Skill

screenshot-desktop

Capture desktop, window, or screen region screenshots using native OS tools

Install

npx claudepluginhub pforret/cli-skills --plugin screenshot-desktop

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Capture desktop/screen screenshots using native OS tools. Supports macOS, Linux, and Windows.

SKILL.md

Similar Skills

design-system

Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.

team-skills-platform

163.7k

ui-demo

Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.

team-skills-platform

163.7k

kotlin-patterns

Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.

team-skills-platform

163.7k

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitApr 2, 2026

Actions

View Source View Plugin View on GitHub View README

screenshot-desktop

Capture desktop/screen screenshots using native OS tools. Supports macOS, Linux, and Windows.

Based on openai/skills/screenshot (Apache 2.0).

Plugin root: this SKILL.md is at skills/screenshot-desktop/SKILL.md inside the plugin. All paths below are relative to the plugin root (two directories up from this file). Resolve them from this file's location before running.

Setup

Run once before first use:

bash <plugin-root>/setup.sh

On macOS, also run the permission preflight:

bash <plugin-root>/scripts/ensure_macos_permissions.sh

Usage

python3 <plugin-root>/scripts/take_screenshot.py [options]

Key options

Option	Description
`--path FILE`	Output file path or directory
`--mode default\|temp`	Save to OS default location or temp dir
`--format png\|jpg\|bmp`	Image format (default: png)
`--region x,y,w,h`	Capture a pixel region
`--active-window`	Capture the focused window only
`--window-id ID`	Capture a specific window by ID
`--app NAME`	macOS: capture windows for an app
`--window-name TEXT`	macOS: match window title substring
`--list-windows`	macOS: list matching window IDs
`--interactive`	Interactive selection (where supported)

Examples

python3 <plugin-root>/scripts/take_screenshot.py --path screen.png
python3 <plugin-root>/scripts/take_screenshot.py --path window.png --active-window
python3 <plugin-root>/scripts/take_screenshot.py --path region.png --region 100,200,800,600
python3 <plugin-root>/scripts/take_screenshot.py --path safari.png --app Safari
python3 <plugin-root>/scripts/take_screenshot.py --mode temp

Windows

powershell -ExecutionPolicy Bypass -File <plugin-root>/scripts/take_screenshot.ps1 -Path screen.png

How Claude Code should use this skill

When the user asks to capture their screen, a window, or an app:

Determine PLUGIN_ROOT by resolving ../.. from this SKILL.md file's path
Ensure macOS permissions if on macOS (bash $PLUGIN_ROOT/scripts/ensure_macos_permissions.sh)
Determine output path (default to screenshot.png in cwd if not specified)
Build the python3 $PLUGIN_ROOT/scripts/take_screenshot.py command from user-specified options
Run via Bash and report the saved file path

Do NOT use this skill for web page URLs — use screenshot-web instead.

Platform details

macOS: Uses screencapture + Swift helpers. Full-screen captures produce one file per display.
Linux: Auto-selects from scrot, gnome-screenshot, or ImageMagick import.
Windows: Uses PowerShell with .NET System.Drawing/Windows.Forms APIs.