1. The user requests image generation using Zimage or ModelScope.
2. The agent extracts the prompt from the user's input.
3. The agent checks for an available API key (environment variable, command line, or default file).
4. The agent executes the `generate_zimage.py` script with the prompt and API key.
5. The script sends the request to the ModelScope API.
6. The agent receives the generated image or an error message.
7. The agent displays the image or provides error feedback to the user, including instructions for Alibaba Cloud account binding if necessary.
SynapticSkills