/search

1 route.

A substring search across every resource, run on the decrypted snapshot of the request. There is no server-side index and there cannot be one.

GET /v1/search

Search everything

A substring match over to-do titles, subtitles, descriptions, subtask titles and tag names; container names, subtitles and descriptions; NOTE titles, bodies and tag names, when the credential also holds notes:read; capture-row text, urls and titles, when it holds inbox:read; and attachment titles, urls and text, when it holds attachments:read (a note’s own attachments too, while the note is live). A note in a folder is searched under its folder (whose name it also matches); a note in a trashed folder is a trashed hit, and its attachments are not searched. A kind the credential cannot read is left out, never refused. A note is matched on its Markdown source, so you find what you typed, except that an embed ([title](note:ID) and the like) is matched by its name, never by the id inside it. ⚠ It runs on the decrypted snapshot in this request: the server has no search index and cannot have one. Ordered by KIND (to-dos, then containers, then notes, then capture rows, then attachments) and by id within a kind, and cursor-paged like every other list. There is no relevance score to sort by. Trashed rows are left out unless include_trashed=true: then a trashed to-do, note or container, or one under a trashed container, comes back with trashed: true (an archived subtree never does, and an attachment or capture row is never returned as trashed).

Scopes: todos:read, containers:read

Query

Response

• 200 SearchResult. One page of hits, with the field each one matched on.

Errors: grant_expired * grant_revoked * unauthorized * forbidden_scope * account_too_large * validation_failed * rate_limited * internal