The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill assembles a structured prompt by interpolating user-provided inputs into a labeled template. This creates a surface where instructions embedded within the user data could potentially influence the image generation model's behavior or bypass intended constraints.
Ingestion points: scripts/image_gen.py accepts user input through various command-line arguments such as --prompt, --use-case, and --constraints.
Boundary markers: The prompt uses simple labeled sections (e.g., "Primary request:", "Constraints:") but lacks robust delimiters or explicit instructions to the model to ignore any control sequences in the user content.
Capability inventory: The skill is capable of writing generated image data to the local file system and making outbound network requests to the OpenAI Image API.
Sanitization: There is no evidence of escaping, filtering, or validation performed on user-supplied text before it is inserted into the prompt template.
[EXTERNAL_DOWNLOADS]: The skill instructions and dependencies list well-known Python packages including openai and pillow. These are standard, reputable libraries for interacting with the OpenAI API and performing image processing tasks, and are fetched from the official Python Package Index (PyPI).

imagegen