Skip to content

Browser & Desktop Automation

Lets the agent drive web pages and Windows desktop apps (Playwright, OpenReverse) via clicks, forms, screenshots and network capture.

Stars
43,969
Licence
MIT
Files
2 files

What it does

Once installed, the agent can open websites, read the clickable elements on a page, fill forms, click buttons, take screenshots and scrape content by running a command-line browser tool (agent-browser, built on Playwright). It can also operate Windows desktop programs such as IDA Pro or x64dbg through the OpenReverse framework, either by reading standard UI controls or by visually identifying buttons on screen, and optionally observe the network traffic those apps generate.

Great for

  • Developers who want an agent to test or scrape a web page and report back
  • Reverse engineers who want an agent to drive GUI tools like IDA Pro or x64dbg
  • Analysts who need repeatable form submissions or page screenshots as evidence

Skip it if

  • People who only need static text extraction from a page, where a simple fetch is enough and this adds unnecessary setup
  • Use on accounts or sites you do not own or have permission to test, since it automates logins, captchas and form submissions and may breach terms of service or law
  • Unattended production use, because the OpenReverse desktop path is labelled experimental with heavy dependencies

Try asking

  • “Open example.com, fill in the contact form with test data, submit it and take a screenshot”
  • “Scrape the product titles and prices from this listing page and save them as a list”
  • “Use OpenReverse to load a binary in IDA Pro and export the function list”
  • “Log into my staging site and check that the dashboard loads after the wait”
Step-by-step guide: how to use Browser & Desktop AutomationInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    browser-automation.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip browser-automation.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip browser-automation.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Always let the agent take a snapshot before clicking so element references are current rather than guessed
  • Use test accounts and throwaway data when automating logins or forms
  • Check that agent-browser close ran afterwards, since the skill warns that unclosed sessions leak processes

What's inside

  • references/playwright-cheatsheet.md
  • SKILL.md

Redistributed unmodified from zhaoxuya520/reverse-skill under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.