MCP Server for Vehicle Fleet Management

ASP.NET Core 8.0 MCP (Model Context Protocol) server providing vehicle tracking, fleet management, and real-time monitoring capabilities through both REST APIs and MCP tools.

Features

• JWT Authentication - Secure token-based authentication with automatic refresh
• Vehicle Tracking - Real-time status, GPS history, and trip analytics
• Conversation Context - AI-powered conversation management with automatic summarization
• MCP Tools - 9 vehicle management tools accessible via Model Context Protocol
• Fleet Analytics - Fleet statistics, vehicle compliance tracking, and daily status monitoring
• Security Guardrails - AI-powered query validation and threat detection
• Audit Logging - Complete audit trail of security events
• Rate Limiting - Configurable rate limits per endpoint
• Clean Architecture - Domain-driven design with clear separation of concerns

Quick Start

Prerequisites

• .NET 8.0 SDK
• SQL Server 2022+
• OpenAI API key (for conversation features)

Installation

1. Clone the repository

git clone <repository-url>
cd MCP-Server-for-Vehicle-Query-main

2. Configure environment variables

cp .env.example .env
# Edit .env with your actual values:
# - OPENAI_API_KEY: Your OpenAI/GitHub Models API key
# - SQL_PASSWORD: Database password (used in appsettings.json via ${SQL_PASSWORD})

3. Configure application settings

cp appsettings.example.json appsettings.json
# Edit appsettings.json with your configuration:
# - ApiSettings: Vehicle API endpoints and credentials
# - ConversationContext: AI conversation settings
# - ConnectionStrings: Database connection string (uses ${SQL_PASSWORD} from .env)

4. Run database migrations

dotnet ef database update

5. Run the application

dotnet run

The server will start at http://localhost:8080

Authentication

Login

Request:

curl -X POST http://localhost:8080/api/auth/login \
  -H "Content-Type: application/json" \
  -d '{
    "username": "your_username",
    "password": "your_password"
  }'

Response:

{
  "accessToken": "eyJhbGc...",
  "refreshToken": "eyJhbGc...",
  "expiresIn": 3600
}

Using Bearer Tokens

Include the access token in all subsequent requests:

curl http://localhost:8080/api/conversation/messages \
  -H "Authorization: Bearer eyJhbGc..."

Token Refresh

The system automatically refreshes tokens within 5 minutes of expiration. Manual refresh:

curl -X POST http://localhost:8080/api/auth/refresh \
  -H "Content-Type: application/json" \
  -d '{
    "refreshToken": "eyJhbGc..."
  }'

Core Concepts

Sessions

Sessions map bearer tokens to conversation contexts:

• Automatically created on first authenticated request
• Persist across server restarts (database-backed)
• Store conversation history and summaries
• Clear via DELETE /api/session/current

Conversation Context

The system maintains conversation history with intelligent context management:

• Sliding Window: Keeps last N messages in context (configurable, default 10)
• Token Tracking: Monitors token usage against budget (default 8000)
• Auto-Summarization: Compresses old messages when threshold reached (default 20 messages)
• Formatted Context: Combines system prompt + summaries + recent messages

MCP Tools

9 specialized tools for vehicle management accessible via /sse endpoint:

• Vehicle registry queries (by plate, ID, group, type)
• Fleet statistics and compliance tracking
• Real-time status monitoring
• Current day statistics (mileage, runtime, max speed)
• GPS history and trip analytics
• Token refresh

See Docs/TOOLS.md for complete catalog.

Configuration

Key settings in appsettings.json:

API Endpoints

{
  "ApiSettings": {
    "VehicleStatusUrl": "https://your-api.com/status",
    "VehicleApiUrl": "https://your-api.com/vehicles",


... [View full README on GitHub](https://github.com/TNHH0704/MCP-Server-for-Vehicle-Query-v2#readme)