demo-redshift-mcp

Executive Question
    ↓
Gradio UI
    ↓
CrewAI Router Agent (query understanding)
    ↓
CrewAI Analysis Agent (data fetching + insights)
    ↓
MCP Tools (GetRenewedCount, GetLeftCount, etc.)
    ↓
Data Layer (pandas + Excel/CSV queries)
    ↓
Response formatted for executives

# Install using uv (recommended)
uv sync

# OR using pip
pip install -r requirements.txt

cp .env.example .env
# Edit .env and add your ANTHROPIC_API_KEY

python -m demo_redshift_mcp.data_generator

# Launch Gradio interface
python -m demo_redshift_mcp

# Opens at http://localhost:7860

python -c "
from src.demo_redshift_mcp.crew_agents import run_customer_migration_analysis
result = run_customer_migration_analysis('How many customers renewed?')
print(result)
"

demo-redshift-mcp/
├── src/demo_redshift_mcp/
│   ├── app.py                 # Gradio UI entry point
│   ├── crew_agents.py         # CrewAI agents + workflow
│   ├── mcp_server.py          # MCP tools definition
│   ├── data_layer.py          # Data access logic
│   ├── data_generator.py      # Mock data generation
│   └── __init__.py
├── data/                      # Generated mock data
│   ├── legacy_product.xlsx
│   ├── new_product_customers.csv
│   └── competitor_coverage.csv
├── INSURANCE_POC_ARCHITECTURE.md
├── pyproject.toml
└── .env

# Ensure you're in the right directory
cd /Users/Balu/Documents/Projects/MyCode/demo-redshift-mcp

# Reinstall dependencies
uv sync

# Generate mock data
python -m demo_redshift_mcp.data_generator

from src.demo_redshift_mcp.data_layer import DataLayer

data = DataLayer()
result = data.customers_renewed()
print(result)