image.pyto process image generation requests. This script uses standard libraries to interact with the Gemini API and save the resulting files to the local disk. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes user-supplied text to generate images. Ingestion points: User input passed as a prompt argument to the execution script in
scripts/generate_image.py. Boundary markers: None identified. Capability inventory: The script performs network requests to Google's API and writes image files to the system via the Pillow library'simg.savemethod. Sanitization: No explicit sanitization or filtering is applied to the user-provided prompt strings.
Audit Metadata