omni-content-explorer

Run omni documents list --include labels -o json, paginate with --cursor until pageInfo.hasNextPage is false, then filter records whose labels array contains the target label. The omni content list command does not support a --labels filter directly.

Call omni dashboards download <identifier> --body '{ "format": "pdf" }' to start an async export job. If a job ID is returned, poll with omni dashboards download-status until completion; supported formats are pdf and png.

Silent failures usually happen when using the id field instead of the identifier field. The id field is null for workbook-type documents, so always use identifier for API calls and URL construction.

Omni CLI profiles authenticate with an API key or OAuth 2.1. If a call returns 401, run omni config login <profile> in an interactive session to complete the browser-based OAuth flow; avoid it in headless or CI environments.

Use omni documents get-queries <identifier> to retrieve the query definitions behind a dashboard's tiles. This lists the fields each tile computes and supports re-running those queries with the omni-query skill.

Some exports fail before a job is created, for example with a 'Cannot use in operator to search for query_id' error. In that case, do not call download-status; report the dashboard identifier and exact API error instead.