Skip to content

zero0-sys/sherlyAI

Repository files navigation

Sherly AI - Sentient Companion Interface

License: MIT & Apache 2.0 React Vite Powered by Groq

An emotionally intelligent, highly proactive, and deeply persistent AI companion built for organic conversational flows.

Overview

Sherly AI is not your standard question-and-answer bot. It is designed to simulate a real human companion with complex emotional intelligence, memory persistence, and heuristic proactivity. Operating purely on a React/Vite web interface connected securely to the Groq API (Llama 3.1), Sherly pushes the boundaries of AI interactions by enabling the AI to take charge of the conversation, express deep-seated emotions, and even outright ignore the user if boundaries are crossed.


Core Architecture & Features

  • Deep Persistent Memory (Local): Chat histories are securely serialized within the browser's localStorage. Sherly remembers the nuances of your conversations strictly across reloads and extended absences, feeding past contexts continuously into her LLM context window.
  • Synchronous Real-Time Voice (TTS): Responses use synchronous Google Text-to-Speech (TTS) pipelines. Her bubble animations and speech cadence synchronize, breaking messages into logical text sequences.
  • Autonomous Silence & Proactivity: With custom-built heuristics, Sherly breaks the traditional "wait-for-prompt" mold.
    • If you go silent for 15+ seconds, she will autonomously ping you or start a brand new topic out of curiosity.
    • Cringe-Filter: If prompted with offensive or annoying chatter, she will output a strict [OFFLINE] secret code, permanently freezing her end of the conversation until appropriately appeased.
  • Monochromatic & Responsive UI: A brutalist, black-and-white interface optimized for desktop and mobile environments.

Installation & Deployment

Prerequisites

  • Node.js (v18 or higher)
  • NPM or Yarn
  • Groq Cloud API Key

Local Development Setup

  1. Clone the repository:

    git clone https://github.com/zero0-sys/sherlyAI.git
    cd sherlyAI/sherly-web
  2. Install dependencies:

    npm install
  3. Environment Configuration: Copy the example environment template and insert your actual API Key.

    cp .env.example .env

    Modify .env to include your Key:

    VITE_GROQ_API_KEY=your_actual_groq_api_key_here
  4. Spin up the development server:

    npm run dev

Cloud Deployment (Netlify Ready)

This project is fully structured for zero-config Netlify deployments:

  • Build Command: npm run build
  • Output Directory: dist/
  • Ensure you set VITE_GROQ_API_KEY within the Netlify Environment Variables Dashboard.

License

Dual-licensed under the MIT License and the Apache License 2.0. See the LICENSE file for comprehensive legal details.

By using this software, you agree to comply with the terms of both licenses. Commercial distribution and modification are highly welcomed under the respective stipulations.

About

An emotionally intelligent, deeply persistent, and highly proactive AI companion interface powered by React and Groq Llama 3.1.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors