Skip to content
Get packs

BrowserWing Browser Control

Lets an agent drive a local browser through an HTTP API: navigate, click, type, and extract page data.

Source Download skill
Repository
MemTensor/MemOS
Stars
12,226
Licence
Apache-2.0
Files
1 file

What it does

Once installed, the agent can operate a web browser for you: open pages, click buttons, fill in forms, read text from pages, and take screenshots. It does this by sending commands to a BrowserWing service running on your own computer at localhost:8080, which controls the browser. The skill teaches the agent which commands exist, how to identify page elements, and a step-by-step workflow to follow.

Great for

  • Developers who want an agent to automate repetitive browser tasks on sites they control or have permission to use
  • Researchers or analysts who need structured data pulled from web pages into a summary
  • People already running the BrowserWing service locally who want an agent to use it

Skip it if

  • Users who do not already run a BrowserWing server; the skill only works against a local API at localhost:8080 and will fail without it
  • Logging into accounts, making purchases or sending messages on your behalf without reviewing each step, because the agent acts on real sites with real effects and the skill has no confirmation step
  • Sensitive sites like banking or email, since the agent may see and type credentials and page content that is not meant to leave your machine

Try asking

  • “Search example.com for 'laptop' and list the first five results with prices and links”
  • “Open a login page and report which interactive elements are available on it”
  • “Take a screenshot of a product page and extract all item names and prices”
  • “Fill out a contact form on a site I own using the values I provide”
Step-by-step guide: how to use BrowserWing Browser ControlInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    browserwing-executor.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip browserwing-executor.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip browserwing-executor.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

  • Tell the agent which site and goal you want, and ask it to confirm before any click that submits forms, deletes data or spends money
  • Make sure the BrowserWing service is running locally and the API key is configured before asking for browser tasks
  • Ask the agent to call the help endpoint first so it uses the commands your installed version actually supports

What's inside

  • SKILL.md

Redistributed unmodified from MemTensor/MemOS under the Apache-2.0 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.