HomeProjectsArticlesPK BoardAbout
Explore Projects

Page Agent

Active
GitHub TypeScript MIT

Description

Page Agent is a JavaScript in-page GUI agent by Alibaba that controls web interfaces with natural language, enabling automated form filling, page navigation, and element interaction.

Key Features

  • In-page JavaScript GUI agent — no browser extension, Python, or headless browser required
  • Text-based DOM manipulation without screenshots or multimodal LLMs
  • Bring your own LLM — works with any OpenAI-compatible API
  • Optional Chrome extension for multi-page agent tasks across browser tabs
  • MCP Server (Beta) to control the agent externally via MCP protocol
  • One-line CDN integration for instant setup and evaluation

Use Cases

💡 SaaS AI copilot embedded directly in web products with zero backend rewrite
💡 Smart form filling for ERP, CRM, and admin systems — 20-click workflows in one sentence
💡 Accessibility layer for any web app via natural language voice commands
💡 Multi-page browser automation across tabs with Chrome extension
💡 MCP-enabled agent clients controlling browser interactions programmatically

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (28.9k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

browser agent typescript automation web

Categories

🌐 Browser Agent

Quick Start

npm install page-agent && import { PageAgent } from 'page-agent'; const agent = new PageAgent({ model: 'qwen3.5-plus', baseURL: 'https://dashscope.aliyuncs.com/compatible-mode/v1', apiKey: 'YOUR_API_KEY' }); await agent.execute('Click the login button')
Visit GitHub Visit Website

Related Projects

Open Operator

2.0k · TypeScript
Stale C

An open-source template for building web agents with Stagehand on Browserbase, providing serverless browser automation for AI agents to safely execute web tasks in the cloud.

browseragenttypescript +2
  • · Open-source web agent template built on Stagehand and Browserbase — converts natural language intent into headless browser operations via a simple agent loop
  • · Serverless browser automation powered by Browserbase — AI agents execute web tasks in isolated cloud browser sessions without managing infrastructure
  • · Stagehand DOM manipulation and state management for precise interaction with web elements and structured data extraction
View on GitHub

BrowserOS

13.5k · TypeScript
Active A+

The open-source Agentic browser that transforms your browser into an AI-powered operating system. Alternative to ChatGPT Atlas, Perplexity Comet, and Dia.

browseragenttypescript +3
  • · AI-native Chromium fork: 53+ browser automation tools for navigation, clicking, typing, and data extraction via natural language
  • · MCP server integration: control the browser from Claude Code, Gemini CLI, or any MCP-capable client
  • · Visual workflow builder: create repeatable browser automations with a drag-and-drop graph interface
View full details → View on GitHub

ScaleCUA

1.1k · Python
Stale C

Open-sourced computer use agents that can operate on cross-platform environments including Windows, macOS, Ubuntu, and Android. ICLR 2026 Oral paper project.

browserpythonagent +1
  • · Cross-platform GUI operation: supports Windows, macOS, Ubuntu, Android, and Web environments
  • · Large-scale training data: operation trajectories across 6 operating systems and 3 task domains
  • · State-of-the-art performance: MMBench-GUI L1-Hard 94.4%, OSWorld-G 60.6%, WebArena-Lite-v2 47.4%
View on GitHub