baoyu-image-gen

AI image generation across providers is automated by routing text prompts to Google, OpenAI, DashScope, or Replicate without manual tool switching. You provide a prompt and output path, and the Skill handles provider selection and model defaults.

Yes, reference images can be used to guide AI image generation. The Skill accepts optional reference images alongside your text prompts, allowing you to direct the visual output while specifying aspect ratios and output sizes.

Default models, providers, quality, and sizes for image generation are configured using project- or user-level EXTEND.md files. This streamlines your workflow by applying consistent defaults across multiple generation requests.

Reading prompts from direct input, files, or stdin is fully supported for AI image generation. This allows you to automate batch processing and integrate the Skill into existing pipelines that output text prompts.

Robust error handling is built into cross-provider image generation, including optional parallel generation to improve throughput. This ensures failures are managed gracefully without interrupting the entire batch processing workflow.

When switching between Google, OpenAI, DashScope, and Replicate for image generation, limitations depend on each specific provider's API constraints. The Skill standardizes prompt handling and sizing controls, but model availability and quality presets vary by provider.