Web + Desktop Automation

Use when the user wants browser automation, web scraping, form filling, clicking, or desktop GUI automation, including mixed workflows that move between web...

This skill appears to do exactly what it claims: automate browsers and desktop GUIs. However, it is capable of controlling your mouse/keyboard, opening local files, and uploading downloaded artifacts — operations that can expose sensitive data or perform unwanted actions if misused. Before installing or running it: 1) Review the three scripts yourself (browser_runner.py, desktop_runner.py, mixed_orchestrator.py) to ensure they only perform the steps you expect. 2) Install and run dependencies in an isolated environment (virtualenv, container, or VM). 3) Do not supply real credentials or sensitive files for initial testing; use dummy accounts/data. 4) Avoid granting broad autonomous execution rights unless you trust the skill and have monitoring in place. 5) If you will run mixed flows that upload or submit files, confirm the destination endpoints are trustworthy. These precautions reduce risk but do not eliminate it.

Type: OpenClaw Skill Name: web-desktop-automation Version: 1.0.0 The skill bundle provides a legitimate framework for Robotic Process Automation (RPA) using standard libraries like Playwright and PyAutoGUI. The scripts (browser_runner.py, desktop_runner.py, and mixed_orchestrator.py) are well-structured templates for browser and desktop interaction without any signs of malicious intent, data exfiltration, or obfuscation. The SKILL.md instructions and reference documentation focus on task classification, error recovery, and safety best practices, such as stopping for user input before irreversible actions.