The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing a local Python script (scripts/generation_api.py) to communicate with the vendor's image generation service.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied text prompts and external image URLs for its core image-to-image and text-to-image functions.
Ingestion points: The --prompt and --reference-images arguments in scripts/generation_api.py are the primary entry points for untrusted data.
Boundary markers: The skill does not currently define explicit delimiters or instructions for the agent to segregate user-supplied content from system commands when constructing API calls.
Capability inventory: The skill utilizes the requests library to perform network operations to giggle.pro and includes a download_images function capable of writing files to the local ~/Downloads directory.
Sanitization: The scripts/generation_api.py script performs format validation on reference image inputs (ensuring they are valid URLs, asset IDs, or Base64 strings) before processing them.

giggle-generation-image