ScamVerifierV6 / README.md
AayanSuleri's picture
Update README.md
b6c94f1 verified
|
Raw
History Blame Contribute Delete
3.71 kB
---
title: ScamVerifierV6
emoji: 💻
colorFrom: red
colorTo: blue
sdk: gradio
sdk_version: 5.42.0
app_file: app.py
pinned: false
license: mit
short_description: Instantly detect and flag potential online scams
---
# Scam-Verifier 🛡️
**Protect yourself from phishing, spam, and scam messages with AI-powered analysis.**
This web app allows users to paste suspicious messages, emails, or URLs and get a **risk assessment**, **red flags**, **recommendations**, and a ready-to-use **report template**.
The backend uses **LangChain + OpenAI GPT** for intelligent scam detection.
---
## Features
* **AI-Powered Scam Detection**
Detects phishing, fake job offers, lottery scams, investment fraud, and other malicious messages.
* **Risk Scoring & Classification**
* HIGH RISK
* MEDIUM RISK
* LOW RISK
* SAFE
* **Red Flags & Recommendations**
Lists specific warning signs and actionable steps to stay safe.
* **Report Template Generator**
Auto-generates a professional report to submit to authorities or platforms.
* **Example Messages**
Test with preloaded phishing, fake job, investment, and lottery scam examples.
* **Clean & Modern UI**
Fully responsive, mobile-friendly, with smooth interactions and copy-to-clipboard support.
---
## How It Works
1. **User Input:** Paste a suspicious message or provide a URL.
2. **AI Analysis:** Backend uses GPT via LangChain to evaluate the message.
3. **Risk Assessment:** Generates a risk score (0–100) and classification.
4. **Red Flags & Recommendations:** Highlights suspicious elements and safety tips.
5. **Report Template:** Provides a ready-to-use scam report template.
---
## Installation
> Note: Designed for deployment on **Hugging Face Spaces**, but can run locally with Flask.
1. Clone the repository:
```bash
git clone <repo-url>
cd scam-signal-verifier
```
2. Create a virtual environment and activate:
```bash
python -m venv venv
source venv/bin/activate # Linux/Mac
venv\Scripts\activate # Windows
```
3. Install dependencies:
```bash
pip install -r requirements.txt
```
4. Set your **OpenAI API key**:
```bash
export OPENAI_API_KEY="your_api_key" # Linux/Mac
set OPENAI_API_KEY="your_api_key" # Windows
```
5. Run the app:
```bash
python app.py
```
6. Open in browser: `http://localhost:7860`
---
## Usage
1. Paste your suspicious message in the text area.
2. Optionally, add a URL to check.
3. Click **Analyze for Scam Signals**.
4. View risk score, red flags, recommendations, and report template.
5. Use the **Copy Report Template** button to quickly save it.
---
## Deployment on Hugging Face Spaces
1. Create a new **Flask Space**.
2. Add `app.py`, `requirements.txt`, and `templates/index.html`.
3. Set your **OpenAI API key** as a secret in HF Spaces.
4. Deploy — the app is live automatically.
---
## Example Messages
* **Phishing Email:** Fake account verification messages.
* **Fake Job Offer:** Work-from-home scams requesting upfront fees.
* **Investment Scam:** Unrealistic “double your money” offers.
* **Lottery Scam:** Fake lottery winnings asking for bank details.
---
## Technologies Used
* **Python 3.10+**
* **Flask** – Backend web framework
* **LangChain** – AI orchestration
* **OpenAI GPT** – Scam detection model
* **HTML/CSS/JS** – Frontend UI
* **FontAwesome** – Icons
---
## Security Notes
* All analysis happens server-side using **OpenAI GPT**.
* No user data is stored permanently.
* Always **verify critical messages independently**.
* This tool is **advisory only**, not legally binding.
---
## License
MIT License © 2025
---
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference