Connect AI assistants to MapTrack with MCP
MapTrack offers a hosted, remote MCP server, so Claude, ChatGPT, Cursor and any other MCP client can read and write your asset data directly. Ask in plain English to find assets, check what maintenance is due, raise and update work orders, and pull reports, all against your live, Australian-hosted MapTrack account.
Trusted by construction, mining and field service teams
What is the MapTrack MCP server?
The Model Context Protocol (MCP) is an open standard, introduced by Anthropic, that lets AI assistants connect to external tools and data in a consistent way. The MapTrack MCP server is a hosted connection that exposes your MapTrack account to any MCP client, so an AI assistant such as Claude or ChatGPT can read your asset register, check maintenance and work orders, create and update records, and run reports, all in plain language. It runs on the same data as the rest of the platform, including integrations and the public API and MapTrack's built-in AI features, so your AI assistant works from one live source of truth instead of a stale export.
Key takeaways
- MCP is an open standard from Anthropic, so the MapTrack MCP server works with any MCP client, including Claude, ChatGPT and Cursor.
- It is read and write: ask it to find assets and to raise or update work orders.
- Every answer and action runs against your live MapTrack account, not a copied export.
- No code to build: connect a client and ask in plain English.
Why connect AI to your asset data
Your asset data should answer questions, not hide them
Teams already run on MapTrack. The MCP server lets the AI assistant they already use reach that data directly, so the answers and actions people need are a single question away.
Answers live in five different screens
A simple question, what is overdue and where is it, means jumping between assets, maintenance, costs and reports. The MapTrack MCP server pulls it together so your assistant answers in one go.
Requests get stuck in an inbox
A field request to raise a job too often waits for someone at a desk. Through MCP, your assistant can create and update work orders directly, so a request becomes an action.
Exports are stale the moment you run them
A spreadsheet pulled this morning is wrong by lunch. The MCP server reads your live MapTrack account, so every answer reflects what is true right now.
Custom integrations need a developer
Wiring an AI tool into your systems usually means code and weeks of work. With a standard MCP server there is nothing to build: connect a client and ask in plain language.
How it works
From connection to answers in four steps
Connect once, then anyone on your team can ask MapTrack questions and take action through the assistant they already use.
Connect your MapTrack account
Authenticate the hosted MapTrack MCP server with your MapTrack account. Your data stays in your account.
Add it to your AI client
Add the MapTrack MCP server to Claude, ChatGPT, Cursor or any other MCP-compatible client your team uses.
Ask in plain English
Query assets, maintenance, costs and reports in natural language. No query builder, no exports, no code.
Let it take action
Ask it to create or update records, such as raising a work order, and the change is recorded against the asset in MapTrack.
What connecting looks like
A standard MCP connection, not a custom build
MapTrack runs a hosted, remote MCP server. Once it is enabled for your account, you add it to your AI client in a few lines and authenticate with your MapTrack account. Here is what a typical configuration looks like.
{
"mcpServers": {
"maptrack": {
"url": "https://mcp.maptrack.com",
"transport": "streamable-http"
}
}
}Example only. The MapTrack MCP server is available on paid plans. Contact our team to enable it and receive your account's connection details.
What your AI assistant can do
Read your data and act on it, in plain language
The MapTrack MCP server covers four jobs across the platform. Here is what each looks like, with examples you could ask today.
Query the asset register
Find assets and read their status, location, custody, specifications and history, across every site.
Check maintenance and work orders
See what is open, scheduled, due or overdue, and drill into the work orders on any asset.
Create and update records
Take action, not just answers: raise work orders, update assets and add notes, recorded against the asset.
Run reports and analytics
Pull the numbers that matter, from utilisation to cost, without building a report by hand.
Vendor-neutral by design
Works with any MCP client
MCP is an open standard, so MapTrack is not tied to one AI provider. Connect the assistant your team already prefers and switch whenever you like, without changing how MapTrack is set up.
Claude
Claude Desktop and claude.ai
ChatGPT
From OpenAI
Cursor
For developer workflows
Any MCP client
Open standard, your choice of assistant
Security and control
Connected to your account, on your terms
The MCP server is a connection to MapTrack, not a copy of your data somewhere else. Here is the short version, and the team can take you through the detail.
Connect with your account
Access is authenticated with your MapTrack account, so the assistant works within MapTrack, not around it.
Your data stays in MapTrack
The MCP server reads and writes your live MapTrack data in place. It is not copied into a separate store.
Australian data residency
Australian data residency is available, in line with the rest of the MapTrack platform.
Every change is recorded
Create and update actions are recorded against the asset and attributed to your account, so you keep a full history of what the assistant did, the same as any change made in MapTrack.
Built on the platform teams already trust
“Thanks to MapTrack we know exactly where our gear is, who has it and when it needs servicing.”
Denis Slaviero
HSEQ Manager at Hacer Group
Put your asset data one question away
See the MapTrack MCP server connected to a live account, and how your team can query assets, raise work orders and pull reports through the assistant they already use.
- Works with any MCP client
- Read and write
- See it on a live account
Works with the rest of the platform
MCP reads and writes everything else in MapTrack
Integrations & API
Connect telematics, accounting and ERP, or build custom links on the public REST API.
Explore integrations & api→AI features
AI form generation, smart data import and inspection summaries built into MapTrack.
Explore ai features→Maintenance & work orders
The work orders your assistant can raise, schedule and close, tracked on the asset.
Explore maintenance & work orders→Asset tracking
The live asset register your assistant reads from: custody, location and history.
Explore asset tracking→Reporting & dashboards
The utilisation, cost and compliance reports your assistant can pull on request.
Explore reporting & dashboards→Alerts & automation
Rule-based alerts and automations for overdue services, expiring tickets and low stock.
Explore alerts & automation→Forms & checklists
Digital pre-starts, inspections and checklists that feed the data your assistant reads.
Explore forms & checklists→MCP server FAQs
Common questions about connecting AI assistants to MapTrack with the Model Context Protocol.
- What is the MapTrack MCP server?
- The Model Context Protocol (MCP) is an open standard, introduced by Anthropic, that lets AI assistants connect to external tools and data in a consistent way. The MapTrack MCP server is a hosted connection that exposes your MapTrack account to any MCP client, so an AI assistant such as Claude or ChatGPT can read your asset register, check maintenance and work orders, create and update records, and run reports, all in plain language against your live MapTrack data.
- Is MapTrack an AI CMMS?
- MapTrack is an asset tracking and maintenance platform with built-in AI features and a hosted MCP server, so you can run a CMMS workflow through an AI assistant in plain language. Rather than a bolt-on chatbot, the MCP server connects your live asset register, work orders and reports to the AI client your team already uses, which makes MapTrack an AI-ready CMMS and asset tracking platform.
- Which AI assistants and clients can connect to MapTrack?
- Because MapTrack provides a standard MCP server, it works with any MCP-compatible client. That includes Claude (Claude Desktop and claude.ai), ChatGPT and Cursor, plus other assistants and agent frameworks that support the Model Context Protocol. You are not locked to a single AI vendor, so you can use the assistant your team already prefers.
- What can an AI assistant actually do through the MCP server?
- Four things. It can query your asset register (find assets, status, location, custody and history), read maintenance and work orders (what is open, due or overdue), create and update records (for example raise a work order, update an asset or add a note), and run reports and analytics (utilisation, costs and compliance summaries). It works on the same data as the rest of MapTrack, so answers and actions stay in sync with your team.
- Can the AI assistant make changes, or only read data?
- Both. As well as answering questions from your live data, the MapTrack MCP server lets an assistant create and update records, such as raising a work order against an asset or updating a field. Write actions run against your account and are recorded against the asset, the same as if a team member had made the change in MapTrack.
- How is the MapTrack MCP server different from a read-only MCP server?
- Some MCP servers only let an assistant read data. The MapTrack MCP server is read and write, so an assistant can both answer questions from your live data and take action, such as raising a work order or updating an asset. Every change is recorded against the asset, so you keep a full history of what the assistant did.
- Is there an MCP server for equipment, plant or fleet asset tracking?
- Yes. The MapTrack MCP server works across your whole asset register, including equipment, plant, tools, vehicles and fleet, not just maintenance records. An AI assistant can find any asset, read its history, and raise or update work orders against it, across every site you run.
- How is the MCP server different from the MapTrack API?
- They serve different jobs on the same data. The MapTrack REST API is built for developers wiring up custom integrations and automations in code. The MCP server is built for AI assistants, so a person can ask MapTrack questions and take actions in natural language with no integration to write. Many teams use both: the API for system-to-system links, MCP for day-to-day questions and actions through an assistant.
- Is the MCP connection secure and private?
- Yes. You connect using your MapTrack account, and your data stays in your MapTrack account rather than being copied elsewhere. Create and update actions are recorded against the asset and attributed to your account, so there is a full audit history. Australian data residency is available, in line with the rest of the platform.
- Does the MapTrack MCP server keep data in Australia?
- Australian data residency is available for MapTrack, in line with the rest of the platform, so your asset and maintenance data can stay onshore. The MCP server reads and writes that data in place rather than copying it into a separate store. Contact our team to confirm the setup for your organisation.
- Is the MCP server included in my plan?
- The MapTrack MCP server is available on paid plans. Once it is enabled for your account, anyone on your team can connect a supported AI client. Contact our team to enable it and confirm the right plan for your asset count.
- Do I need to be a developer to use it?
- No. Once the server is enabled for your account, connecting is a one-time setup in your AI client, such as Claude, ChatGPT or Cursor, and after that anyone on your team can ask in plain English. The setup is a few lines of configuration, not custom code.
Bring AI to your asset data
Connect Claude, ChatGPT or any MCP client to MapTrack and let your team query assets, manage work orders and run reports in plain language. Book a demo to see it on a live account, or get pricing for your asset count.
- Any MCP client
- Read and write
- Australian data residency