.svg){kind=small}
AI-Powered Fleet Management with Model Context Protocol
The first fleet management platform with MCP integration. Connect your entire operation directly to AI assistants like Claude—enabling intelligent automation, instant insights, and conversational fleet control.
Enterprise-Grade AI Integration
How AI Fleet Management Works
From your fleet data to conversational control in four simple steps. No complex setup, no coding required.
Your Fleet Data
Customers, vehicles, rides, and transactions live in your Levy Fleets database—secure and always up-to-date.
MCP Server Connects
Our MCP server provides a secure bridge, translating AI requests into authenticated database operations.
AI Understands Context
Claude, ChatGPT, or your custom AI gains full understanding of your fleet through 11 specialized tools.
You Ask, AI Acts
Ask questions in plain English. Get instant answers, trigger actions, and automate workflows conversationally.
Your Fleet Data
Customers, vehicles, rides, and transactions live in your Levy Fleets database—secure and always up-to-date.
MCP Server Connects
Our MCP server provides a secure bridge, translating AI requests into authenticated database operations.
AI Understands Context
Claude, ChatGPT, or your custom AI gains full understanding of your fleet through 11 specialized tools.
You Ask, AI Acts
Ask questions in plain English. Get instant answers, trigger actions, and automate workflows conversationally.
Secure, Standards-Based Integration
Built on the Model Context Protocol by Anthropic—an open standard for AI-to-data connections.
Your Fleet Data
Customers, Vehicles,
Rides, Transactions
Levy MCP Server
Secure Bridge
11 AI Tools
AI Assistants
Claude, ChatGPT,
Custom Agents
Service Role Auth
Supabase service role with row-level security
Audit Logging
Every operation logged with full context
Open Protocol
MCP standard by Anthropic—no vendor lock-in
Traditional Dashboard vs AI-Powered
See how MCP transforms everyday fleet operations from clicks to conversations.
"Find customer john@example.com""What's the status of V-1247?""Refund ride R-8847, customer complaint""How much revenue did we make today?""Show vehicles under 20% battery not in use""Unlock vehicle V-1247"Multiple screens & clicks
Single conversation
Training required for new staff
Natural language—anyone can use
Manual, repetitive tasks
Automated AI workflows
Limited to dashboard hours
24/7 AI availability
Why MCP Changes Everything
Model Context Protocol enables AI assistants to understand and interact with your fleet in ways that were previously impossible. Here's what that means for your operation.
Conversational Fleet Control
Ask questions in plain English. Get instant answers about customers, vehicles, rides, and revenue—no dashboards or reports needed.
Real-Time Data Access
AI assistants connect directly to your live fleet data. Current vehicle locations, active rides, customer details—all accessible instantly.
Secure & Controlled
Service role authentication with Supabase. All operations are logged and auditable. You control what data AI can access.
Instant Operations
Issue refunds, add wallet credits, send vehicle commands—all through natural conversation. No form-filling or menu navigation.
Automate Complex Workflows
Build AI agents that handle customer support, fleet monitoring, and operational tasks. Chain multiple operations seamlessly.
Intelligent Insights
Ask for fleet statistics, revenue breakdowns, or ride patterns. AI analyzes your data and surfaces actionable insights.
Complete Fleet Control at Your AI's Fingertips
Every tool is designed for AI-first interaction. Natural language in, structured data out. Click any tool to see example usage.
