Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Generates new images or edits existing ones using Google's Gemini 3 Pro Image API. Supports text-to-image and image-to-image workflows at 1K, 2K, or 4K resolution. Runs locally via a Python script with a draft-iterate-final workflow to avoid wasting high-resolution quota.
The built-in draft-at-1K-then-finalize-at-4K workflow saves API quota by locking the prompt before committing to high-resolution output.
When to use it
- Generate a product mockup from a text description
- Remove an unwanted object from a photo
- Convert a photo to a watercolor painting style
- Create a 4K hero image for a blog post
- Adjust the lighting or color palette of an existing image
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: edit my photo 'storefront.jpg' to make the sky look stormy
- 1Check that storefront.jpg exists in the current directory
- 2Run generate_image.py with --input-image storefront.jpg, --prompt 'Change ONLY: sky to dramatic storm clouds. Keep identical: subject, composition, lighting, color palette, background, and overall style.', --resolution 1K for a draft
- 3Review the draft output path with the user
- 4Re-run with adjusted prompt if needed, then finalize at 2K or 4K once prompt is confirmed
Edited PNG saved to current directory, e.g. 2026-03-16-11-45-00-stormy-storefront.png
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.