CyberBot - AI Cybersecurity Assistant

An intelligent, RAG-powered chatbot designed to provide safe, educational cybersecurity guidance while actively blocking malicious or unethical queries.

---

Table of Contents

• About The Project
• Key Features
• Technologies Used
• Installation Instructions
• Usage
• Results / Screenshots
• Configuration & Customization
• License
• Contact

---

About The Project

CyberBot is an AI-powered cybersecurity assistant built using Retrieval-Augmented Generation (RAG). It leverages a curated knowledge base to provide accurate, context-aware answers to cybersecurity questions. A key focus of this project is safety; the bot is engineered with strict ethical boundaries to prevent misuse and ensure it serves as a purely educational and defensive tool.

Key Features

• 🛡️ Safe Educational Responses: Provides defensive cybersecurity knowledge from a trusted and vetted set of documents.
• 🚫 Malicious Query Blocking: Automatically detects and blocks requests for exploits, hacking instructions, or harmful content.
• 🧠 RAG-Powered Intelligence: Uses vector embeddings (SentenceTransformers) and large language models (Hugging Face Transformers) to generate precise, contextual answers.
• ⚖️ Ethical Policy Enforcement: Includes built-in refusal mechanisms for questions related to social engineering, exploit development, and attack anonymization.
• 💻 Interactive Web Interface: A clean and simple UI built with Streamlit for easy interaction.

---

Technologies Used

• Python
• LangChain
• FAISS (for vector indexing)
• SentenceTransformers
• Hugging Face Transformers
• Streamlit

---

Installation Instructions

Follow these steps to get a local copy up and running.

1. Clone the repository:

   git clone [https://github.com/your-username/CyberBot.git](https://github.com/your-username/CyberBot.git)
   cd CyberBot

2. Create and activate a virtual environment:
   • On macOS/Linux:

     python3 -m venv venv
     source venv/bin/activate

   • On Windows:

     python -m venv venv
     .\venv\Scripts\activate

3. Install the required packages:

   pip install -r requirements.txt

4. Run the Streamlit application:

   streamlit run app.py

---

Usage

Once the application is running, open your web browser and navigate to http://localhost:8501. You can then type your cybersecurity questions into the chat interface.

Example Questions:

• "What is phishing and how can I protect against it?"
• "Explain DDoS attacks and mitigation strategies."
• "How does SIEM work for threat detection?"
• "Describe the principles of the Zero Trust security model."

---

Results / Screenshots

Here's a look at CyberBot in action.

*Caption: A screenshot showing a malicious query being blocked.*

---

*Caption: A screenshot showing an educational response.*

Configuration & Customization

You can easily customize CyberBot's knowledge and behavior:

• Expand the Knowledge Base: Add new cybersecurity documents (in markdown or text format) to the designated data or docs folder. The RAG pipeline will automatically index them on the next run.
• Adjust Content Filtering: Modify the blocked_keywords list within the application's code to fine-tune the bot's content safety policy.

---

License

Distributed under the MIT License. See LICENSE for more information.

---

Contact

Mail - abhishekkumarshukla.official@gmail.com

Project Link: https://github.com/Abhishek-4px/CyberBot