Generate images
and finish them in the same call chain

1. 1

   Choose an image agent

   An agent is a saved configuration: engine, size, style, and the prompt template your team agreed on. List the ones your workspace has and keep the id.

   curlresponse

   Copy

   export AUTOSTUD_API_KEY="sk_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
   
   # An image agent is a saved configuration, not a model name.
   curl "https://autostud.ai/api/v1/agents?agent_type=image&limit=10" \
     -H "Authorization: Bearer $AUTOSTUD_API_KEY"

2. 2

   Generate

   Send the prompt, plus any variables the template expects. The response carries the image and the credits it spent.

   curlNoderesponse

   Copy

   curl -X POST https://autostud.ai/api/v1/actions/images.generate \
     -H "Authorization: Bearer $AUTOSTUD_API_KEY" \
     -H "Content-Type: application/json" \
     -H "Idempotency-Key: sku-4471-packshot" \
     -d '{
       "agent_id": "a91d5f27-6c0b-4e83-97aa-2b1e4d70c6f9",
       "prompt": "A ceramic mug seen three-quarter, soft studio light",
       "variables": {
         "product": "ceramic mug",
         "color": "matte navy"
       }
     }'

3. 3

   Finish the file

   Cut the background out, apply an instruction-based edit, or reframe it to another format. Same key, same ledger, no second integration.

   cut outeditwebhook

   Copy

   curl -X POST https://autostud.ai/api/v1/actions/tools.remove_background \
     -H "Authorization: Bearer $AUTOSTUD_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "image_url": "https://cdn.autostud.ai/generations/b6c30a19.../image.png"
     }'

Questions developers ask first