Image Gen

The image-generation engine of the product-shots ecosystem. A single, parameterised entry point: routes the request to the correct API endpoint based on model family (OpenAI's /v1/images/generations + /v1/images/edits vs Gemini's /v1/chat/completions multimodal), preprocesses reference images, and returns a saved file path.

Primary backend: OmniMaaS / Cloubic image gateway — https://api.omnimaas.com/v1. The gateway is OpenAI-SDK-compatible and unifies access to GPT image 2 and the Gemini Nano Banana family behind one auth token. See:

• GPT image 2 via OmniMaaS: https://docs.cloubic.com/docs/zh-CN/image-generation/image-openai
• Gemini via OmniMaaS: https://docs.cloubic.com/docs/zh-CN/image-generation/image-gemini

Fallback backends: any other OpenAI-SDK-compatible image gateway works — set PRODUCT_SHOTS_IMAGEGEN_BASE_URL + PRODUCT_SHOTS_IMAGEGEN_API_KEY, or keep the legacy CANVASFLOW_IMAGEGEN_* env vars for migrated installs.

Onboarding (First-Use Setup)

Before generating any image, this skill MUST verify an API gateway is configured. If env vars are missing, surface the instructions below to the user verbatim — never silently fall back to a hard-coded key, never ask the user to fork the repo, never ask them to paste a key into chat.

Detection logic

Run on every invocation. The resolution order: