Root, the AI assistant
Root, an AI assistant for maintenance, shows its working.
An agent with real tools over your records, a scope it cannot widen, and an approval in front of every change.
What it does
Grounded answers with citations
Questions are answered from the tenant’s own records and published procedures, retrieved through a search index that covers every record in the tenant. When an answer draws on retrieved records, those records are shown beside it, so an engineer can check it against the source instead of taking it on trust.
Two retrieval tools, not one pipeline
The agent chooses between searching by meaning and querying the equipment knowledge graph, in a bounded reasoning loop. Tenant and organisational scope are fixed on the server before it sees a tool, so the model cannot reach outside them.
Nothing changes without your approval
Root is offered only the actions your role allows, and a proposal missing a required field cannot become a card. A proposal changes nothing: it waits as a card for your approval.
Allow-listed actions, approved one by one
Root may propose an action from a server-side allow-list on any turn. Nothing runs until you approve that specific action, and each is approved on its own. The approved change goes through the same permission-checked path the dashboard screen uses.
Root-cause explanations from work history
Ask why an asset keeps failing and the retrieval runs over that asset’s own closed job cards, certificates and status history, with those records shown so you can check the answer against them.
Downloadable reports and draft checklists
Root produces reports, extracts and draft plans as documents you can download. It is instructed to build them only from records retrieved out of your own tenant and to say so plainly when a fact is missing. Every one is a draft for a competent person to review, not an authority to act on.
Chat attachments are read, not absorbed
A file attached to a conversation is extracted to text for that conversation only. Promoting it into durable, searchable knowledge is a separate action you approve.
Your own history, under your control
Conversations and generated documents are owner-private. You browse, reopen, archive and delete your own.
How it fits the rest of the platform
Root owns no business rules. It reads what every other module publishes and, when you approve a change, runs it through that module’s own rules. Adding a module to a tenant’s subscription widens what Root can answer about, with no work on Root itself.
Questions about Root
Can Root change data on its own?
No. Root can only propose actions from a server-side allow-list, and only ones your role allows. Nothing executes until a person approves that specific action, one at a time.
Where do its answers come from?
From the tenant’s own records, retrieved through the tenant’s own search index and equipment knowledge graph. Where an answer draws on retrieved records, those records are shown with it.
Does it see data outside my organisational scope?
No. Tenant and organisational scope are bound server-side into the tools the agent can call, rather than asked for in the prompt.
See it against your own equipment data.
A walkthrough on your asset classes, your certificate disciplines, your job cards, not a canned demo.