Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated description for improved clarity and feature detail. - Added repository link and open-source details. - Documented new file upload security restrictions: `upload_asset` now only accepts files from specific allowed directories. - Added a dedicated Security section outlining file and API key requirements. - Minor updates to usage examples to reflect new path restrictions.
What This Skill Does
Perstudio routes natural language prompts to AI models for image, video, and audio generation. Supports text-to-image, image transformation, product photography, portrait creation, upscaling, inpainting, style transfer, sticker generation, video, and text-to-speech. Token costs vary by operation type.
Automatically routes requests to the appropriate model, so you don't need to know which model handles portraits versus video versus style transfer.
When to use it
- Generate product shots from text descriptions
- Animate a still photo with subtle motion
- Convert a portrait photo into a stylized avatar
- Create stickers or icons from reference images
- Upscale low-resolution images to 4x
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: turn my photo into a watercolor painting
- 1Upload the source image with upload_asset, specifying the file path
- 2Receive the returned asset_id from the upload response
- 3Call generate_sync with the watercolor intent and the input_image_asset_id
- 4Wait 15-90 seconds for generation to complete
- 5Return the output image URL to the user
A watercolor-style version of the original photo
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.