ai-image-editorlisted
Install: claude install-skill popnow-ai/popnow-skills
# Prompt-based image editing
Image-to-image editing takes one or more input images plus an instruction and returns a new image. PopNow exposes the edit endpoints of the leading models behind one key, so the agent can pick the best editor for the change.
## Models
These are the PopNow model paths at the time of writing. Confirm them with `models` before use; newer versions appear there first.
| Model path | Best for |
|---|---|
| `google/nano-banana-2.1/edit` | General edits, fast |
| `google/nano-banana-pro/edit` | Precise 4K edits, layouts |
| `openai/gpt-image-2.5-sunburst/edit` | Text and logo changes |
| `bytedance/seedream-5.0-pro/edit` | Up to ten references, consistency |
| `black-forest-labs/flux-2-max/edit` | Photoreal relighting |
| `alibaba/qwen-image-3.0-pro/edit` | Region-specific text edits |
## Prompt tips
- Input images must be URLs the API can fetch; check the schema for the field name (often `image_urls` or `images`).
- Write edits as: change X, keep Y. Example: change the sofa to green velvet, keep the room, light and camera angle.
- For product photos on a new background, give the product image first and describe the scene second.
- Only edit photos the user has rights to; get consent before editing a real person's likeness.
## Setup and cost
Requires Python 3.8+ and a PopNow API key in `POPNOW_API_KEY`. Create a dedicated key at https://popnow.ai/api/console/keys and restrict its models, IPs and spending caps. Keep the key in the agent's secret or