generate-images

name	generate-images
description	Generate AI illustrations for course cards and upload to Vercel Blob
user-invocable	false
allowed-tools	Read, Write, Edit, Glob, Grep, Bash

Generate Images

Generate AI illustrations for selected course cards using GPT Image 1.

Usage

/generate-images <course-slug>

Example: /generate-images claude-architect

Workflow

1. Read .env to load OPENAI_API_KEY and BLOB_READ_WRITE_TOKEN.
2. Scan cards for image: "TODO: <description>" markers.
3. For each card with a TODO image: a. Use the description to generate an illustration via GPT Image 1. b. Upload the PNG to Vercel Blob at path courses/<slug>/images/<card-id>.png. c. Update the card's frontmatter: replace image: "TODO: ..." with image: "<blob-url>". d. Update .enrichment-hashes.json with the image description hash.
4. Use illustrations selectively — only where a visual genuinely aids understanding (architecture diagrams, flow charts, concept maps).
5. Run the media validator after completion:

   npx tsx scripts/validate-media.ts courses/<slug>
   

GPT Image 1 API

curl https://api.openai.com/v1/images/generations \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "<description>",
    "n": 1,
    "size": "1024x1024",
    "quality": "medium"
  }'

Vercel Blob Upload

curl -X PUT "https://blob.vercel-storage.com/courses/<slug>/images/<card-id>.png" \
  -H "Authorization: Bearer $BLOB_READ_WRITE_TOKEN" \
  -H "Content-Type: image/png" \
  -H "x-api-version: 7" \
  --data-binary @image.png