Designing-an-Agentic-AI-System-with-MCP-Integration

Sample application showing Agentic AI using MCP Server connecting to REST API, PostgreSQL Database and Document Storage

Setup

UV Installation and Setup

1. Install UV
   To install UV, run the following command:

   pip install uv

2. Verify Installation
   Confirm that UV is installed by running:

   uv --version

Using pyproject.toml

Setting Up a Virtual Environment and Installing Dependencies

1. Create a Virtual Environment
   Run the following command to create a virtual environment:

   python -m venv venv

2. Activate the Virtual Environment

   • On Windows:

     venv\Scripts\activate

   • On macOS/Linux:

     source venv/bin/activate

3. Install Dependencies from
   Use pip to install dependencies specified in the pyproject.toml file:

   pip install .

4. Verify Installation
   Ensure all dependencies are installed correctly:

   pip list

5. Run Test Cases
   Ensure all test cases are passed using pytest:

   pytest

Populate Financial Txn Data

1. Run the Script to Populate Data
   Execute the following command to populate sample financial transaction data:

   python util/populate_data.py

Configure MCP Client - Claude for Windows

1. Locate the Sample Configuration File
   Use the provided config/sample_claude_desktop_config.json file as a template.

2. Copy the Configuration File
   Place the file in the following directory:

   ~\AppData\Roaming\Claude

3. Rename the File
   Rename the file to claude_desktop_config.json.

4. Update Configuration Details

   • Open the claude_desktop_config.json file in a text editor.

   • Update the Alpha Vantage API Key with your key.

   • Update the Database Connection Details with the appropriate credentials and connection string.

Restart Claude Desktop

Restart Claude Desktop, you might have to kill tasks from Task manager. Note that once you restart it,

1. You should be able to see number of tools along with hammer icon.

2. This will start python program for each tool in background.