AgentSkillsCN

Generate Image

生成图像

SKILL.md

Generate Situation Image

Generate an illustration for a learning situation using Gemini AI.

Arguments

  • <description>: English description of the situation (e.g., "A customer buying items at a Japanese convenience store")
  • <destination>: Output image file path (e.g., app/assets/situations/convenience_store.png)

Instructions

Run the Python script using Bash:

bash
python3 scripts/generate-image.py --description "<description>" --destination "<destination>"

Prerequisites

  • GEMINI_API_KEY or GOOGLE_API_KEY must be set in environment
  • Install dependencies: pip install -r scripts/requirements.txt

Notes

  • The script auto-creates destination directories if they don't exist
  • Edit PROMPT_TEMPLATE in scripts/generate-image.py to customize the image style
  • Output format matches the file extension in destination path