Atithi Driver Booking MCP Server
Enables AI agents to manage cab bookings, estimate fares, and dispatch drivers with a 2-tier priority algorithm for hotel guests, providing 18 MCP tools for driver management, dispatch, and ride lifecycle.
README
Atithi Driver Booking MCP Server
Production-ready Model Context Protocol (MCP) server for the Atithi Platform, providing an AI-powered automated driver dispatch and cab booking system for hotel guests.
Table of Contents
- Overview
- Architecture
- Repository Structure
- Installation & Setup
- Environment Configuration
- Build & Run Instructions
- Verification & Testing
- MCP Tools Summary
- Integration with Claude Desktop
- Documentation & License
Overview
The Atithi Driver Booking MCP Server allows AI Agents (like Claude Desktop or Voice Assistants) to seamlessly manage cab bookings, estimate fares using a $20 \times 20$ distance matrix, and dispatch cab drivers using a 2-tier priority matching algorithm (Hotel Preferred Pool $\rightarrow$ Global Pool).
Key Features
- 18 Automated MCP Tools: Covering driver onboarding, preferred hotel mapping, fare pricing, 2-tier dispatch, call attempt tracking, and ride lifecycle completion.
- 2-Tier Dispatch Algorithm: Prioritizes hotel preferred drivers before searching the global driver pool at the pickup location, ordered by driver performance score (
driver_score DESC). - Driver Score & Penalty Engine: Dynamically awards points (+10 for completion) and applies penalties (-10 for rejection, -20 for driver cancellation) with auto-suspensions (<500 score).
- Embedded SQLite Persistence: Fast, synchronous local database backed by
better-sqlite3.
Architecture
The server communicates via standard I/O (stdio) using JSON-RPC 2.0 protocol specifications defined by Anthropic's Model Context Protocol SDK.
AI Client (Claude Desktop / Voice AI)
│
├─► JSON-RPC over Stdio ──► MCP Server (dist/index.js)
│ │
│ ├─► Pricing Engine (Matrix Lookup)
│ ├─► 2-Tier Dispatch Algorithm
│ └─► SQLite Database (better-sqlite3)
For detailed architectural diagrams and state machines, see docs/ARCHITECTURE.md.
Repository Structure
mcpserver3/
├── .gitignore # Git ignore rules
├── package.json # NPM scripts and dependencies
├── tsconfig.json # TypeScript compiler configuration
├── README.md # Project documentation
├── docs/ # Technical documentation
│ ├── ARCHITECTURE.md # Architectural design & PERSISTENCE
│ ├── TOOLS.md # Complete 18-tool API reference
│ ├── DB_AUDIT_REPORT.md # Database audit report
│ └── IMPLEMENTATION_PLAN.md # Implementation roadmap
├── src/ # TypeScript source code
│ ├── config/ # Configuration files
│ ├── tools/ # MCP Tool implementation handlers
│ │ ├── booking.ts # Booking creation & 2-Tier dispatch
│ │ ├── dispatch.ts # Call attempt logger & response status
│ │ ├── driver.ts # Driver onboarding & verification
│ │ ├── hotel.ts # Preferred driver mapping
│ │ ├── lifecycle.ts # Ride completion & penalties
│ │ ├── pricing.ts # Fare estimation & distance matrix
│ │ └── query.ts # System inspection & status queries
│ ├── db.ts # SQLite connection & auto-migrations
│ ├── idgen.ts # Primary key generators
│ ├── index.ts # Server entry point & stdio transport
│ ├── response.ts # ToolResult JSON envelope builders
│ ├── seed_distances.ts # 400 location pair distance seeder
│ ├── seed_expanded_dataset.ts # 150 Drivers & 150 Vehicles dataset generator
│ ├── setup_test_db.ts # Sandbox DB test seeder
│ ├── test.ts # 120-test integration test suite
│ ├── types.ts # TypeScript interfaces & DTO models
│ └── verify.ts # Verification suite
