OneWhistle/dowhistle-mcp-server

DoWhistle MCP Server

Submission Details

• Team Name: DoWhistle
• Team Members: Raja Appachi , Nishanth M, Mithran Raja, Nitin Murali
• Hackathon Theme: Theme 2 - Agents and tools need a reliable interface layer—secure, permissioned, and observable—to interact with real-world APIs.

Project Structure

mcp-server/
├── config/
│   └── settings.py              # Configuration management
├── utils/
│   └── http_client.py          # HTTP client with retry logic
├── agents/
│   ├── __init__.py
│   ├── search.py               # Search functionality
│   ├── auth.py                 # Authentication tools
│   ├── whistle.py              # Whistle management
│   └── user.py                 # User settings management
├── tests/
│   ├── __init__.py
│   ├── test_agents.py          # Agent tests
│   └── conftest.py             # Test configuration
├── app.py                      # Entry point
├── pyproject.toml              # Dependencies
├── Dockerfile                  # Container configuration
├── docker-compose.yml          # Multi-container setup
├── .env.example                # Environment variables template
└── README.md                   # Documentation

Agents and Tools

1. Search Agent (agents/search.py)

• Tool: search
• Purpose: Search functionality through Express API

2. Authentication Agent (agents/auth.py)

• Tools:
  • sign_in: User authentication
  • verify_otp: OTP verification
  • resend_otp: Resend OTP code
• Purpose: Handle all authentication flows

3. Whistle Agent (agents/whistle.py)

• Tools:
  • create_whistle: Create new whistle reports
  • list_whistles: List whistles with pagination
• Purpose: Complete whistle management

4. User Agent (agents/user.py)

• Tools:

  • toggle_visibility: Control user visibility

• Purpose: User preference management

Installation and Setup

Prerequisites

• Python 3.11+
• Access to your Express.js API
• Docker (optional)

Local Development

1. Clone and setup:

git clone https://github.com/OneWhistle/dowhistle-mcp-server
cd dowhistle-mcp-server

uv sync

source .venv/bin/activate  # On Windows: .venv\Scripts\activate

1. Configuration:

cp env.example .env
# Edit .env with your settings

1. Run the server:

uvicorn app:app --host 0.0.0.0 --port 8000 

Docker Deployment

1. Build and run:

docker compose up -d --build 
(or)

docker-compose up --build 

2. Production deployment:

docker-compose -f docker-compose.yml -f docker-compose.prod.yml up -d

Configuration

All configuration is managed through environment variables:

Demo Video

https://www.youtube.com/watch?v=wOn_jo6Re60

What We'd Do with More Time

With more time, we would focus on:

Develop a dedicated web and mcp-client:

• DoWhistle-MCP-Gateway: Containing the MCP Client and integration with OpenAI for agentic capabilities.

Repo: DoWhistle-MCP-Gateway

• DoWhistle-MCP-Web: Housing the ChatBot interface for user interaction.

Repo: DoWhistle-MCP-Web

Smithery Link for MCP Server

👉 Smithery dowhistle-mcp-server