baoyu-xhs-images
Generates Xiaohongshu (Little Red Book) infographic series with 11 visual styles and 8 layouts. Breaks content into 1-10 cartoon-style images optimized for XHS engagement. Use when user mentions "小红书图片", "XHS images", "RedNote infographics", "小红书种草", or wants social media infographics for Chinese platforms.
89
81%
Does it follow best practices?
Impact
97%
2.15xAverage score across 6 eval scenarios
Passed
No known issues
Evaluation results
100%
81%Xiaohongshu Infographic Series Planning
File structure and outline generation
Directory slug format
100%
100%
xhs-images parent
0%
100%
Source file naming
28%
100%
analysis.md YAML frontmatter
0%
100%
Image count range
0%
100%
Auto style selection
71%
100%
Three strategy files
0%
100%
Outline YAML frontmatter
0%
100%
Different styles per strategy
37%
100%
Strategy A image count
0%
100%
Strategy B image count
0%
100%
Strategy C image count
0%
100%
Image filename format
0%
100%
Cover is sparse
0%
100%
Ending is sparse or balanced
25%
100%
94%
31%Prepare Image Generation Prompts for a Xiaohongshu Series
Prompt file assembly and reference chain
Prompt file count
100%
100%
Prompt file naming
50%
100%
Portrait aspect ratio
100%
100%
Hand-drawn style requirement
62%
100%
Text must be hand-drawn
37%
100%
Style section included
100%
100%
Layout section included
100%
100%
Content section per image
85%
100%
Language matches content
100%
83%
Nano banana pro ending
0%
100%
Cover generated without ref
100%
100%
Subsequent images use ref
0%
100%
Session ID format
0%
0%
100%
76%Set Up XHS Image Generator Preferences
EXTEND.md preferences configuration
Correct file location
0%
100%
Correct filename
0%
100%
Schema version field
0%
100%
Watermark enabled
0%
100%
Watermark content
0%
100%
Watermark position
100%
100%
Preferred style name
0%
100%
Preferred layout
0%
100%
Language field
0%
100%
Custom style entry
37%
100%
Custom style color palette
37%
100%
Custom style description
100%
100%
Valid YAML syntax
100%
100%
100%
82%Concert Tour Infographic Series
Screen-print style prompts and auto-selection
Auto-selection: screen-print
0%
100%
Flat color blocks stated
0%
100%
Color count constraint
0%
100%
Halftone patterns
0%
100%
No handwritten fonts
0%
100%
Bold condensed typography
0%
100%
Paper grain texture
0%
100%
Watermark section in prompts
100%
100%
Prompt ending phrase
0%
100%
Portrait 3:4 aspect ratio
28%
100%
Style section from preset
0%
100%
Content section per prompt
42%
100%
Correct prompt count
100%
100%
88%
10%Update and Expand a Xiaohongshu Infographic Series
Image modification workflow and file backup rules
Backup of original prompt-02
0%
0%
Updated prompt-02 content
100%
100%
New prompt-04 created
100%
100%
Former prompt-04 renumbered to 05
100%
100%
Outline updated to 5 images
100%
100%
generate-changes.sh uses --ref for images 2, 4, 5
0%
100%
generate-changes.sh excludes image 1
100%
100%
Prompt-04 ends with nano banana pro
100%
100%
Prompt-05 position updated
100%
100%
New prompt-04 portrait aspect ratio
100%
100%
New prompt-04 hand-drawn instruction
100%
100%
100%
31%Revised Healthy Recipe Infographic Series
Directory conflict resolution and completion report
Conflict resolution: timestamp suffix
0%
100%
Original directory untouched
100%
100%
Completion report exists
100%
100%
Report: topic field
50%
100%
Report: mode field
50%
100%
Report: style field
100%
100%
Report: layout field
100%
100%
Report: image count
100%
100%
Report: location field
71%
100%
Report: file checklist
100%
100%
3 prompt files exist
100%
100%
Prompt files end with nano banana pro
0%
100%
outline.md image_count matches
100%
100%
- Repository
- jimliu/baoyu-skills
- Commit
9eb032a
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.