NVD Vulnerability Chatbot

This project is designed to create a chatbot that interacts with the National Vulnerability Database (NVD) to provide information about Common Vulnerabilities and Exposures (CVEs). The chatbot uses OpenAI's GPT-4 model to understand user queries and extract relevant filters to query the NVD API.

Project Structure

Files and Directories

• .env: Contains environment variables, including API keys for OpenAI and the NVD API.
• chatbot.py: Defines the main chatbot functionality, initializes the language model, and handles user interaction.
• constants.py: Defines constant mappings and a prompt template used for extracting filters from user queries.
• req.txt: Lists the dependencies required for the project.
• tool.py: Defines the tool for querying the NVD database and extracting filters from user queries.
• util/utilities.py: Contains utility functions for constructing API queries and filtering API responses.

Setup

1. Clone the repository:

   git clone <repository-url>
   cd <repository-directory>

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the dependencies:

   pip install -r req.txt

4. Set up environment variables:

   • Create a .env file in the root directory of the project.
   • Add your OpenAI API key and NVD API key to the .env file:

     OPENAI_API_KEY="your_openai_api_key"
     API_URL="https://services.nvd.nist.gov/rest/json/cves/2.0"
     API_KEY="your_nvd_api_key"
     

Usage

Run the chatbot by executing the following command:

python chatbot.py

Key Components

chatbot.py

• Initializes the language model using OpenAI's GPT-4.
• Creates an agent with a custom tool for querying the NVD database.
• Defines the chatbot() function, which handles user interaction and processes queries using the agent.

tool.py

• Imports necessary libraries and loads environment variables.
• Defines the language model and prompt template for query extraction.
• Implements the extract_filters(user_query: str) -> Dict function to extract filters from user queries using the language model.
• Implements the query_online_database(query: str) -> Dict function to query the NVD database with extracted filters and return filtered API responses.

utilities.py

• construct_api_query(filters: Dict) -> str: Builds an API query string from extracted filters.
• filter_response(data: Dict) -> List[Dict]: Filters the API response to include only relevant fields.

constants.py

• PARAMETER_MAPPING: A dictionary mapping parameter names to their descriptions.

• PROMPT_TEMPLATE: A template for generating prompts to extract filters from user queries.

License

This project is licensed under the MIT License.