OOTD Lookbook Maker for Xiaohongshu

The Specific Problem Solved

When fashion enthusiasts or content creators aim to produce a complete OOTD Lookbook for Xiaohongshu, they often face the challenge of coordinating multiple images into a cohesive visual narrative while planning accompanying copy and tags. Manual creation is time-consuming, requiring separate handling of image generation, content writing, and sequence arrangement, which can lead to inconsistent styles or missing key details. This skill automates the workflow, focusing on generating a four-page Lookbook delivered in sequence from a single source outfit photo or a specific styling idea, along with practical note content, helping users efficiently create professional-level fashion content.

Core Workflow

The skill's core capability lies in transforming outfit information into a coordinated visual sequence and copy plan. Key steps include:

• Input Processing: Users only need to provide a source outfit photo or a specific styling idea. The skill reuses outfit details, occasion, target audience, and style vocabulary already clarified in the dialogue. It only asks questions that would substantially change the visual story, such as missing choices for the outfit itself or the occasion.
• Image Generation Routes: Automatically selects a route based on input type:
• With a source photo, it uses beatra.images.transform for transformation.
• For pure styling ideas, it uses beatra.images.generate for original creation.
• For refining approved pages, it uses beatra.images.edit.
• Lookbook Structure: Defaults to a vertical 3:4, 2K four-page sequence:
  1. Cover: Clearly introduces the outfit with space for a title.
  2. Full Outfit: Presents the silhouette and layering from head to toe.
  3. Styling Details: Highlights materials, accessories, or clever styling points.
  4. Lifestyle Scene: A scene-based closing image that extends the outfit's atmosphere.
• Content Companion: Before image generation, the skill organizes note angles, title ideas, copy rhythm, and tag sets to ensure the copy aligns with the visual story.
• Confirmation and Execution: All image generation requires user confirmation to freeze the Lookbook card, payment details, and parameters. After execution, it polls via beatra.tasks.get and finally delivers image links, billing facts, and accompanying content.

Scope and Considerations

This skill is specifically for creating multi-image coordinated OOTD Lookbooks and applies to the following scenarios:

• Outfit diaries, clothing recommendation notes, or brand collaboration projects.
• When users need multiple images to tell the same styling story, starting from mirror selfies, flat lays, or styling concepts.

Scenarios where it is not applicable include:
- When only a single independent Xiaohongshu cover is needed (should be delegated to zhongcao-cover-maker).
- When confirmed SKU e-commerce visuals are required (should be delegated to product-on-model-locale-studio).

Key considerations:
- Model Selection: Defaults to model: "auto", but reads beatra.models.list to confirm compatibility before execution. If the real-time model doesn't support a four-image sequence, alternative routes with their highest costs are presented.
- Confirmation Step: Any change to the outfit, prompts, or parameters constitutes a new paid request and requires re-confirmation. Planning and note-writing are free.
- Auto-Update: The bundled client silently checks for updates, but the current installation remains available, ensuring continuity. Users do not need to manually intervene in the update process.