huashu-image-upload
Skill3 starsChanged 6 months agoMITView on GitHub
One-click generation and upload of article illustrations to an image host, with automatic Markdown link insertion. Use when the user mentions "illustrations", "article images", "upload images", or "image for the article".
We split this skill into the rules it gives an agent
14 rules. About 0 are common, 13 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈13
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
After article proofreading is complete and illustrations are needed
- Distinctive: line 16
When the article content is complete but images are missing
- Distinctive: line 25
Auto-upload to image host: Call ImgBB to obtain permanent links
- Distinctive: line 27
Verify display: Ensure all image links are valid
- Distinctive: line 33
Maximum: No more than 10 (to avoid excessive interruption to reading flow)
- Distinctive: line 56
Method: Use WebFetch to obtain Wikimedia Commons image links
- Distinctive: line 63
Method: Call API to generate image, returns a temporary URL
- Very specific: line 83
Use the project's existing /tools/upload_image.py script
- Distinctive: line 112
Use permanent network links (not local paths)
- Distinctive: line 113
Write a meaningful image description (alt text)
- Distinctive: line 114
Keep descriptions concise and focused on the image content
- Distinctive: line 144
Before executing an illustration task, confirm:
- Distinctive: line 160
Read the full article, mark key sections
- Distinctive: line 217
A: Not recommended. Local paths break when pasted into the WeChat editor — always use network links.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Image Generation & Upload
A professional capability for automatically generating illustrations for WeChat articles, uploading them to an image host, and inserting Markdown links.
When to Use This Skill
This skill is automatically loaded when the following scenarios are detected:
- The user says "add illustrations now", "help me with illustrations", or "add images"
- After article proofreading is complete and illustrations are needed
- When the article content is complete but images are missing
Core Features
- Analyse illustration requirements: Determine 5–8 illustrations needed based on article content
- Source images from:
- Public domain works (Wikimedia Commons, Google Arts & Culture)
- AI-generated images (Volcano Engine doubao-seedream API)
- Free stock libraries (Unsplash, Pexels)
- Auto-upload to image host: Call ImgBB to obtain permanent links
- Insert Markdown: Insert image links at the appropriate positions in the article
- Verify display: Ensure all image links are valid
Illustration Quantity Guidelines
- Recommended: 5–8 images
- Minimum: 3 images (cover + 2 body images)
- Maximum: No more than 10 (to avoid excessive interruption to reading flow)
Illustration Placement Strategy
Mandatory placement:
- ✅ Cover image (hero image): Below the title — always required
- ✅ Core sections: 1 image per key section
Optional placement:
- Theory support (e.g., referencing a classic work)
- Data visualisation (comparison charts, graphs)
- Case supplements (product screenshots, examples)
Image Source Priority
1️⃣ Public Domain Works (Highest Priority)
- Sources: Wikimedia Commons, Google Arts & Culture
- Suitable for: Classic artwork, historical portraits, ancient book covers
- Advantages: Free, no copyright issues, high quality
- Examples: Van Gogh's Starry Night, Kant portrait
- Method: Use WebFetch to obtain Wikimedia Commons image links
2️⃣ AI-Generated (Recommended)
- Source: Volcano Engine doubao-seedream-4-0-250828
- Suitable for: Cover images, concept art, abstract themes
- Advantages: Original, customisable, fast
- Method: Call API to generate image, returns a temporary URL
3️⃣ Free Stock Libraries
- Sources: Unsplash, Pexels, Pixabay
- Suitable for: Cover images, concept art, background images
- Advantages: Free, high quality, commercially usable (CC0)
- Method: WebFetch search + download
4️⃣ Screenshots / Official Assets (Citation required)
- Sources: YouTube, Bilibili, product websites, film posters
- Suitable for: Case images (product screenshots, video stills)
- Note: Cite the source; use under fair use principles
- Method: Remind user to take their own screenshot, or generate an AI concept image as a substitute
Image Host Upload Workflow
Core Script
Use the project's existing /tools/upload_image.py script
Upload Method
# Call upload script (via Bash)
python3 /Users/alchain/Documents/writing/tools/upload_image.py <image URL or local path>
# Handle web images (e.g., AI-generated temporary links)
python3 /Users/alchain/Documents/writing/tools/upload_image.py "https://example.com/ai-generated.jpg"
# Handle local images
python3 /Users/alchain/Documents/writing/tools/upload_image.py "/Users/alchain/Pictures/image.png"
# Script automatically returns ImgBB permanent linkFallback Mechanism
- ✅ Prefer uploading to ImgBB image host (permanent links)
- ⚠️ If upload fails, automatically use the original URL as a fallback
- 📌 See
/tools/README.mdfor image upload script documentation
Markdown Insert Format
- Use permanent network links (not local paths)
- Write a meaningful image description (alt text)
- Keep descriptions concise and focused on the image content
Image Specifications
Size Specifications
- Cover image: 1200 x 600px (16:9)
- Body images: 800–1200px wide
- Person/product: 800 x 800px (1:1)
File Specifications
- Format: JPG (photos) / PNG (illustrations)
- Size: < 500KB
- Naming: lowercase, semantic, in English (e.g.,
mrbeast.jpg,starry-night.jpg)
Path Specifications
- ✅ Network link:
https://i.ibb.co/xxxxx/image.jpg(recommended, permanent) - ❌ Local path: No longer used (breaks when pasted into the WeChat editor)
Copyright Notes
- ✅ Public domain works (author deceased 70+ years)
- ✅ CC0-licensed images (Unsplash, etc.)
