A smart macOS utility that automates high-quality wallpaper rotation using preference-based learning and multi-source image integration.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mwc
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install mwc using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Mac Wallpaper Changer?
The Mac Wallpaper Changer is a sophisticated automation tool designed for macOS users who want a dynamic and personalized desktop environment. As a standout entry in the Openclaw Skills collection, it leverages multiple high-quality image sources like Unsplash, Bing Daily, and Picsum to ensure your background is always fresh and inspiring.
Beyond simple rotation, this skill incorporates a preference engine that learns from your ratings. By analyzing your favorites and specific keywords, it provides intelligent recommendations, transforming your desktop into a curated gallery that evolves based on your aesthetic tastes.
Mac Wallpaper Changer Use Cases
- Automatically updating your macOS desktop background daily with high-definition imagery.
- Searching and applying wallpapers based on specific moods or themes like "Shanghai rain" or "mountain sunset".
- Building a personal database of favorite wallpapers with a structured scoring system.
- Scheduling wallpaper changes using cron jobs for a seamless, hands-free experience.
How Mac Wallpaper Changer Works
- The user triggers a change request via a Python script, optionally specifying a theme or source.
- The skill fetches images following a priority hierarchy: Unsplash (High Quality/Theme) -> Bing (Daily Featured) -> Picsum (Fallback).
- The selected image is downloaded and applied to the macOS desktop, with specific compatibility handling for macOS Tahoe (26+) using WallpaperAgent refreshes.
- Users can rate the current wallpaper, saving the metadata and score into a local Parquet database.
- The recommendation engine analyzes stored preferences to suggest or automatically apply similar high-ranking images.
Mac Wallpaper Changer Setup
Ensure you have macOS and the uv Python package manager installed. Then, configure the environment and run the core scripts:
# Clone the repository and navigate to the directory
cd mac-wallpaper-changer
# Change to a random high-definition wallpaper from Unsplash
uv run scripts/change.py
# Search for a specific theme
uv run scripts/change.py --query "mountain sunset"
# Set up a daily cron job (example for 10:30 AM)
(crontab -l ; echo "30 10 * * * cd /path/to/mac-wallpaper-changer && uv run scripts/change.py") | crontab -
Mac Wallpaper Changer Data Schema & Taxonomy
The skill maintains a structured data directory at ~/wallpaper-daily/ to manage assets and learning models:
| Path | Description |
|---|---|
YYYY-MM-DD/ |
Daily folders containing downloaded wallpaper images. |
preferences.parquet |
A columnar database storing wallpaper paths, scores, and metadata for the recommendation engine. |
logs/ |
Detailed execution logs for troubleshooting and audit trails. |
references/ |
Configuration files for API sources (Bing/Unsplash) and embedding schemas. |
Mac Wallpaper Changer Advanced Features
- Multi-source prioritization with fallback logic to ensure the desktop never fails to update.
- Semantic recommendation system using embedding configurations for advanced preference matching.
- Full CLI suite for preference management, including statistical reporting and score adjustments.
- Built-in troubleshooting routines that handle macOS-specific UI refresh issues (e.g., WallpaperAgent resets).
- Location-based wallpaper configuration support for localized content.
SKILL.md
Loading
Related Openclaw Skills
AI Status Skill
A real-time monitoring skill that tracks AI provider health, model availability, pricing, and LLM benchmark leaderboards.
MediaIO AIGC Generate
A unified interface to generate and edit high-fidelity AI images and videos through the Media.io OpenAPI ecosystem.
CN Web Search
A zero-API, multi-engine search aggregator for AI agents covering Chinese social media, global finance, and technical documentation.
Ride Receipts Gateway LLM
A local, privacy-focused pipeline that fetches ride-sharing receipts from Gmail and uses an LLM-powered Gateway to extract structured data into SQLite.
Flashcard
A sophisticated spaced repetition learning tool that allows users to create decks and optimize memory retention through intelligent card scheduling.
Celebrity Hot Money Tag
An AI-powered analytical tool that identifies famous Chinese traders from market data to provide win rates, trading styles, and investment recommendations.
Featured*
