Oryonix AI

Your autonomous browser copilot.

What is Oryonix AI? Oryonix AI is an open-source, autonomous browser agent that lives right inside your browser (Brave, Chrome etc.) side panel. Simply tell it what you want in plain English, and watch it navigate websites, fill forms, extract data, manage multiple tabs, and complete complex web workflows entirely on its own! Unlike cloud-based automation tools, Oryonix AI operates under a privacy-first, client-side model. Your browsing data never leaves your machine. Start immediately with our pre-configured, highly capable models provided directly out of the box, or run local models offline via local gateways for complete offline privacy! --- KEY FEATURES Natural Language Interface No coding, no complex CSS selectors, and no XPath required. Just describe your goal in natural English (e.g., "Compare prices of three headphones on Amazon and compile them into a markdown table"). Oryonix AI parses your intent, understands page structures, and dynamically self-corrects if layouts change. Multi-Tab & Tab Group Control Navigate, open, close, and switch between multiple browser tabs simultaneously. The agent automatically creates and organizes managed tabs under a beautiful, color-coded Chrome tab group labeled "OryonixAI", keeping your personal workspace clean and untouched. Local & Cloud Model Integration (Zero Vendor Lock-In) Start automating immediately without complex configurations, or choose total offline execution: • Pre-Configured Cloud Model: Start right away! Our default cloud-hosted model is pre-configured and ready to run out of the box. No personal API keys needed. • Offline Local Models: Run 100% private, zero-cost, and offline-capable models locally on your machine via standard local gateways (such as Ollama). • Bring Your Own Credentials: Securely connect your own API keys to query standard cloud-based AI providers. All credentials are saved strictly inside your browser's private local storage. Rich Interactive Output & Charts Oryonix AI doesn't just return plain text. Get beautiful markdown summaries, structured data, and interactive visual charts (Bar, Line, Area, and Pie charts) rendered directly in your side panel using Recharts! Custom Project Guardrails Create tailored project-specific configurations. Configure custom rules or target states once, and every subsequent action automatically adheres to these guidelines—no repetitive typing needed. Privacy-First Architecture • Zero Cloud Footprint for Local: Everything runs client-side in your browser. • Zero Data Collection: No telemetry, no analytics, no third-party trackers. • Local Sandbox Storage: Local settings, preferences, and chat history are kept securely inside Chrome’s private local storage. • Open Source: Fully auditable, community-driven codebase licensed under the MIT License. --- EXAMPLE TASKS TO TRY • "Find the cheapest roundtrip flights to Tokyo for next month and compile the results." • "Go to GitHub, summarize the top trending repositories for TypeScript, and format them in a table." • "Help me fill out this job application form using my resume details." • "Navigate to the documentation, locate the Quick-Start section, and summarize the setup instructions." • "Compare pricing of a product across 3 different e-commerce pages." --- HOW TO START IN 3 EASY STEPS 1. Open the Oryonix AI side panel from your Chrome extensions toolbar. 2. Select your preferred model from the bottom dropdown (use the pre-provided cloud connection or select your local offline model). 3. Type your task in the chat window, click send, and watch your new autonomous copilot handle the work! Experience the future of the web with the ultimate client-side browser automation companion. Tell it what to do. Watch it work. --- USEFUL LINKS & RESOURCES • Official Website: https://oryonix-ai.vercel.app • GitHub Repository: https://github.com/Subhankar-Patra1/Oryonix-ai • Privacy Policy: https://oryonix-ai.vercel.app/privacy.html • Terms of Use: https://oryonix-ai.vercel.app/terms.html • Early Access Form: https://oryonix-ai.vercel.app/early-access/index.html