🚀 POSA AI - Open Source Browser Agent

Control browsers with natural language - Built with Flutter, Python, and AI

🌐 Website • 📚 Documentation • 💬 Discord • 🏢 Nexios

📥 Download POSA AI

Version: 1.0.5 | Size: 150 MB | Updated: June 17, 2025

📸 Screenshots

Initialization

*Please wait while the Posa AI backend downloads all required dependencies.*

Main Dashboard

*Clean and intuitive interface for browser automation*

Agent in Action

*Real-time visual feedback as AI navigates the web*

🎯 About

POSA AI is an open-source browser automation agent that lets you control web browsers using natural language. Built with Flutter for cross-platform desktop support and Python's browser-use framework for intelligent web automation.

Why POSA AI?

🤖 AI-Powered - Uses GPT-4, Claude 3, Gemini, and DeepSeek
🌐 Cross-Platform - Windows, macOS, Linux support via Flutter
🔓 Open Source - Fully transparent, community-driven development
🎯 Real-Time - Watch AI navigate with visual indicators
🛠️ Extensible - Built on open standards (Socket.IO, Playwright)
📊 95% Success Rate - Proven reliability in browser automation
⏰ 24/7 Availability - Always ready for your automation needs

✨ Features

Feature	Description
🗣️ Natural Language Commands	Just describe what you want in plain English
👁️ Visual Feedback	See every click and action in real-time with indicators
🧠 Multi-Model AI Support	Choose between GPT-4, Claude 3, Gemini, DeepSeek
📑 Multi-Tab Management	Handle complex workflows across multiple tabs
🔄 Real-Time Logs	Stream execution logs live as tasks run
💾 Secure Local Storage	API keys stored securely per model
🌍 Multilingual	Full support for Arabic and English
🚀 Auto-Setup	Automatic Python backend and dependency installation

Technical Capabilities

✅ Automatic Python backend setup with progress tracking
✅ Playwright Chromium auto-installation
✅ Socket.IO real-time bidirectional communication
✅ Supabase authentication & user management
✅ Cross-platform desktop builds (Windows/macOS/Linux)
✅ Intelligent DOM + Visual parsing for accuracy
✅ Task history and quota management

🚀 Quick Start

System Requirements

Component	Requirement
OS	Windows 10/11 (64-bit), macOS 10.15+, or Linux
Memory	4GB RAM minimum, 8GB recommended
Storage	3GB free disk space
Browser	Chrome or Edge for automation
Python	3.10+ (auto-detected)

📥 Download & Install

Windows (Recommended)

File: POSA.AI.1.5.SETUP.exe | Size: 150 MB
SHA256: 8e7e9b3e954f3ef00d65b7038ec812dfaf49d3d5cde76a1b06904581d5d1138b

Installation Steps

Download the installer from the button above
Run POSA.AI.1.5.SETUP.exe as administrator
Follow the installation wizard
Launch POSA AI from your desktop or start menu
Configure your API keys in settings
Start automating! 🎉

Alternative Downloads

Platform	Link	Status
🍎 macOS	Coming Soon	In Development
🐧 Linux	Coming Soon	In Development
🤖 Android	Future Release	Planned
📱 iOS	Future Release	Planned

Other Installation Methods

🔧 Build from Source (Advanced)

# Clone the repository
git clone https://github.com/Hmida71/POSA_AI.git
cd POSA_AI

# Install Flutter dependencies
flutter pub get

# Install Python dependencies
cd backend
pip install -r requirements.txt
python -m playwright install chromium
cd ..

# Run the app
flutter run -d windows  # or macos, linux

See Building from Source for detailed instructions.

