story-cover

To generate a web novel cover, this tool analyzes your book title and genre positioning, mapping them to target platform-specific visual styles and color palettes. It automatically constructs an English image prompt with typography rules and calls the GPT-Image-2 API to create the final cover.

Yes, you can create a readable book title on an AI-generated cover image. The process constructs an English prompt containing specific title and author typography readability rules, then performs iterative quality checks on the generated image to ensure text legibility before saving.

The best way to automate cover creation is by reading book metadata directly from workspace context files like `.active-book` and genre documents. It uses this context to map genre visualization rules, generates the image via GPT-Image-2, and saves the cover to the standard long-form project directory path.

GPT-Image-2 generates the cover image by processing an English prompt that encodes Chinese typography readability and layout rules. The generation process includes iterative quality checks specifically for text legibility, ensuring the rendered title and author names meet platform standards.

Yes, you need a GPT_IMAGE_API_KEY to generate platform-specific web novel covers. This key is required to authenticate requests to the OpenAI images generation endpoint, which produces the cover image based on the constructed English prompt and platform style mapping.

A generated web novel cover fails the quality check when the image lacks style matching, genre alignment, compositional balance, or text readability. The system performs iterative checks across these dimensions and automatically regenerates and optimizes the prompt until the cover passes all quality standards.