Skip to main content
Lists the documents in a document collection, or in every collection in the workspace, together with how far indexing has got. Display name “List Documents”. Agents use it to see what exists before searching, to answer questions like “is there a police report?”, to page through every document of one type, and to check whether index_documents has finished.

​
Authentication and enablement

  • No integration needed.
  • Off by default. Add it to an agent’s tools.
  • The workspace must have document collections switched on (runtime flag document_collections.enabled).

​
Inputs

  • collection (optional): collection key(s), matched case-insensitively.
  • scope (optional): collection (the default when collection is given) or workspace.
  • filters (optional): doc_type, date_from / date_to (YYYY-MM-DD), metadata, file_id, source_agent_group_id, source_task_id, source_task_source.
  • limit (optional): default 50, maximum 200.
  • offset (optional): for paging.

​
Output

  • Indexing progress for each named collection: files indexed out of the total, pages indexed, files still pending, and every file that was skipped or failed with the reason. “Indexing complete” appears only once nothing is pending.
  • A line per document: file name, file_id, type, date, page count, status and title.
  • “showing X-Y of N”, plus the offset for the next page.

​
Limits and side effects

  • Read-only.
  • Same visibility as read_document.

​
Expected errors

  • Document collections not enabled.
  • scope set to library. Library documents are listed on the Documents page instead.
  • A malformed filter.