🎬 Demo

Example Commands You Can Try

"Go to Amazon and search for Samsung Galaxy S25"
"Find the best laptop under $1000 on eBay"
"Compare prices for iPhone 15 across 3 websites"
"Book a flight to NYC for next weekend"
"Fill out this form with my information"
"Download the top 5 research papers on AI from Google Scholar"
"Monitor price changes for this product"

Supported Use Cases

🛒 E-commerce Automation - Price monitoring, inventory checks, automated purchasing
📊 Data Collection - Web scraping, research automation, content aggregation
🧪 Testing & QA - Automated browser testing, regression testing
📈 Market Research - Competitor analysis, trend monitoring
⚡ Productivity - Repetitive task automation, form filling
🔍 Content Discovery - Information gathering and analysis

🏗️ Architecture

📐 System Architecture Diagram

┌─────────────────────────────────┐
│     Flutter Desktop App         │
│  ┌───────────────────────────┐  │
│  │  UI Layer                 │  │
│  │  - browser_agent_screen   │  │
│  │  - Model selection UI     │  │
│  │  - Real-time logs display │  │
│  └───────────────────────────┘  │
│  ┌───────────────────────────┐  │
│  │  State Management         │  │
│  │  - Provider pattern       │  │
│  │  - backend_manage.dart    │  │
│  └───────────────────────────┘  │
│  ┌───────────────────────────┐  │
│  │  Services                 │  │
│  │  - Socket.IO Client       │  │
│  │  - SharedPreferences      │  │
│  └───────────────────────────┘  │
└─────────────────────────────────┘
            │
            │ Socket.IO (localhost:5000)
            ▼
┌─────────────────────────────────┐
│     Python Backend              │
│  ┌───────────────────────────┐  │
│  │  Socket.IO Server         │  │
│  │  - app.py / app_secure.py │  │
│  │  - Task orchestration     │  │
│  └───────────────────────────┘  │
│  ┌───────────────────────────┐  │
│  │  Browser Automation       │  │
│  │  - browser-use framework  │  │
│  │  - Playwright             │  │
│  │  - Chromium control       │  │
│  └───────────────────────────┘  │
│  ┌───────────────────────────┐  │
│  │  AI Model Integration     │  │
│  │  - OpenAI API (GPT)       │  │
│  │  - Anthropic API (Claude) │  │
│  │  - Google API (Gemini)    │  │
│  │  - DeepSeek API           │  │
│  └───────────────────────────┘  │
└─────────────────────────────────┘
            │
            ▼
┌─────────────────────────────────┐
│     Supabase Backend            │
│  - Authentication               │
│  - User management              │
│  - Plans & quotas               │
│  - Task tracking                │
└─────────────────────────────────┘

Technology Stack

Frontend:

Flutter 3.x (Desktop support)
Provider (State management)
SharedPreferences (Local storage)
Socket.IO Client

Backend:

Python 3.10+
Socket.IO Server
Playwright (Browser automation)
browser-use framework
AI Model APIs (OpenAI, Anthropic, Google, DeepSeek)

Database:

Supabase (Auth, User management, Plans & Quotas)

💻 Usage

Basic Workflow

Launch POSA AI and sign in with Supabase Auth
Navigate to AI Browser Agent screen
Enter your task (e.g., "Search for Python tutorials on YouTube")
Select AI model and version:
- Gemini - gemini-2.0-flash (default), gemini-1.5-pro
- GPT - gpt-4, gpt-3.5-turbo
- Claude - claude-3-opus, claude-3-sonnet
- DeepSeek - Various versions
Provide API key (stored locally per model)
Toggle options:
- ☑️ Keep browser open after completion
- ☑️ Enable verbose logging
Click Start - Watch real-time execution
View results in the results panel

Advanced Features

Auto Backend Setup - First run installs all dependencies automatically
Model-Specific API Keys - Each model has separate key storage
Multi-Language Support - English and Arabic interfaces
Task History - Review past automations
Quota Management - Track usage limits per plan

🔑 API Keys

Get Your API Keys

Provider	Models	Get Key	Free Tier
🤖 OpenAI	GPT-4, GPT-3.5	Get Key	$5 credit
🧠 Anthropic	Claude 3 Opus, Sonnet	Get Key	Limited
🔮 Google	Gemini 2.0, 1.5 Pro	Get Key	Free
🚀 DeepSeek	DeepSeek V2	Get Key	Free tier

API Key Setup

Click the Settings icon in POSA AI
Select your preferred AI model
Paste your API key
Keys are stored locally and securely
Start automating!

