Image storage embedding split

Skill fabioc-aloha/Alex_Skill_Mall/plugins/media-graphics/image-storage-embedding-split

Storing images at embed size loses the original:From its SKILL.md

Install

npx -y skills add fabioc-aloha/Alex_Skill_Mall --skill image-storage-embedding-split

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, 459 tokens by cl100k_base, as published. Nobody here has run it

Image Storage vs Embedding Split

The Problem

Storing images at embed size loses the original:

Can't upscale later
Can't use for different contexts
Quality degradation is permanent

The Solution

Keep originals at full resolution. Embed at reduced size.

const sharp = require('sharp');

async function processImage(inputPath, outputDir) {
  const filename = path.basename(inputPath, path.extname(inputPath));
  
  // Store original at full resolution
  const original = await sharp(inputPath)
    .resize(512, 512, { fit: 'inside' })
    .toFile(path.join(outputDir, `${filename}-512.png`));
  
  // Create embed version at reduced size
  const embed = await sharp(inputPath)
    .resize(256, 256, { fit: 'inside' })
    .toBuffer();
  
  // Return both
  return {
    originalPath: path.join(outputDir, `${filename}-512.png`),
    embedDataUri: `data:image/png;base64,${embed.toString('base64')}`,
    embedSize: 256
  };
}

Storage Schema

{
  "images": [
    {
      "id": "avatar-001",
      "originalPath": "assets/avatar-001-512.png",
      "embedDataUri": "data:image/png;base64,iVBOR...",
      "embedSize": 256,
      "originalSize": 512
    }
  ]
}

Token Savings

Size	Base64 Size	Token Estimate
512px	~42KB	~10,500 tokens
256px	~13KB	~3,250 tokens
Savings	~70%	~7,250 tokens

Verification

Original file exists at full resolution
Embedded version is at reduced size
Schema documents both dimensions
Can regenerate embed from original

When to Apply

AI context windows with images
Visual memory systems
Any system with storage/display size trade-offs