title	Aivon AI
emoji	🤖
colorFrom	indigo
colorTo	blue
sdk	docker
pinned	false

Aivon AI - Browser Automation Agent 🤖

📝 Short Description

Aivon AI is an action-oriented browser assistant that autonomously executes complex web workflows, manages background media, and securely controls a live browser to perform real-world tasks.

📖 Long Description

Aivon AI is a high-performance, fully autonomous browser automation agent designed to bridge the gap between AI reasoning and direct digital action. Unlike standard chatbots that only generate text, Aivon AI actively controls a web browser on your behalf. Powered by LangChain and Groq's lightning-fast inference models, it follows strict multi-step reasoning workflows—for example, it can navigate to Google, extract relevant search snippets, synthesize the information, and then automatically compose and send an email with those findings.

It features a "Mini-Mode" floating dashboard, live in-browser status overlays, and self-healing browser management to ensure high stability even in cloud or headless environments. Whether it's playing a YouTube video in the background or performing deep data gathering, Aivon AI acts as your digital proxy.

🛠️ Technologies Used

Layer	Technology	Description
AI Engine	Groq (Llama-3.1-8B)	Provides lightning-fast, highly accurate tool-calling and strict sequential reasoning.
Orchestration	LangChain (AgentExecutor)	Manages multi-step workflows, tool bindings, and prompt chaining.
Automation	Puppeteer (Stealth Plugin)	Controls the headless/headful Chromium browser while bypassing basic bot detection.
Backend	Node.js + Express	Handles API endpoints, command parsing, and robust error management.
Frontend	Vanilla HTML/CSS/JS	Delivers a clean, responsive SaaS-style dashboard with a pop-out "Mini-Mode".

Setup

1. Clone & Install

git clone https://github.com/sunnan-riaz/AI-Agent.git
cd AI-Agent
npm install

2. Configure Environment

Edit the .env file with your credentials:

GROQ_API_KEY=gsk_your_groq_api_key_here
GMAIL_EMAIL=your_email@gmail.com
GMAIL_PASSWORD=your_app_password
PORT=3000

⚠️ Gmail App Password: Use a Gmail App Password (not your main password).
Go to Google Account → Security → App Passwords

3. Get Groq API Key

• Visit Groq Console to get a free API key.
• Create a new API key and paste it in .env

4. Run

npm start

Visit http://localhost:3000

📦 Standalone Build (.exe)

You can package the agent into a single Windows executable:

1. Run npm run build
2. Look in the dist/ folder for AI-Agent.exe
3. Double-click to run! (Ensure your .env is in the same folder)

Usage Examples

Command	Action
send email to ali@gmail.com msg Hello bro!	Sends email via Gmail
email john@example.com subject Meeting message See you at 3pm	Sends email with custom subject
check weather today	Opens Google and searches weather
what's the weather	Opens Google and searches weather

Project Structure

AI-Agent/
├── server.js          # Express server + AI parser
├── automation/
│   ├── sendEmail.js   # Gmail Puppeteer automation
│   └── checkWeather.js # Weather search automation
├── public/
│   ├── index.html     # Chat UI
│   ├── style.css      # Styles
│   └── app.js         # Frontend logic
├── .env               # Credentials (not committed)
└── package.json