Find Any Site Document, RFI, or Submittal in Seconds

No. Mirage reads the documents where your team already keeps them, in SharePoint or your document store, and answers questions against them. Your document system stays the system of record; Mirage is the layer that reads it. For documents held in Procore, Procore's API terms exclude copying a project's whole store into an outside index, so Mirage works on the documents a workflow needs, record by record, or with Procore's own tools.

It searches indexed project documents and returns the answer with an exact citation: document name, page, and revision number. It doesn't summarize or infer beyond what the source document actually states.

Yes. Field teams access it through the credentials they already use, on devices already in use on site. No separate login, and no new app to install on top of what's there.

The conflict is surfaced, not resolved automatically. Mirage flags the documents and the specific values that disagree, and routes it to the person responsible for that document to decide.

Scanned documents are indexed the same way as native files. Handwritten field notes are the one input type where accuracy drops, and flagged answers reflect that lower confidence.

It updates as documents are added or revised, so answers reflect the current version by default. Every answer states which revision it was pulled from, so nobody has to double-check separately.

4 to 6 weeks to a production deployment: document infrastructure audit and indexing first, then a parallel pilot on one active site before rollout to the rest.

Pricing starts with a fixed-price pilot indexing one active project, so you see real answers before committing to anything larger. After that, it moves to an annual subscription priced by volume, quoted in your first conversation with us.