/describe2 creditsMidjourney onlyDescribe
Reverse-engineer Midjourney prompts from any image. Submit an image URL and receive four distinct prompt suggestions that would reproduce the image's style, subject, and composition. Essential for building prompt libraries from reference images, understanding how Midjourney 'sees' an image, and matching the aesthetic of existing artwork. Midjourney only.
Request Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| image_url | string | Required | Publicly accessible URL of the image to describe. |
| webhook_url | string | Optional | URL where we POST the completed task result. |
| webhook_secret | string | Optional | Sent as x-webhook-secret header in webhook delivery. |
Example Request
curl -X POST https://api.journeyapi.com/api/v1/describe \
-H "Authorization: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"image_url": "https://example.com/photo.jpg",
"webhook_url": "https://your-server.com/webhook"
}'Immediate Response
Returns immediately with a task ID. Use /fetch or webhooks to get the result.
{
"task_id": "ff0i8e00-i29l-41n4-k716-446655440000",
"task_type": "describe",
"image_url": "https://example.com/photo.jpg",
"content": [
"1️⃣ a cyberpunk cityscape with neon lights reflecting on wet streets, rain falling, futuristic buildings --ar 16:9",
"2️⃣ neon-lit urban landscape at night, reflective wet pavement, sci-fi architecture --ar 16:9",
"3️⃣ a rainy night in a futuristic city, glowing neon signs, dark atmospheric mood --ar 16:9",
"4️⃣ cyberpunk street photography, neon reflections, dystopian urban environment --ar 16:9"
]
}Response Fields
| Field | Type | Description |
|---|---|---|
| task_id | string | UUID of the describe task. |
| task_type | string | Always "describe". |
| image_url | string | The image URL that was described (echoed back). |
| content | string[] | Array of 4 numbered description/prompt suggestions. |
Tips
- 1Use describe to reverse-engineer prompts from existing images.
- 2The returned prompts can be fed directly into /imagine.
- 3Great for understanding what Midjourney "sees" in an image.
Quirks & Gotchas
Content items are prefixed with emoji numbers (1️⃣, 2️⃣, etc.).
Expert Tips & Best Practices
Building a prompt library from reference images
Run /describe on curated reference images and save the output prompts. Over time, this builds a library of proven prompt patterns for specific styles, subjects, and moods. When you want to match an aesthetic from a reference image you found, describe it first, then use one of the returned prompts as a starting point for /imagine — often more effective than writing a prompt from scratch.
Understanding Midjourney's visual vocabulary
The prompts returned by /describe reveal exactly which artistic, stylistic, and compositional terms Midjourney associates with an image. Use these terms to strengthen your own prompts. If you see an artist name, art style, or technical photography term in the description that you didn't know about, try adding it to your own prompts to achieve similar results.