Skip to content
Get packs

Browser Automation via Hive Browser CLI

Drive the user's running Chrome from the terminal using hive-browser, with screenshot-based clicking.

Source Download skill
Repository
aden-hive/hive
Stars
11,132
Licence
Apache-2.0
Files
1 file

What it does

This skill teaches an AI agent how to control a real Chrome browser that the user already has open, using a command-line tool called hive-browser that connects through a browser extension. The agent takes screenshots, clicks and types at proportional positions on the page, and reads page elements, while following rules about not restarting or killing the browser and handling timeouts. It also covers choosing the right Chrome profile when several accounts are connected and points to related LinkedIn skills.

Great for

  • Developers building agents that must operate on websites through a logged-in Chrome session
  • Automation users who need to interact with pages whose elements are hidden in shadow DOM or hard to select
  • Teams running several agents that share one browser connection and need clear lifecycle rules

Skip it if

  • Users who want a headless or isolated browser for scraping, since this skill only attaches to the user's own Chrome and forbids launching browser processes
  • Sensitive account work without supervision, because the agent acts inside logged-in sessions and could post or message as the user
  • Environments without the Hive runtime and hive-browser CLI installed, where the instructions cannot be followed

Try asking

  • “Open my LinkedIn-connected Chrome profile acct-jpn and check the inbox for new messages”
  • “Take a screenshot of the current page and click the Send button at its visible position”
  • “Scan this dashboard page for all rows in the table without freezing the browser”
  • “Fill in the rich-text comment box on this site even though the selectors cannot find it”
Step-by-step guide: how to use Browser Automation via Hive Browser CLIInstall, 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.

    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 check the profile label echoed in the JSON output, and pass --browser-profile when more than one account is connected
  • If a command times out, ask the agent to wait and retry once with a smaller request rather than restarting the browser
  • Use screenshots with fractional coordinates for pages with shadow DOM or zoomed containers, where selectors often fail

What's inside

  • SKILL.md

Redistributed unmodified from aden-hive/hive 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.