Skip to content

nithindas-k/ChatScope

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 

Repository files navigation

ChatScope Logo

ChatScope

Advanced WhatsApp Chat Analytics & AI Insights

Live Demo


ChatScope is a powerful, privacy-first web application designed to transform your raw WhatsApp .txt export logs into stunning visualizations and deep psychological insights. By leveraging modern data parsing and advanced AI Language Models, ChatScope reveals hidden patterns in your conversations without ever compromising your data privacy.


Core Features

Dashboard & Activity Maps

  • Instant Overviews: View total message counts, active participant leaderboards, and top emojis at a glance.
  • Timeline Analysis: Track message frequencies across days and months to see when your conversations peak.
  • Hourly Heatmaps: Beautiful bar and pie charts map out the busiest hours of the day and who dominates the conversation.

Linguistic & Word Analysis

  • Vocabulary Tracking: Discover exactly which words are used the most by each specific participant.
  • Emoji Usage: A detailed breakdown of the most frequently used emojis.
  • Typing Habits: Calculates average message lengths to reveal who sends short bursts and who writes long paragraphs.

Neural AI Insights (Powered by Groq)

  • Psychological Profiling: Generates dynamic personality archetypes for participants based on their messaging style.
  • Sentiment Analysis: Calculates the overall emotional tone of the timeline (e.g., 85% Positive, 10% Banter, 5% Conflict).
  • Communication Dynamics: The AI strictly analyzes the interaction style, reading between the lines without storing the raw text.

Interactive AI Chat Analyst

  • Chat with your Chat: An interactive terminal where you can ask free-form questions about the conversation history (e.g. "Who texts first usually?" or "What is our most talked about topic?").
  • Context-Aware: The AI handles regional languages, Manglish, and heavy slang natively.

Strict Privacy System (Zero-Retention)

We believe your private conversations shouldn't be trained on or sold.

  • Memory-Only Processing: When you upload your .txt file, it is read entirely in temporary RAM. The .txt file is never saved to our servers' hard drives.
  • Ephemeral Database Sessions: Upon processing, the anonymous numerical statistics and only a snippet of the last 50 messages are temporarily held in our MongoDB database.
  • Safe AI: We use Groq API (Llama 3.1) which strictly prohibits training their models on our user requests. We do not use ChatGPT for data privacy reasons.
  • Instant "Clear Data" Wipe: The UI provides a manual "Clear Data" button which fires an immediate deleteOne() backend command, instantly purging your temporary session from our database and wiping your local browser cache. Leaving zero trace.

Technology Stack

Frontend

  • React 18 (Vite + TypeScript)
  • Tailwind CSS & Shadcn UI (Component Library)
  • Framer Motion (Smooth Micro-animations & Transitions)
  • Zustand (Local, persistent State Management)
  • Recharts (Data Visualization)
  • WhatsApp Dark Theme deeply integrated across the UI.

Backend

  • Node.js & Express.js (TypeScript)
  • MongoDB + Mongoose (Session Management)
  • Groq SDK (AI LLM inference)
  • Multer (In-memory file buffer processing)

Running ChatScope Locally

Prerequisites

  • Node.js (v18+)
  • MongoDB running locally or a MongoDB Atlas URI
  • A free API key from Groq Console

1. Setup Backend

cd backend
npm install

Create a .env file in the backend/ directory:

PORT=3000
MONGODB_URI=mongodb://127.0.0.1:27017/chatscope
GROQ_API_KEY=your_groq_api_key_here

Run the server:

npm run dev

2. Setup Frontend

cd frontend
npm install

Create a .env file in the frontend/ directory (if using variable routing):

VITE_API_URL=http://localhost:3000

Run the client:

npm run dev

3. Usage

Simply visit http://localhost:5173, click Upload Chat, drop your WhatsApp .txt file, and dive into the analytics!

About

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages