browser-automation

Automates browser interactions with Playwright and Windows desktop applications with OpenReverse.

7|4|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/dbx0/skills --skill browser-automation-dbx0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-automation
Source: https://github.com/dbx0/skills/tree/main/skills/_tooling/browser-automation
Command: npx skills add https://github.com/dbx0/skills --skill browser-automation-dbx0

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, agent-browser, and includes scripts (resource) and references (resource) components.

What problem does it solve? Manually driving browsers or Windows GUI tools during penetration testing and reverse engineering is slow and error-prone. This Skill provides a unified entry point for automating web pages (clicking, form filling, scraping, screenshots, login flows) and desktop applications (IDA Pro, x64dbg, Wireshark) via structured UIA control or vision-driven CUA interaction. ## Core Features & Use Cases - Browser automation with Playwright/agent-browser: open pages, snapshot interactive elements, click, fill forms, submit payloads, and capture evidence during web penetration tests. - Desktop automation with OpenReverse: operate Windows GUI tools through UIA (structured controls) or CUA (vision-driven) modes, with built-in mitmproxy network observation for desktop traffic capture. - On-demand bootstrap: a setup script installs Node.js, agent-browser, and Playwright browsers automatically, with guided manual installation for OpenReverse. - Use Case: During a web pentest, automatically log into a target application, inject XSS or SQLi payloads into forms in bulk, wait for responses, and snapshot results — or automate IDA Pro to batch-load samples and export function lists while observing its network traffic. ## Quick Start Ask the agent to open a target URL with agent-browser, snapshot the interactive elements, fill in the login form, submit it, and confirm the redirect.

Frequently Asked Questions about browser-automation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser form filling and clicking with Playwright?

Use the agent-browser CLI: open the URL, run snapshot -i to get interactive element references like @e1, then use click and fill commands with those references. Always run agent-browser close afterward to avoid process leaks.

What tool should I use for Windows desktop application automation?

Use OpenReverse for Windows GUI automation. Choose UIA mode for standard Windows controls like buttons and text boxes, or CUA mode for vision-driven interaction with complex interfaces like IDA's disassembly view.

Playwright vs OpenReverse: which one for my automation task?

Use Playwright when the target is a web page inside a browser. Use OpenReverse when the target is a Windows desktop application. If a workflow needs both, such as a desktop app that loads web content, combine them.

Can I capture network traffic from a desktop application?

Yes, OpenReverse includes a network observation lane built on mitmproxy. Use proxy mode when the application supports proxy configuration, or local capture mode when it does not, and export results as HAR files.

Why does my element reference stop working in agent-browser?

Element references become invalid when the page refreshes or navigates. Run snapshot -i again after any page change to get fresh references before interacting with elements.

What are the limitations of UIA mode for desktop automation?

UIA mode only works with standard Windows controls and fails on applications with self-drawn or custom-rendered interfaces. In those cases, switch to CUA mode, which uses screenshots and coordinate-based vision interaction.