- ✅ AI-generated images (confirm tool licence)
- ⚠️ Screenshots/official assets require source attribution
Illustration Checklist
Before executing an illustration task, confirm:
- Article has been proofread and content is finalised
- Number of illustrations determined (5–8)
- Illustration positions confirmed (cover + core sections)
- All images obtained (AI-generated / public domain / stock)
- All images uploaded to ImgBB image host
- Image links are permanent network links (starting with https://)
- Images inserted in Markdown (using network links)
- Image descriptions (alt text) have been filled in
- All image links verified as valid (accessible)
Execution Steps (5-Step Standard Workflow)
Step 1: Analyse article and determine illustration requirements
- Read the full article, mark key sections
- List illustration requirements (6–8 items)
- Determine the position and purpose of each image
Step 2: Source / generate images
Based on source priority order:
- Public domain works → WebFetch to retrieve
- AI-generated → Call API
- Free stock → WebFetch search
- Screenshot assets → Remind user to provide
Step 3: Upload to image host, obtain permanent links
# Call upload script for each image
python3 /Users/alchain/Documents/writing/tools/upload_image.py <image URL>Step 4: Insert images into article
Step 5: Verify display
# Check all image references
grep -n "!\[" "article-path.md"
# All image links should start with https://Reference Resources
/tools/README.md— Image upload script documentation/tools/upload_image.py— Image upload script/writing/AI-image-generation-API.md— AI image generation API documentation/writing/images/illustration-best-practices.md— Illustration best practices
Frequently Asked Questions
Q1: Will AI-generated image temporary links expire?
A: Yes. That's why you must call upload_image.py to upload to ImgBB and get a permanent link.
Q2: What if the image host upload fails?
A: The script has a fallback mechanism and will automatically use the original URL. However, it's recommended to check your ImgBB API configuration.
Q3: Does every article need illustrations?
A: Illustrations are strongly recommended, especially for long articles (3,000+ words). Short articles (under 1,000 words) may only need a cover image.
Q4: Can local image paths be used?
A: Not recommended. Local paths break when pasted into the WeChat editor — always use network links.
Technical Dependencies
- Python script:
/tools/upload_image.py(already in place) - API configuration: ImgBB API key (configured in
~/.zshrc) - AI generation API: Volcano Engine doubao-seedream (optional)
- Tools: WebFetch (for retrieving public domain images)
Success Examples
- "Why the Best Content Is Useless": 6 illustrations (2 public domain + 4 AI-generated)
- "DeepSeek-OCR In-Depth Review": 5 illustrations, all uploaded to ImgBB, copy-pasted to WeChat without issue
Last Updated: 2025-11-07 Applicable Projects: WeChat article writing Maintained by: Huashu
By Huashu | AI Native Coder · Independent Developer WeChat Official Account "Huashu" | 300k+ followers | AI Tools & Productivity Flagship products: Kitty Catchlight (AppStore Paid Rankings Top 1) · Play DeepSeek with One Book
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- alchaincyf/huashu-skills★ 2ksame name
- dqtx760/skills-manage★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- huashu-agent-swarmSkill
- huashu-article-editSkill
- huashu-article-to-xSkill
- huashu-data-proSkill
- huashu-designSkill
- huashu-douyin-scriptSkill
- huashu-info-searchSkill
- huashu-material-searchSkill
- huashu-md-to-pdfSkill
- huashu-prompt-saveSkill
- huashu-proofreadingSkill
- huashu-researchSkill
- huashu-script-polishSkill
- huashu-slidesSkill
- huashu-speech-coachSkill
- huashu-topic-genSkill
- huashu-video-checkSkill
- huashu-video-outlineSkill
- huashu-wechat-imageSkill
- huashu-xhs-imageSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

