Skip to content

jonymusky/ai-browser-automation

Repository files navigation

AI Browser Automation

Tests Lint npm version NPM Downloads

A TypeScript package that combines Selenium WebDriver with AI capabilities for intelligent browser automation.

Overview

This package provides an easy way to automate browser interactions using AI capabilities. It supports multiple LLM providers (OpenAI, Ollama) and can automatically detect and interact with web elements.

For a detailed list of changes and versions, see our Changelog.

Features

  • Selenium-based browser automation
  • AI-powered element detection and interaction
  • Support for multiple LLM providers (OpenAI, Ollama, Grok)
  • Screenshot capture capability
  • Flexible step-based automation configuration

Installation

npm install ai-browser-automation
# or
pnpm add ai-browser-automation

Usage

import { AiBrowserAutomation } from 'ai-browser-automation';

const automation = new AiBrowserAutomation({
  llmProvider: 'OpenAI',
  apiKey: 'your-api-key',
  browser: 'chrome',
  headless: true
});

const steps = [
  {
    action: 'navigate',
    description: 'Go to Google',
    url: 'https://google.com'
  },
  {
    action: 'write',
    description: 'Search for something',
    solve_with_ai: true
  }
];

const result = await automation.execute(steps);

Examples

For detailed examples with screenshots and execution outputs, see our Examples Documentation.

Running an example

You can find example scripts in the examples directory. To run a specific example:

pnpm run-example google

About

AI-powered browser automation using Selenium WebDriver

Topics

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published