Skip to content

Manipulense is an AI-powered conversation analyzer that detects manipulation tactics in chats and explains them in human language.

Notifications You must be signed in to change notification settings

ErrVoid/Manipulense

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🛡️ Manipulense – Conversation Manipulation & Influence Detector

“Exposing hidden persuasion tactics in your everyday chats.”

🚀 Repo Stats

Stars Forks Issues Pull Requests Last Commit


🪄 Easter Egg for Hackathon Judges

👉 Did you know?
Every time you refresh this repo, you’re resisting manipulation 😉

fun gif


🌍 Problem We’re Solving

In daily chats (WhatsApp, Messenger, Discord…), people often get influenced or manipulated without realizing it.

  • Saying “yes” under peer pressure.
  • Agreeing because of guilt-tripping or flattery.
  • Being persuaded through fear or authority pressure.
  • Falling victim to gaslighting in toxic conversations.

There is no tool today that can read a conversation and explain:

  • Why did this person agree?
  • What tactics were used to influence them?
  • Was the response genuine, or was it forced/manipulated?

✨ Our Solution: Manipulense

Manipulense is an AI-powered conversation analyzer that detects manipulation tactics in chats and explains them in human language.

💬 Upload a chat log → ConvGuard highlights manipulative lines and shows you why they were manipulative.

📊 Visualize influence → See how pressure built up across the conversation.

🔎 Learn tactics → Understand guilt-tripping, flattery, fear, and gaslighting as they happen in text.


🎯 Key Features

  • 📑 Upload Chats – Import WhatsApp exports or paste raw text.
  • 🧠 AI-Powered Detection – Uses Hugging Face pipelines to detect emotions + manipulation tactics.
  • 🔎 Explainable Insights – Every manipulative line comes with an explanation.
  • 🎨 Modern Flutter UI – Dark theme, micro-animations, chat bubbles, and interactive highlights.
  • Fast Python Backend – Built with FastAPI + Hugging Face for NLP.

🛠 Tech Stack

Layer Tech Used
Frontend Flutter (Dart) – beautiful modern UI with animations
Backend Python (FastAPI) – NLP processing
AI Models Hugging Face Transformers (emotion + intent + persuasion detection)

🚀 Impact

  • 🛡 Safety & Awareness – Teens, women, and vulnerable groups can detect manipulation in relationships.
  • 🤝 Negotiation Training – Learn persuasion dynamics in business or sales chats.
  • 🧠 Mental Health – Identify toxic conversational patterns early.
  • 🎓 Education – A tool for teaching psychology, persuasion, and communication skills.

Manipulense is not just a project — it’s a step toward transparent, safer communication in the digital world.


📸 App Screenshots (UI Preview)


🧩 How It Works

  1. Input → Upload a chat log (WhatsApp export or pasted text).
  2. Preprocessing → Split into speaker turns.
  3. NLP Analysis
    • Emotion detection
    • Manipulation tactic classification (guilt, flattery, fear, gaslighting…)
  4. Influence Mapping → Detect when a person changes stance (No → Yes).
  5. Output
    • Highlighted chat bubbles (red = manipulative, green = neutral).
    • Explanations for each manipulative line.

⚙️ Installation & Setup

Frontend (Flutter)

git clone https://github.com/ErrVoid/Manipulense/
cd Manipulense/flutter_app
flutter pub get
flutter run

📱 Quick Install (APK)

Or you can directly install the APK from the releases page:

👉 Download APK v1.0.0


Made with ❤️ by ErrVoid & Piyushsakhuja


About

Manipulense is an AI-powered conversation analyzer that detects manipulation tactics in chats and explains them in human language.

Topics

Resources

Stars

Watchers

Forks

Contributors 2

  •  
  •