Extract product text and design details from images to generate Fantui-style prompts.
npx skills add https://github.com/ghostltx/codex-skills --skill baokuan-tupian
Please help me install this Agent Skill. Skill: baokuan-tupian Source: https://github.com/ghostltx/codex-skills/tree/main/baokuan-tupian Command: npx skills add https://github.com/ghostltx/codex-skills --skill baokuan-tupian
Identify the product subject from user-provided images and record all visible image content, including text, labels, and layout cues to guide automated generation.
Provide a product image (preferably white-background); the skill will analyze it, extract visible text and features, create a premium ecommerce prompt, and generate the final image with RH-GPT-IMAGE-2-I2I.
High-intent search queries and answers about installing and using this skill.
To convert a white-background product image, the skill identifies the product subject, extracts visible text, builds a structured Fantui-style prompt, and generates a premium 2k resolution image via RH-GPT-IMAGE-2-I2I.
Yes, you can reverse engineer an Amazon reference image. The skill analyzes product recognition details and extracts text, labels, and layout cues to automatically craft a high-conversion Fantui-style ecommerce prompt.
The best way to generate high-conversion visuals is using an automated workflow that extracts product features, constructs a prompt with a title, subtitle, and 3-5 callouts, and renders the final image via RH-GPT-IMAGE-2-I2I.
Yes, the workflow supports custom text and callouts. It captures all visible text from your reference image and enforces a structured prompt containing a title, subtitle, and 3-5 callouts for the final image generation.
A white-background image is preferred but not strictly required. The skill identifies the product subject and extracts all visible design details and text to guide the automated generation of the final image.
Missing text or layout cues occur if the source image lacks clear visibility. The skill extracts all readable text and features to build the prompt, so providing a high-contrast white-background image ensures accurate capture.
iterationlayerConvert product photos to Amazon-compliant 2000×2000px JPEGs with white backgrounds.
ghostltxAutomate Amazon product-image generation across modes A, B, C, and D.
RASCLAWValidate structured prompt JSON for product fidelity before image generation.
Yukine5377Analyze text or image input to generate infographic illustrations via Gemini.
