Core concepts: workspaces, templates, cases, fields
Four concepts explain almost everything in Document Blueprint — workspaces group cases, templates shape outputs, cases hold data, and fields define what gets extracted or written.
Last updated
What you need
- No prerequisites — this article defines the vocabulary used everywhere else
The four pillars
Workspace
The top-level container for one document process. Owns an inbox, a dashboard, a set of templates, and the rules that route files into cases. Workspaces live inside an organization (the tenant layer that holds members, roles, and billing).
Template
The blueprint for a generated document. Defines fields, their outputs, page placements, and output naming. A template carries up to two tracked capabilities: Extract (pulls AI-extracted data from incoming files — set when any field has source='extracted' and at least one source tab references a source PDF) and Generate (assembles output pages via the template's source tabs). Form questions (fields with source='question') are configured on the fields themselves and aren't a separately tracked capability.
Case
A unit of work — one invoice, one permit, one contract. Holds files, extracted data, activity history, and generated outputs. Sharing happens through two peer panels: Team (workspace members) and Clients (external collaborators), both scoped by file category visibility.
Field
A named concept on a template with a source (question / extracted / static) and one or more outputs (each producing a value in case.data). The stable output keys are the data paths into case.data — renaming a field or output doesn't migrate data. See glossary for the output kinds.