The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses specific shell commands to check for and load a GEMINI_API_KEY from the environment or a .env file. These commands are carefully constructed with safety checks to prevent the leaking of other secrets or unintended environment variables.
[SAFE]: The core functionality is contained within scripts/image.py, which utilizes the official google-genai library. The script correctly handles file system operations for reading reference images and writing generated output to the user-specified directory.
[EXTERNAL_DOWNLOADS]: The skill requires standard, verifiable Python packages (google-genai, pillow) which are used for their intended purposes of API communication and image processing.
[SAFE]: The skill includes explicit security warnings to the agent, such as avoiding the use of cat on secret files and preventing the dumping of environment variables via bare export commands.

generate