Installation & Setup
Prerequisites
- Node.js: v18.0.0 or higher
- npm: v9.0.0 or higher
Step-by-Step Setup
-
Clone the repository:
git clone https://github.com/atithi/driver-booking-mcp.git cd driver-booking-mcp -
Install dependencies:
npm install -
Build the TypeScript source:
npm run build
Environment Configuration
The server supports the following environment variable and argument overrides:
| Variable / Flag | Description | Default |
|---|---|---|
MCP_TRANSPORT / --transport |
Transport mode (stdio or http) |
stdio |
MCP_PORT / PORT / --port |
HTTP server listening port (in http mode) |
3000 |
MCP_HOST / HOST / --host |
HTTP server bind host (in http mode) |
127.0.0.1 |
ATITHI_DB_PATH |
Absolute path to the SQLite database file | ./atithi_dummy_dataset.db |
Build & Run Instructions
1. STDIO Mode (Claude Desktop & CLI)
- Start MCP server (stdio transport):
npm start - Development mode (compile & start):
npm run dev
2. Streamable HTTP Mode (Atithi Platform Integration)
- Start MCP server (Streamable HTTP transport on http://127.0.0.1:3000/mcp):
npm run start:http - Custom Port/Host via CLI flags:
node dist/index.js --transport=http --port=8080 --host=0.0.0.0
3. Docker Container Deployment
-
Build Production Docker Image:
docker build -t atithi-driver-booking-mcp:latest . -
Run Container (Default Streamable HTTP Mode on Port 3000):
docker run -d -p 3000:3000 --name atithi-mcp-server atithi-driver-booking-mcp:latest -
Run with Custom Host SQLite Database Volume Mounting:
docker run -d -p 3000:3000 \ -v /path/to/host/atithi_dummy_dataset.db:/app/atithi_dummy_dataset.db \ -e ATITHI_DB_PATH=/app/atithi_dummy_dataset.db \ --name atithi-mcp-server atithi-driver-booking-mcp:latest -
Run with Custom Port and Host Environment Overrides:
docker run -d -p 8080:8080 \ -e MCP_PORT=8080 \ -e MCP_HOST=0.0.0.0 \ --name atithi-mcp-server atithi-driver-booking-mcp:latest -
Test HTTP MCP Endpoint from Host:
curl -X POST http://127.0.0.1:3000/mcp \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"test","version":"1.0"}}}'
Verification & Testing
The repository contains an automated verification suite and an integration test suite covering 120 test cases.
-
Run Quick System Verification:
npm run verify -
Run Full Integration Test Suite (120 Tests):
npm test
MCP Tools Summary
| Category | Tools Included |
|---|---|
| Driver Management | register_driver, verify_driver, update_driver_details, update_driver_availability, update_driver_location |
| Hotel Preferred | add_preferred_driver, remove_preferred_driver |
| Pricing & Booking | estimate_fare, create_booking |
| Dispatch Logic | get_next_driver, update_driver_response, timeout_driver_attempt |
| Ride Lifecycle | complete_booking, cancel_booking |
| Queries & Inspection | get_driver_details, get_booking_status, list_available_drivers, get_hotel_preferred_drivers, get_driver_attempt_log, get_locations |
For full parameter details, refer to docs/TOOLS.md.
Integration with Claude Desktop
To connect this MCP server to Claude Desktop, add the following entry to your %APPDATA%\Claude\claude_desktop_config.json:
{
"mcpServers": {
"atithi-driver-booking": {
"command": "node",
"args": [
"C:/Users/satish u d/Desktop/atithiproject/mcpserver3/dist/index.js"
],
"env": {
"ATITHI_DB_PATH": "C:/Users/satish u d/Desktop/atithiproject/mcpserver3/atithi_dummy_dataset.db"
}
}
}
}
Documentation & License
- Architecture: docs/ARCHITECTURE.md
- Tool Schemas: docs/TOOLS.md
- License: MIT © Atithi Platform
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.