page-agent

Embed an Alibaba in-page GUI agent into web applications.

4|Updated May 14, 2026
One-click install
npx skills add https://github.com/liumeixin/hermes-skills --skill page-agent-liumeixin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: page-agent
Source: https://github.com/liumeixin/hermes-skills/tree/main/web-development/page-agent
Command: npx skills add https://github.com/liumeixin/hermes-skills --skill page-agent-liumeixin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires page-agent, and includes scripts (resource) components.

What problem does it solve?

This Skill enables the integration of the Alibaba page-agent into your web application, providing an AI-powered in-page GUI agent for end-users to interact with your application using natural language.

Core Features & Use Cases

  • AI-Powered GUI Interaction: Users can drive the UI with natural language commands like "click login, fill username as John".
  • Embeddable Solution: Can be embedded into web apps as a single <script> tag or npm package.
  • Use Case: For web developers who want to add an AI assistant to their SaaS dashboard, modernize legacy web apps, or create interactive demos.

Quick Start

Import the page-agent library and wire it up with your own LLM endpoint:

import { PageAgent } from 'page-agent'
const agent = new PageAgent({
    model: 'qwen3.5-plus',
    baseURL: 'https://dashscope.aliyuncs.com/compatible-mode/v1',
    apiKey: process.env.LLM_API_KEY,
    language: 'en-US',
})
agent.panel.show()

Frequently Asked Questions about page-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add an AI copilot to my web application?▼

To add an AI copilot to your web application, embed the page-agent library via a script tag or npm package and configure it with an OpenAI-compatible LLM endpoint. This allows users to execute natural language commands on the webpage UI.

Can I use natural language commands to drive web UI without Python?▼

Yes, you can drive web UI with natural language commands without Python or a headless browser. The page-agent executes instructions like "click login" directly in the browser by connecting to an OpenAI-compatible LLM endpoint.

Do I need a Python backend to run an in-page GUI agent?▼

No, you do not need a Python backend to run an in-page GUI agent. The page-agent operates entirely with JavaScript and an OpenAI-compatible LLM endpoint, eliminating the need for Python or headless browser dependencies.

What is the best way to modernize legacy web apps with an AI assistant?▼

The best way to modernize legacy web apps with an AI assistant is to embed an in-page GUI agent via a single script tag. This enables natural language interaction with existing UI elements without requiring complex backend rewrites.

How to embed an in-page GUI agent into a SaaS dashboard?▼

To embed an in-page GUI agent into a SaaS dashboard, import the page-agent JavaScript library and initialize it with your LLM API key and base URL. Call the show method to display the interactive AI copilot panel for users.