← Back to Skills Marketplace
ouyangabel

Linux Desktop Control

by ouyangAbel · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
1594
Downloads
0
Stars
6
Active Installs
1
Versions
Install in OpenClaw
/install linux-desktop
Description
Linux desktop automation and control. Use when: (1) taking screenshots of the screen or windows, (2) controlling mouse and keyboard, (3) managing windows, (4...
README (SKILL.md)

Linux Desktop Control

Automate and control your Linux desktop using command-line tools. Capture screenshots, control mouse/keyboard, and manage windows.

When to Use

  • Take screenshots of screen or specific windows
  • Automate mouse movements and clicks
  • Simulate keyboard input
  • Manage and interact with windows
  • Build desktop automation workflows

Prerequisites

Install required tools:

sudo apt-get install scrot xdotool xclip x11-utils x11-apps

Quick Start

Take Screenshot

python3 scripts/linux-desktop.py screenshot
# Output: ~/screenshot_20240224_203901.png

List Windows

python3 scripts/linux-desktop.py list

Move Mouse and Click

python3 scripts/linux-desktop.py move 500 300
python3 scripts/linux-desktop.py click

Type Text

python3 scripts/linux-desktop.py type "Hello World"

Commands

screenshot [path]

Capture a screenshot of the entire screen.

Examples:

# Save to default location (/tmp/screenshot_YYYYMMDD_HHMMSS.png)
python3 scripts/linux-desktop.py screenshot

# Save to custom path
python3 scripts/linux-desktop.py screenshot ~/desktop.png

window [window_id] [path]

Capture a screenshot of a specific window.

Examples:

# Screenshot active window
python3 scripts/linux-desktop.py window

# Screenshot specific window
python3 scripts/linux-desktop.py window 0x12345678 ~/window.png

active

Get information about the currently active window.

python3 scripts/linux-desktop.py active
# Output: 🖥️ Active Window
#         ID: 0x12345678
#         Title: Terminal

list

List all visible windows.

python3 scripts/linux-desktop.py list
# Output: 🪟 Found 5 windows:
#         1. 0x12345678 - Terminal
#         2. 0x87654321 - Chrome

move \x3Cx> \x3Cy>

Move mouse cursor to specified coordinates.

Examples:

python3 scripts/linux-desktop.py move 100 200
# Moves mouse to (100, 200)

python3 scripts/linux-desktop.py move 500 300
# Moves mouse to center of 1000x600 area

click [button]

Click mouse button at current cursor position.

Button values:

  • 1 - Left button (default)
  • 2 - Middle button
  • 3 - Right button

Examples:

python3 scripts/linux-desktop.py click
# Left click

python3 scripts/linux-desktop.py click 3
# Right click

type \x3Ctext>

Type text at current cursor position (must be in focused window).

Examples:

python3 scripts/linux-desktop.py type "Hello World"

python3 scripts/linux-desktop.py type "ls -la"

python3 scripts/linux-desktop.py type "sudo apt update"

key \x3Ckeyspec>

Press keyboard keys.

Common keys:

  • Return - Enter key
  • Escape - Escape key
  • Tab - Tab key
  • BackSpace - Backspace
  • Delete - Delete
  • Up, Down, Left, Right - Arrow keys
  • Home, End, Page_Up, Page_Down
  • F1 through F12
  • Ctrl+c, Ctrl+v, Ctrl+a, Ctrl+z - Key combinations

Examples:

python3 scripts/linux-desktop.py key Return

python3 scripts/linux-desktop.py key Escape

python3 scripts/linux-desktop.py key Ctrl+a

python3 scripts/linux-desktop.py key F5

screen

Get screen information.

python3 scripts/linux-desktop.py screen
# Output: 🖥️ Screen Info
#         Resolution: 1920x1080

Automation Examples

Basic Automation

# Move mouse, click, type, and press enter
python3 scripts/linux-desktop.py move 100 100
python3 scripts/linux-desktop.py click
python3 scripts/linux-desktop.py type "ls -la"
python3 scripts/linux-desktop.py key Return

Web Search Automation

# Open browser, navigate to Google, search
python3 scripts/linux-desktop.py move 100 50
python3 scripts/linux-desktop.py click
python3 scripts/linux-desktop.py type "https://www.google.com"
python3 scripts/linux-desktop.py key Return
sleep 2
python3 scripts/linux-desktop.py type "how to make money online"
python3 scripts/linux-desktop.py key Return

Screenshot Workflow

# Take screenshot before and after action
python3 scripts/linux-desktop.py screenshot /tmp/before.png
python3 scripts/linux-desktop.py key F5  # Refresh
sleep 1
python3 scripts/linux-desktop.py screenshot /tmp/after.png

