OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Browser Agents

Skills — browser-act

Skills — browser-act is a public MIT-licensed Python repository that provides a browser automation CLI for AI agents. It combines local or cloud-controlled real-browser execution with web extraction, authenticated account workflows, parallel task handling and independent browser sessions.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Skills — browser-act — OpenViST project profile in Browser Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use browser-act when an AI agent must interact with websites through real browser sessions, especially for extraction, authenticated workflows or concurrent account-specific tasks. Choose its local mode for local Chrome state and direct agent control, or its cloud mode for managed execution.

Who it is for

  • Developers building AI-agent workflows that need real-browser control.
  • Teams extracting web data or automating account-based browser tasks.
  • Builders running concurrent jobs or separate browser sessions for multiple accounts.
  • Users who need local Chrome login-state reuse or cloud-managed execution.

Documented uses

  • Extract data from websites, including protected pages.
  • Automate authenticated web workflows.
  • Run multiple browser tasks concurrently without mixing session state.
  • Keep separate account workflows in isolated sessions.
  • Hand blocked browser steps to a human and resume agent control.
Before you choose

Requirements & limits

The repository is public and MIT-licensed.

  • Local mode can reuse Chrome login state through profile import or CDP attachment.
  • Cloud mode builds and runs a reusable scraping Bot from a described website, filter and field request.
  • Stealth modes distinguish fresh privacy sessions from stable identities for logged-in parallel browsing.
  • Indexed text state and indexed interaction are designed to reduce the need for DOM parsing.
  • Concurrent sessions use explicit naming and ownership to reduce agent conflicts.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run an indexed local browser workflow

Open a browser session, inspect its indexed state and interact with a page through the documented CLI pattern.

Before you start

  • browser-act installed in the local environment.
  • A target page URL.
  • A named session for the workflow.

Steps

  1. Open the target URL in a named browser-act session.

  2. Request the session state to obtain indexed interactive elements.

  3. Select an element by its returned index.

  4. Provide input using the indexed interaction pattern.

Expected outcome

The agent can progress through a browser workflow using compact indexed state and actions.

Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
2 October 2026
Original website title
browser-act/skills · Official homepage title · retained in its source language
How we review sourcesSuggest a correction