This project is a versatile AI-powered dashboard designed for efficient information retrieval from CSV files and Google Sheets. It combines web search capabilities with AI-driven query processing to extract specific information based on user-defined prompts.
The AI Agent for Data-Driven Query Processing empowers users to:
Upload a CSV file or connect to a Google Sheet for data processing. Dynamically query data using customizable placeholders. Perform automated web searches to fetch missing information. Leverage Grok (xAI) for natural language query processing and parsing. Export results in a user-friendly format, such as CSV or directly update Google Sheets.
#Features CSV/Google Sheets Integration: Seamless data upload and preview. Dynamic Querying: Use placeholders to tailor queries for your data. Automated Web Search: Fetch relevant data using Google Custom Search API. LLM Integration: Enhance results with AI-based processing. Export Options: Download extracted results or update Google Sheets.
├── main.py # Main application file
├── google_api.py # Google Sheets API utilities
├── requirements.txt # Python dependencies
├── README.md # Project documentation
├── .env # (optional) Store API keys and environment variables
└── static/ # Static assets (if any)
- Clone the Repository
git clone https://github.com/ark5234/AI-Agent-Project.git
cd AI-Agent-Project
2️. Install Dependencies Ensure you have Python 3.8+ installed. Then, install the required libraries:
pip install -r requirements.txt
3️. Set Up API Keys Create a .env file in the project directory and add the following keys:
GROK_API_KEY=<Your_Grok_API_Key>
GOOGLE_API_KEY=<Your_Google_API_Key>
SEARCH_ENGINE_ID=<Your_Custom_Search_Engine_ID>
Alternatively, export these variables in your terminal:
export GROK_API_KEY=<Your_Grok_API_Key>
export GOOGLE_API_KEY=<Your_Google_API_Key>
export SEARCH_ENGINE_ID=<Your_Custom_Search_Engine_ID>
4️. Run the Application Start the Streamlit dashboard:
streamlit run main.py
Access the application in your browser at http://localhost:8501.
Step 1: Data Input Upload CSV: Use the Upload CSV File option to browse and upload your data. Google Sheets: Enter the Google Sheets URL and the desired sheet name to fetch data.
Step 2: Select Main Column After uploading or fetching data, choose the column you want to process (e.g., company names).
Step 3: Define Queries Input your query using placeholders like {entity}, which will dynamically reference rows in your selected column. Example: Get the contact email for {entity}
Step 4: Retrieve Results If data exists locally, it will filter and display relevant rows. If missing, it will perform a web search and query Grok for advanced AI-driven processing.
Step 5: Export Data Download the results as a CSV. Update Google Sheets directly.
1️. Grok (xAI) For AI-based natural language query processing. Integrated via openai Python library with the grok-beta model.
2️. Google Sheets API For reading and updating Google Sheets. Integrated using googleapiclient library.
3️. Google Custom Search API For automated web searches to fetch missing data.
Variable Name Purpose
GROK_API_KEY API key for Grok (xAI).
GOOGLE_API_KEY API key for Google APIs.
SEARCH_ENGINE_ID Custom Search Engine ID for CSE.
Ensure all environment variables are set before running the application.
Vikrant Kawadkar (@ark5234)
Feel free to raise an issue or submit a pull request for contributions!
If you have any questions, feedback, or suggestions, feel free to reach out:
Author: Vikrant Kawadkar
Email: [email protected]
This Google Drive folder contains:
A project walkthrough video demonstrating the working of the application.
Downloaded query results in CSV and TXT formats for reference.
https://drive.google.com/drive/folders/1-Zyg5JMycjNvxj0AHvtRh5KsdxMmn58C?usp=sharing
This project is licensed under the MIT License.