Skip to main content
Two engineers talking over a laptop in a plant control office, with process screens and piping behind themAll Features

Ask your assets anything

AssetLab ships a native MCP server, so Claude, ChatGPT and Microsoft Copilot connect straight to your asset data. Pull reports, create work orders, and check compliance in plain English - answers in seconds, not hours.

Setting it up? Connection steps for Claude, ChatGPT and Copilot

Skip the dashboards

Ask in plain English and get answers grounded in your live asset data. No menus to navigate, no filters to set, no training required - just a conversation.

Work gets done in the chat

Describe an issue and the AI creates a fully detailed work order with the right priority and assignment. Reports, lookups, and inventory checks work the same way.

You set the boundaries

Read-only or read-write, scoped resource by resource, with every call logged against its key. The AI works inside the limits you set - never around them.

Talk to your asset data like a colleague

An Administrator creates a scoped API key in Settings, then you add AssetLab to Claude or ChatGPT as a connector, or to Microsoft Copilot through a Copilot Studio agent - no coding, no plugins to install. From that moment, your assets, work orders, compliance records, and costs answer in plain English, grounded in your live data.

  • Native MCP server - one connector URL and your API key
  • One scoped key per person or purpose, revocable instantly
  • Ask follow-up questions in the same conversation
  • Build custom workflows on the full REST API - documented at app.assetlab.ca/docs with an interactive reference
How the AI connection works →
AI assistant · connected to AssetLabLive
One question, one answer

"What assets need replacement this year?"

23 assets across 3 sites have passed their expected useful life. The highest priority is the HVAC system at Building A - risk score Critical, estimated replacement cost $145,000. Want the full replacement report?

No dashboards to navigate, no filters to set - just ask
Asset query · over MCP
The AI finds the answer
"What's past its useful life?"
HVAC-04 · Building ACritical · $145K
Boiler B2 · Main CampusHigh · $98K
Elevator EL-02 · Building AHigh · $74K
RTU-7 · Westside CampusMedium · $32K
Assets past expected life23 · 3 sites
One question - not an afternoon of filters and exports

From question to work order in one sentence

The AI navigates AssetLab for you. Ask for a report and it pulls the numbers. Describe an issue and it creates a fully detailed work order with the right priority and assignment - then asks if you want the building manager notified.

  • Any report in plain English - replacements, compliance, costs
  • Work orders created with priority, assignee, and due date
  • Asset condition, warranty, and history in one question
  • Stock levels and reorder points across every storeroom
  • Inspection form templates authored from a manufacturer's manual - over MCP or the REST API
Explore work order management →

Powerful AI, on your terms

You stay in control. Every connection is authenticated with a key you create, every permission is set by you, and every call is logged. Read-only for reports or read-write for work orders - entirely up to you.

  • Decide exactly what the AI can see and do
  • Grant read or write access resource by resource
  • Every call logged with the identity of the key that made it
See reporting & dashboards →
Security & control
The AI works inside your limits
Read-only or read-write - you decide
Access scoped resource by resource
AssetLab never trains AI on your data
The MCP server stores no records
Every call logged against the key that made it
AI actions audited100%
Authenticated connections only - revoke a key and access stops instantly

And everything else you'd expect

Native MCP server

Connect Claude, ChatGPT or Microsoft Copilot

Plain-English queries

Any report without dashboards or filters

AI work order creation

Priority, assignment, and due date set

Asset lookups

Condition, warranty, and history in seconds

Trend analysis

Cost and repair patterns over time

Inventory checks

Stock levels across every storeroom

Scoped permissions

Read or write access, resource by resource

Full audit trail

Every AI call logged against its key

Full REST API

Build your own workflows and integrations

Interactive API reference

Docs you can read alongside your key's scopes

Forms via API & MCP

Let an assistant author inspection templates

“I spend a lot of time walking this facility, and repairs and issues present themselves constantly. Being able to use voice-to-text and upload photos on the spot - and have that translate directly into a structured work order for the right staff member - means I'm not carrying a mental list all morning. The work gets captured when I see it, and it gets to the right person quickly.”
Mark Perkins
Facility Operations Manager, Calgary Winter Club
Read the full story →

AI Integration FAQ

Common questions about connecting AI assistants to your CMMS, how the MCP server works, and how your data stays under your control.

What is AI in CMMS software?
AI in a CMMS lets you work with your asset data through conversation instead of menus. With AssetLab, an assistant like Claude, ChatGPT or Microsoft Copilot connects to your live records, so you can pull reports, look up assets, check compliance, and create work orders by asking in plain English - no dashboards to navigate and no training required.
Which AI assistants work with AssetLab?
Claude, ChatGPT and Microsoft Copilot connect to AssetLab through its native MCP server, as can any other MCP-compatible client - no coding and no plugins to install. Copilot connects through an agent built once in Copilot Studio and published to Copilot chat. Teams that want custom workflows can also build on AssetLab's full REST API.
What is an MCP server?
MCP (Model Context Protocol) is an open standard that lets AI assistants connect securely to outside systems. AssetLab ships its own MCP server, so the AI queries your asset data through a secure, authenticated channel in real time. There are no bulk data exports, and the MCP server passes records through without storing them.
Is my asset data used to train AI models?
AssetLab does not use your data to train AI models. An assistant sees only the records its API key's scopes allow, and how it handles what it receives is governed by your agreement with that AI provider. AI access is off until an Administrator creates a key for it.
Can AI make changes to my CMMS data?
Only if you allow it. You choose read-only access for reports or read-write access for creating work orders, and you grant read or write access resource by resource. Every call the assistant makes is logged with the identity of the key that made it, and revoking the key cuts access instantly.
Do I need technical skills to set it up?
No coding is needed. An Administrator first creates a scoped API key under Settings → API Keys. You then add the connector URL to Claude or ChatGPT and paste the key when AssetLab asks for it; Microsoft Copilot connects through an agent built once in Copilot Studio. From there, anyone who can type a question can use the full CMMS through conversation.
Where is the API documented?
At app.assetlab.ca/docs, alongside an interactive API reference you can read against the scopes of your own key. The REST API covers the same records the MCP server does - assets, work orders, inspection form templates, day plans (work order schedules), and more - and the docs' What's new page lists each release with a link to its documentation.
A maintenance technician with a tablet inspecting pumps in an industrial plant

Built for the people who make things happen.