Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- No visible changes detected in this version. - Content, features, and documentation remain consistent with the previous release.
What This Skill Does
Generates YouTube thumbnails at 1280x720 via the each::sense API. Supports styles like reaction faces, before/after splits, listicles, gaming, and product reviews. Uses session IDs to maintain consistent channel branding across multiple requests.
Dedicated thumbnail prompting with built-in CTR-focused defaults removes the trial-and-error of adapting general image generators to YouTube's specific format and feed constraints.
When to use it
- Generating a shocked-face thumbnail for a viral reaction video
- Creating a before/after split thumbnail for a room makeover video
- Producing a consistent thumbnail series for a tech review channel
- Rapidly testing two or three thumbnail variations before publishing
- Building a gaming channel thumbnail with RGB and action composition
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Generate a shocked-face thumbnail for my reaction video with space for text on the right
- 1Constructs a POST request to https://sense.eachlabs.run/chat with the X-API-Key header
- 2Sets mode to 'max' for higher quality output
- 3Includes prompt specifying exaggerated shock expression, bright yellow and red background, and text space on the right
- 4Sends the request and streams the server-sent event response
- 5Returns the generated 1280x720 thumbnail image
A high-contrast 1280x720 thumbnail image with a shocked face and clear right-side space for text overlay
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.