Pop-cam NanoBanana Skill

Generate images with the Pop-cam NanoBanana API. The model supports any visual style — photorealistic, cinematic, cartoon, illustration, watercolor, 3D render, and anything else you can describe in a prompt.

This skill supports two modes:

• Image-to-image: send a base64-encoded photo and a prompt describing how to transform it.
• Text-to-image: send only a prompt field (no image) to generate an image from scratch.

Both modes cost 1 credit per generation and use the same endpoint.

Authentication

All generation requests require a Pop-cam API token (pk_ prefix) in the Authorization header.

If the user does not have a token yet, walk them through these steps in order:

1. Sign up at https://www.pop-cam.com/sign-up (skip if they already have an account).
2. Sign in at https://www.pop-cam.com/sign-in.
3. Create an API token at https://www.pop-cam.com/developer — label it "OpenClaw". Copy the token immediately; it is only shown once.

Store the token and send it on every request:

Authorization: Bearer pk_YOUR_TOKEN

If unsure whether the user is signed in, call the auth-status endpoint first:

curl https://www.pop-cam.com/api/openclaw/auth-status

The response tells you whether sign-in is needed and provides the relevant URLs.

Skill Definition JSON

For a machine-readable version of this entire skill (features, endpoints, schemas, examples), call:

curl https://www.pop-cam.com/api/openclaw/skill

Generation Endpoint

POST https://www.pop-cam.com/api/v1/nanobanana

Headers

Request Body

Image-to-Image Example

Transform an existing photo:

curl -X POST https://www.pop-cam.com/api/v1/nanobanana \
  -H "Authorization: Bearer $POPCAM_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "image": "data:image/png;base64,iVBOR...",
    "prompt": "Enhance this photo with cinematic lighting and shallow depth of field"
  }'

Text-to-Image Example

Generate an image from a text prompt only — no source photo needed:

curl -X POST https://www.pop-cam.com/api/v1/nanobanana \
  -H "Authorization: Bearer $POPCAM_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A photorealistic portrait of a man in a leather jacket standing on a city rooftop at golden hour, natural lighting, shallow depth of field"
  }'

Synchronous Response (200)

{
  "job_id": "a1b2c3d4-...",
  "image_url": "https://cdn.pop-cam.com/generated/user_id/a1b2c3d4.png",
  "download_url": "https://...r2.cloudflarestorage.com/...?X-Amz-Signature=...",
  "used_prompt": "A photorealistic portrait of a man in a leather jacket standing on a city rooftop at golden hour, natural lighting, shallow depth of field",
  "credits_remaining": 22,
  "mode": "text-to-image"
}

Important: Use download_url to fetch the generated image. It is a presigned URL valid for 1 hour that works regardless of storage bucket privacy settings. The image_url is the permanent stored reference but may not be directly accessible if the bucket is private.

The mode field is "text-to-image" or "image-to-image" depending on which path was used.

Webhook / Async Mode

Include webhook_url to get an immediate HTTP 202 response with a job_id. Pop-cam POSTs the result to your URL when generation completes (retries up to 3 times with exponential backoff).

Initial response (202):

{
  "job_id": "a1b2c3d4-...",
  "status": "processing",
  "message": "Your image is being generated. Results will be POSTed to your webhook_url."
}

Webhook delivery payload includes download_url (presigned, 1-hour expiry) for fetching the image.

Error Handling

Credits

• Each generation (both modes) costs 1 credit.
• New accounts receive starter credits for free.
• Every response includes credits_remaining.
• Purchase more at https://www.pop-cam.com/checkout.

Prompt Tips

The model is not limited to any single style. The output depends entirely on your prompt. Here are example styles you can request:

For the most realistic output, include details like lighting conditions, camera lens, depth of field, and environment in your prompt.

Quick Reference