The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the well-known Python packages google-genai and pillow for its core operations.
[COMMAND_EXECUTION]: The skill executes a Python script via uv run to interact with the Gemini API, which is the primary intended function of the skill.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external user inputs.
Ingestion points: The script scripts/generate_image.py ingests user prompts and image files provided via command-line arguments.
Boundary markers: User-provided text is passed to the API call without structural delimiters or specific instructions to the model to ignore embedded commands.
Capability inventory: The skill is capable of writing files to the disk and making network requests to the Gemini API endpoint.
Sanitization: No input validation or sanitization is applied to the user-supplied prompts or images before they are processed by the script and the model.

nano-banana-pro