Skip to content
B

browser-act

Browser automation CLI for AI agents

by browser-actFree

6.1K
#2401in AI Skills
0In stores

Description

browser-act is an agent skill from the BrowserAct skills collection, listed on skills.sh. It is a browser automation command-line tool for A…

Highlights

  • Browser automation, extraction and screenshots
  • Several browsers in parallel with proxies
  • Confirmation gate before sensitive actions
  • Headed mode with remote assistance

Ratings & Reviews

No ratings yet

Be the first to rate browser-act.

People also visit

A

agent-browser

Fast, persistent browser automation

AI Skills
Get
H

hyperframes

HTML-based video composition

AI Skills
Get
P

ponytail

Lazy senior dev mode: the smallest change

AI Skills
Get
G

gstack

Fast headless browser for QA testing, deployment verification

AI Skills
Get
C

caveman

Ultra-compressed communication mode cutting token usage ~75% while preserving

AI Skills
Get
G

graphify

Any question about a codebase, its architecture, file relationships

AI Skills
Get

browser-act FAQ

What does browser-act do?

It gives an AI agent a browser automation command-line tool for navigation, extraction, screenshots and form filling.

How do I install it?

Run: npx skills add https://github.com/browser-act/skills --skill browser-act. The CLI itself is installed with uv tool install browser-act-cli --python 3.12.

Will it act without asking?

No. A confirmation gate needs explicit user approval before a browser is created or deleted, and before sensitive operations.

Which tools does it work with?

Cursor, Claude Code, Codex and Windsurf, according to the page.

Has it been checked for security?

skills.sh lists three audits: Gen Agent Trust Hub Pass, Socket Warn and Snyk Warn. A Warn or Fail is worth reading before you install.

What are the best browser-act alternatives?

Popular alternatives to browser-act include agent-browser, hyperframes and ponytail. See all browser-act alternatives

Developer Stuff

Badges