knowledgeBases
Load the SOPs, manuals, reports, spec sheets and records you already own. A knowledgeBase sets exactly what the platform is allowed to answer from, so every answer stands on sources you approved and names the one it came from.
knowledgeBasesYou already own the answer. That is the frustrating part.
Nothing here is a knowledge gap. It is a retrieval gap, and it costs the same.
The answer is in the archive. Nobody can say which file.
A 1985 hydrostatic test record, a superseded coating spec, the report that settled this exact question four years ago. It is all in there, which is not the same as being findable.
The folder nobody has opened since the merger.
Two drives, three naming conventions and a shared inbox. New hires do not know it exists; the people who do are the ones you least want interrupted.
A general AI tool will answer from anywhere.
Point a public assistant at a technical question and it will happily answer from the internet, from a superseded edition, or from nothing at all, in the same confident tone.
The sources, governed.
Your documents become the only thing the platform answers from: scoped, current, and cited.
Bring the documents as they are
PDFs, spreadsheets, drawings, scanned manuals and records ingest as-is, into your own tenant. No re-foldering project, no cleanup phase before you see value, and nothing used to train outside models.
Scope is the product
A knowledgeBase defines what the platform may answer from. Approve a set of sources and every answer stands on those, so the wrong edition and the wrong document simply are not in play.
Everything downstream inherits it
Chat, viewXpert, Agentic Apps and tutorXpert all run on these knowledgeBases. Fix a source once and every surface is fixed, because there is only one place the answers come from.
Point it at the folder your documents live in.
Not an import, a connection. The knowledgeBase keeps pace with the source, and you decide what is in scope.
Scope is not a setting. It is the whole idea.
This is the line between a knowledgeBase and a search box pointed at a drive.
Role-scoped
Decide which teams reach which knowledgeBases. Commercial terms stay with the people who negotiate them; the field sees the field manual.
Current, not historical
Connect the folder your documents already live in and the knowledgeBase keeps pace with it. Supersede an edition and answers move with it.
It tells you rather than reaching
Ask something the approved sources do not cover and it says so. That is a property of the scope you set, not a setting you have to remember to turn on.
Answers name the source they came from, and you can open it and read it yourself, that is viewXpert. The access model, SSO and audit side lives on Enterprise.
Connect the systems you already run on
knowledgeXpert reads from where your documents already live and writes results back to your systems of record, no rip-and-replace.
Connector availability varies by plan and deployment, and we add connectors, including custom systems via API, to fit your stack.
One Hub. Every governed tool.
knowledgeBases is the substrate. The other six governed tools all answer from it.
Agentic Apps
Workflow automation: turn a request you repeat into a finished, cited deliverable.
Explore →All of it behind one launcher. See The Hub →
Straight answers.
How much cleanup do we have to do first?
None to start. Documents ingest as they are, in the folder structure you already have. We do look at how an archive is organised before it is loaded, because how a document was ingested decides how precisely an answer can cite it, and that is far cheaper to get right at ingest than afterwards.
Is our data used to train AI models?
No. Your documents stay in your isolated tenant, are never used to train public AI models, and are never shared with other customers. You can export them, and if you leave we delete the tenant within the timelines in your contract.
Can different teams have different knowledgeBases?
Yes, and that is the normal setup. Role-based permissions decide who reaches which knowledgeBases, down to an individual source, and every access is logged. Enterprise covers the SSO and administration side of that.
What about standards and codes we do not own?
Public and regulatory material can sit alongside your own records, from the Marketplace. Copyrighted industry standards such as API, ASME and AMPP/NACE are not ours to host and we do not. An answer names which source it came from either way, so you can always see whether you are standing on your own record or a published one.
How current does a knowledgeBase stay?
Connect a folder in Google Drive, SharePoint, OneDrive or Dropbox and documents stay in step with it as they change. You decide what is in scope; nothing is pulled in that you have not approved.
See your own documents become a cited knowledge base.
Thirty minutes, your own documents, every answer cited to its source.