Tips

  • Always check the active window before typing
  • Use sleep commands between actions for reliability
  • Take screenshots to verify state changes
  • Test commands one by one before building complex workflows
  • Use window list to find specific window IDs for targeting

Troubleshooting

"Command not found" errors:

sudo apt-get install scrot xdotool xclip x11-utils x11-apps

Permission denied:

  • Ensure you're running in a graphical session (X11 or Wayland)
  • Some actions require focus on the target window

Mouse doesn't move:

  • Check if another application is grabbing the mouse
  • Try moving the mouse manually to see if it's responsive

Security Notes

  • This skill can control your desktop - use with caution
  • Don't automate sensitive actions without verification
  • Always review automation scripts before running
Usage Guidance
This skill appears to do what it says (control and automate a Linux desktop), but review and be cautious before installing or granting it autonomous use. Specific recommendations: - Don't run it with untrusted inputs: the Python script constructs shell commands and calls subprocess.run(..., shell=True) — text you pass (for typing, window IDs, paths) can lead to command injection. Treat it as unsafe for use with untrusted data. - If you will use it, prefer invoking it manually rather than allowing autonomous agents to call it. Autonomous agents could type commands or click things that perform sensitive actions. - Consider patching before use: replace shell=True invocations with subprocess.run([...], shell=False) and properly validate/escape user inputs (window IDs, file paths, typed text). Ensure file paths are sanitized and text is passed as program arguments rather than inline shell strings. - Install required system packages listed in SKILL.md and also ImageMagick (for 'convert'), and run the tool in a non-privileged graphical session. Avoid running in an environment where sensitive windows (password managers, terminals with secrets) are focused. - If you want the safety of this skill but lack the skills to audit/fix it, do not enable autonomous invocation and ask the publisher for a fixed version that avoids shell=True and documents all native dependencies.
Capability Analysis
Type: OpenClaw Skill Name: linux-desktop Version: 1.0.0 The skill bundle provides legitimate Linux desktop automation capabilities. However, the `scripts/linux-desktop.py` script uses `subprocess.run(..., shell=True)` and constructs several commands (e.g., `key`, `window`, `screenshot`, `type`) with user-provided arguments without adequate sanitization or quoting. This creates critical shell injection vulnerabilities (Remote Code Execution) where a malicious prompt or user input could execute arbitrary commands on the host system. There is no evidence of intentional malicious behavior, but the severe vulnerabilities make it suspicious.
Capability Assessment
✓ Purpose & Capability
Name/description, SKILL.md, and the included script all describe Linux desktop automation (screenshots, mouse/keyboard, window management). Required tools called by the script (scrot, xdotool, xwd/convert, xdpyinfo) align with that purpose. SKILL.md lists most prerequisites but omits the ImageMagick 'convert' dependency used in the script.
⚠ Instruction Scope
SKILL.md instructs running the bundled Python script and installing packages — that's within scope. However the script builds shell commands using user-provided values and calls subprocess.run(..., shell=True) without sanitization (e.g., type, window id, file path), creating straightforward command-injection vectors. The instructions also encourage typing arbitrary text (including commands like 'sudo apt update'), which combined with autonomous or scripted use increases risk of unintended privileged actions.
✓ Install Mechanism
No install spec (instruction-only) — lowest install risk. The skill does include a Python script that invokes system binaries; nothing in the package automatically downloads remote code or writes installers to disk.
✓ Credentials
The skill requests no environment variables, credentials, or config paths — appropriate for a local desktop automation tool.
✓ Persistence & Privilege
always:false and user-invocable:true (normal). disable-model-invocation is false (default) — the skill can be invoked autonomously, which is platform normal but increases the practical risk because the skill controls the desktop; there is no request to modify other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install linux-desktop
  3. After installation, invoke the skill by name or use /linux-desktop
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: Linux desktop automation with screenshot, mouse, keyboard, and window control
Metadata
Slug linux-desktop
Version 1.0.0
License —
All-time Installs 7
Active Installs 6
Total Versions 1
Frequently Asked Questions

What is Linux Desktop Control?

Linux desktop automation and control. Use when: (1) taking screenshots of the screen or windows, (2) controlling mouse and keyboard, (3) managing windows, (4... It is an AI Agent Skill for Claude Code / OpenClaw, with 1594 downloads so far.

How do I install Linux Desktop Control?

Run "/install linux-desktop" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Linux Desktop Control free?

Yes, Linux Desktop Control is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Linux Desktop Control support?

Linux Desktop Control is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Linux Desktop Control?

It is built and maintained by ouyangAbel (@ouyangabel); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191113 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments