Claude-skill-registry create-icon
install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/create-icon" ~/.claude/skills/majiayu000-claude-skill-registry-create-icon && rm -rf "$T"
manifest:
skills/data/create-icon/SKILL.mdsource content
create-icon
Centralized utility for creating high-quality Stream Deck icons with consistent aesthetics.
Features
- Automated Fetching: Uses
to find and download icons from Flaticon or other free providers.surf - Pre-flight Standardization: Automatically resizes to 72x72 and ensures PNG format.
- State Generation:
: High-contrast white.inactive
: Vibrant "active" color (default: cyan/yellow).active
- Collaborative Workflow: Search for multiple candidates and pick the best one via personal interview.
Usage
# Fetch a NEW icon by keyword ./run.sh fetch "folder" --name "my_folder" # Generate states for an EXISTING icon ./run.sh generate "/path/to/icon.png" --name "my_icon" # COLLABORATIVE search and selection ./run.sh collaborative "camera" --name "my_camera" # DIRECT AI-driven creation ./run.sh create "terminal prompt inside a rounded rectangle" --name "my_terminal"
How It Works
- Fetch: Uses
to search for icons, picks the first result, and downloads it.surf - Process: Uses ImageMagick (
) or Pillow to:convert- Resize to 72x72.
- Use
or tinting to create the two required states.+level-colors
- Output: Saves to the local
directory of the calling project.icon/
Commands
| Command | Description |
|---|---|
| Search web and download icon |
| Create states from existing icon |
| Interactive search and selection |
| AI-driven icon generation |
Output Structure
icon/ ├── my_icon_inactive.png # White/gray (72x72) ├── my_icon_active.png # Colored state (72x72) └── my_icon_original.png # Source file
Dependencies
skill for web fetchingsurf- ImageMagick or Pillow for image processing