Customer Operations
Customer Lookup
levy_customer_lookupFind any customer by email, phone, or customer number. Get wallet balance, ride history, and account details.
levy_customer_lookup(email: "john@example.com")Customer Search
levy_customer_searchSearch customers with filters—status, registration date, ride count, spending totals, and more.
levy_customer_search(min_rides: 50, created_after: "2024-01-01")Wallet Credit
levy_wallet_creditAdd credits to customer wallets. Process refunds, promotional credits, or corrections with full audit trail.
levy_wallet_credit(customer_id: "...", amount_cents: 2500, credit_type: "promotional")Vehicle Operations
Vehicle Lookup
levy_vehicle_lookupFind vehicles by number, UUID, or IoT IMEI. Get status, battery level, location, and current rider.
levy_vehicle_lookup(vehicle_number: "V-1247")Vehicle Search
levy_vehicle_searchSearch fleet by status, battery level, location zone, or model. Filter available vs. rented vehicles.
levy_vehicle_search(status: "available", max_battery: 20)Vehicle Command
levy_vehicle_commandSend IoT commands—lock, unlock, or beep any vehicle. Perfect for support scenarios or fleet management.
levy_vehicle_command(vehicle_number: "V-1247", command: "unlock")Ride Operations
Ride Lookup
levy_ride_lookupGet complete ride details including customer, vehicle, pricing breakdown, and optional GPS history.
levy_ride_lookup(ride_id: "...", include: ["location_history"])Ride Search
levy_ride_searchSearch rides by customer, vehicle, date range, status, amount, or refund status. Full pagination support.
levy_ride_search(min_amount_cents: 5000, date_from: "2024-01-15")Ride Refund
levy_ride_refundIssue refunds directly to customer wallets. Specify custom amounts or refund full ride cost.
levy_ride_refund(ride_id: "...", reason: "Customer complaint")Fleet Analytics
Fleet Statistics
levy_fleet_statsGet comprehensive fleet metrics—vehicle counts by status, ride totals, revenue, and averages.
levy_fleet_stats(date_from: "2024-01-08", date_to: "2024-01-14")Active Rides
levy_active_ridesList all currently active rides with customer, vehicle, duration, and real-time location data.
levy_active_rides(limit: 50)See MCP in Action
From customer support to fleet automation, see how AI + MCP transforms everyday fleet operations.
AI Customer Support
Customer says their scooter won't unlock. Email: john@example.com
I found John's account. He has an active ride on vehicle V-1247. Let me check the vehicle status...
The vehicle shows as locked with 78% battery. I'll send an unlock command now.
Vehicle V-1247 unlocked successfully
Automated Fleet Monitoring
Show me all vehicles with battery below 20% that aren't currently rented
Found 12 vehicles with low battery. 8 are at the downtown hub, 4 scattered across zones A and B.
Create a collection route for the maintenance team
Here's an optimized route hitting all 12 vehicles. Estimated collection time: 2.5 hours.
Proactive Issue Detection
Alert: Ride R-8847 has been active for 4+ hours
Checking ride details... Customer Sarah M. started at 2pm, current location shows the vehicle stationary for 90 minutes.
I'll send a check-in SMS and flag this for support review.
Instant Reporting
How did we do last week vs the week before?
Last week: 2,847 rides, $14,230 revenue, 94% completion rate. That's up 12% in rides and 15% in revenue compared to the prior week.
What drove the increase?
The downtown zone saw 340 additional rides, mostly during evening hours. Looks like the new bar district parking zone is performing well.
Works With Your AI Stack
MCP is an open protocol. Our server works with any AI assistant or framework that supports the Model Context Protocol standard.
Claude (Anthropic)
First-class MCP support in Claude Desktop, Claude Code, and Claude API. Full tool execution capabilities.
- Claude Desktop App
- Claude Code CLI
- Claude API with Tools
ChatGPT (OpenAI)
Use GPT models with function calling to interact with your fleet through our tool definitions.
- Custom GPT Actions
- Assistant API
- Function Calling
Custom AI Agents
Build your own AI agents using our MCP server. Python, JavaScript, and any MCP-compatible client.
- Python SDK
- Node.js SDK
- Any MCP Client
CRM & Support Tools
Integrate AI-powered fleet queries directly into Zendesk, Intercom, Salesforce, and more.
- Zendesk Integration
- Intercom Bots
- Salesforce Einstein
MCP is an open standard developed by Anthropic for connecting AI assistants to external data sources. It ensures secure, standardized, and auditable AI-to-data interactions.