🛠️ Building from Source

Prerequisites

# Install Flutter (https://flutter.dev/docs/get-started/install)
flutter doctor

# Install Python 3.10+
python --version

# Clone the repository
git clone https://github.com/Hmida71/POSA_AI.git
cd POSA_AI

Build Instructions

🪟 Windows

# Install dependencies
flutter pub get
cd backend && pip install -r requirements.txt && cd ..

# Build
flutter build windows --release

# Output: build/windows/runner/Release/

🍎 macOS

# Install dependencies
flutter pub get
cd backend && pip install -r requirements.txt && cd ..

# Build
flutter build macos --release

# Output: build/macos/Build/Products/Release/

🐧 Linux

# Install dependencies
flutter pub get
cd backend && pip install -r requirements.txt && cd ..

# Build
flutter build linux --release

# Output: build/linux/x64/release/bundle/

Create Windows Installer

# 1. Build the Windows release
flutter build windows --release

# 2. Copy backend files
cp -r backend/ installer_source/app_files/backend/

# 3. Open setup_script.iss in Inno Setup Compiler
# 4. Compile to create POSA-AI-Setup.exe

🤝 Contributing

We ❤️ contributions! Join our community of developers.

How to Contribute

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

Areas We Need Help

🌍 Translations - Add more languages
🎨 UI/UX - Improve the interface
🐛 Bug Fixes - Find and fix issues
📚 Documentation - Improve guides
🧪 Testing - Write more tests
🚀 Features - Implement roadmap items

View Contributing Guide →

👥 Community

Join Our Growing Community!

💬 Discord - Chat with developers and get support
🐛 Issues - Report bugs and request features
💡 Discussions - Share ideas and ask questions
🌟 Star us - Show your support!

🗺️ Roadmap

✅ Completed

Multi-model AI support (GPT, Claude, Gemini, DeepSeek)
Real-time browser control with visual indicators
Cross-platform desktop support (Windows/macOS/Linux)
Visual feedback system
Supabase authentication
Multi-tab management
Real-time log streaming

🚧 In Progress (Q1 2025)

Mobile support (Android/iOS)
Chrome extension
Cloud deployment option
Plugin system architecture

🔮 Planned (Q2-Q3 2025)

Vote on features →

🐛 Troubleshooting

Python not found

# Windows - Add Python to PATH
setx PATH "%PATH%;C:\Python310"

# Linux/macOS
export PATH="/usr/local/bin/python3:$PATH"

# Verify
python --version

Playwright Chromium missing

# Auto-installed on first run, but if needed:
python -m playwright install chromium

# With dependencies
python -m playwright install --with-deps chromium

Port 5000 already in use

# Windows
netstat -ano | findstr :5000
taskkill /PID <PID> /F

# Linux/macOS
lsof -ti:5000 | xargs kill -9

Socket.IO connection failed

Verify backend is running (check for "Server running on port 5000")
Ensure localhost:5000 is accessible
Check firewall/antivirus settings
Try restarting the application

📄 License

This project is licensed under the MIT License

Read Full License →

🙏 Acknowledgments

Created by: Hmida71 & abenkoula71
Company: Nexios
Built on browser-use framework
Powered by Flutter
Browser automation via Playwright

👨‍💻 Maintainers

Maintainer	Role	GitHub
Hmida71	Creator & Nexios Founder	@Hmida71
abenkoula71	Creator & Nexios Founder	@abenkoula71

Developed by Nexios

📊 Project Stats

⭐ Star us on GitHub — it motivates us a lot! ⭐

Made with ❤️ by the POSA AI community

🌐 Website • 🏢 Nexios • 📱 GitHub • 💬 Discord

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
.dart_tool		.dart_tool
.idea		.idea
android		android
assets		assets
backend		backend
build		build
installer_source		installer_source
ios		ios
lib		lib
linux		linux
macos		macos
web		web
website posa ai		website posa ai
windows		windows
.flutter-plugins-dependencies		.flutter-plugins-dependencies
.gitignore		.gitignore
.metadata		.metadata
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.md		README.md
analysis_options.yaml		analysis_options.yaml
posa_ai_app.iml		posa_ai_app.iml
project_structure.txt		project_structure.txt
pubspec.lock		pubspec.lock
pubspec.yaml		pubspec.yaml
template_app.iml		template_app.iml

License

Hmida71/POSA_AI

Folders and files

Latest commit

History

Repository files navigation