AttributeGroup

One tag's share of the whole matching corpus.

Keyed by `(workspace, tag)`, because a TagId belongs to one workspace's table and the tags array is per table too. A multi-workspace query gets a group per table rather than one merged heading.

The id is the one a live table can name. A tag merged into another re-points no node ([ADR-176]), so nodes go on carrying the merged-away id; the store folds those counts into the surviving tag before answering, so a merge shows as one group rather than splitting into a named half and a nameless one. The consequence worth knowing: the id here is not necessarily an id any row's koru.tag value spells.

[ADR-176]: ../../../docs/decisions/ADR-176-200.md#adr-176-consenton--telemetry-consent-defaults-on-and-a-discoverable-opt-out-is-what-buys-it

Type object

Shape

Field	Type	Description
workspacerequired	WorkspaceId	Which workspace's tag table names this id.
tagrequired	string	The `TagId` in storage form. Look it up in `tags` for a name and color.
countrequired	integer (uint, >= 0)	Distinct nodes carrying it across the whole matching corpus - not the page. A node with two tags counts once in each group, so the counts sum to more than `rows.len()` even on a complete page.
trashedrequired	integer (uint, >= 0)	How many of `count` sit in a binned document. Binned rows are counted and flagged rather than dropped, exactly as they are in `rows` - a heading that hid them would disagree with the list underneath it.

Example

{
  "workspace": {},
  "tag": "string",
  "count": 0,
  "trashed": 0
}

The contract has no example values, so example bodies are generated from each schema: every field is shown with a placeholder of its type.

Generated from contract version 1. Last updated November 12, 2026.