Learn more about MCP →What Our Customers Say
Fleet operators are transforming their operations with AI-powered management.
“MCP changed how we handle support. What used to take 5 minutes per ticket now takes 30 seconds. Our agents just ask Claude and get instant answers.”
“The ability to query fleet data conversationally is incredible. I can get insights that would have taken an hour of spreadsheet work in seconds.”
“We built a custom AI agent that monitors our fleet 24/7 and auto-escalates issues. MCP made it possible in a weekend.”
Trusted by innovative fleet operators worldwide
Four Steps to AI-Powered Fleet Management
No complex integrations. No developer required. Get from zero to AI-powered in under 15 minutes.
Sign Up for Levy Fleets
Create your account and set up your fleet. Import vehicles, configure zones, and invite your team.
Enable MCP Integration
Go to Settings → Integrations → MCP. Generate your credentials and copy your server configuration.
Connect Your AI Assistant
Add the MCP server to Claude Desktop, Claude Code, or your custom AI agent. We provide copy-paste configs.
Start Asking Questions
Query customers, check vehicles, process refunds, get reports—all through natural conversation.
Frequently Asked Questions
Everything you need to know about AI-powered fleet management with MCP.
MCP (Model Context Protocol) is an open protocol developed by Anthropic that allows AI assistants to securely connect to external data sources. In fleet management, MCP enables AI to access real-time fleet data, execute operations like refunds and vehicle commands, and provide instant insights—all through natural language conversation. It's the bridge that makes AI truly useful for fleet operations.
Levy Fleets provides an MCP server that connects your fleet data to AI assistants like Claude. When you ask a question like "Find customer john@example.com", the AI uses our MCP tools to query your database securely, returning real-time information. For actions like "Refund ride R-1234", the AI executes the operation through authenticated API calls. It's like giving your AI assistant direct access to your fleet—with guardrails.
Absolutely. Security is built into every layer. The MCP integration uses service role authentication with Supabase, meaning all operations require valid credentials. Every API call is logged for audit purposes. Write operations (like refunds) have built-in limits ($100 max) and require explicit reasons. You control exactly which data and operations the AI can access, and all connections use encrypted protocols.
Our MCP server works with Claude (via Claude Desktop, Claude Code CLI, and the API with native MCP support), ChatGPT (via Custom GPT Actions and the Assistant API with function calling), and any custom AI agent built with MCP-compatible SDKs in Python, JavaScript, or other languages. The open protocol ensures compatibility with current and future AI platforms.
No. MCP integration is included with all Levy Fleets plans at no additional cost. You get full access to all 11 AI tools covering customer operations, vehicle management, ride handling, and fleet analytics. The only costs are your own AI API usage (Claude, ChatGPT, etc.), which you control directly.
Yes! That's one of the most powerful aspects of MCP. You can build AI agents that combine multiple tools for complex workflows—like an agent that monitors for long-running rides, checks customer history, sends notifications, and escalates to support. Our MCP server provides the building blocks; you create the automations that fit your operation.
Getting started is simple: 1) Sign up for Levy Fleets, 2) Enable MCP in your account settings to get your credentials, 3) Configure your AI assistant (Claude Desktop, custom agent, etc.) with the MCP server details, 4) Start asking questions and running operations. We provide setup documentation and our team can help with initial configuration.
Our 11 tools cover four categories: Customer Operations (lookup, search, wallet credits), Vehicle Operations (lookup, search, IoT commands like lock/unlock/beep), Ride Operations (lookup, search, refunds), and Fleet Analytics (statistics, active rides monitoring). Read operations provide instant data access; write operations execute real actions with audit trails.
Ready to Add AI to Your Fleet?
Be among the first to leverage AI-powered fleet management. Our MCP integration is included with all Levy Fleets plans—no additional cost.
Questions? Email us at sales@levyelectric.com