← Discover
Discover · GitHub · Page Agent↑ +163 today

Manage web interfaces with natural language

Developed by Alibaba, page-agent is an in-page graphical user interface agent that allows controlling web interfaces with natural language commands. Written in TypeScript, this tool offers a browser-based control mechanism to automate web interactions.

Updates

  • September 6, 2026: Stars 28,366 → 29,000, latest release v1.12.4 (September 6, 2026).
  • August 2, 2026: Stars 20,032 → 28,366, latest release v1.12.2 (July 16, 2026).

What you get

  • No browser extensions or Python required
  • Text-based DOM rendering instead of screenshots
  • Integration with your own AI models

Installation

Installation via NPM
npm install page-agent

If you don't write code

🤖 Paste this into your AI agent (Claude Code · Codex · Antigravity)

I want to manage interactions on the web page with natural language commands using PageAgent. I want it to perform operations such as clicking buttons on the page, filling out forms or entering data using the artificial intelligence model I have determined.

Related dictionary terms

Page AgentUser InterfaceGUIAgentArtificial Intelligence
Who it is forIt is suitable for developers who want to add artificial intelligence assistants that can be controlled with natural language to their web applications.
LicenseMIT

Links

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-26: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.