Find readable documents in the active scope. Use mode=list for paginated browsing by folder or tag UUID; use mode=search for hybrid text and semantic retrieval with tag names.

Read one document by UUID. Choose detail for content and metadata, markdown for portable export, or versions for saved revisions and snapshots. All views are read-only.

Create or update one document. Use action=create for a new document or action=update for an existing one; batch_documents handles the same placement change across multiple existing documents. Creation requires write access; content edits require edit access, and placement changes may require write access. Content or placement changes queue normal indexing.

Move one document to trash without purging content or versions. Use batch_documents(action=trash) for multiple explicit IDs. Requires write access and an explicit delete request; permanently_delete_document is a separate irreversible action.

List visible folders, categories, or tags in the active workspace or category scope. Set kind=folders to inspect root folders or children of parentId.

Create or update a folder. action=update can rename or move an existing folder; the API enforces category boundaries and queues affected documents for reindexing.

Delete a folder while preserving its documents; child folders and documents are detached under the API rules. Requires write access.

Create or update a workspace category. Full workspace write access is required; category-scoped credentials cannot manage categories.

Delete a workspace category and detach its documents without deleting them. Requires full workspace write access; category-scoped credentials are denied.

Create or update a workspace tag. Full workspace write access is required; renaming or recoloring a tag causes affected documents to be reindexed.

Delete a workspace tag and remove its document assignments. Full workspace write access is required; documents themselves remain.

Add or remove an existing tag assignment on one document. Use batch_documents for the same tag action across multiple explicit IDs. Requires edit access to each document; action=remove preserves the tag itself.

Save a named snapshot of an existing document before a planned change. The snapshot can later be restored with restore_document_version.

Restore content from a version or snapshot listed by read_document(view=versions). This changes current content and queues indexing; it does not restore a trashed document.

Explore graph relations from readable seed documents. mode=neighbors traverses one document without a query; mode=search ranks relations from one or more seed IDs with optional query text.

Read current indexing and pipeline status for one document without starting work. Use refresh_document_index only when an explicit retry is intended.

Queue an explicit asynchronous reindex for one document; use batch_documents(action=refresh_index) for multiple explicit IDs. Routine content and placement changes already schedule indexing; check get_document_index_status first.

List soft-deleted documents visible to the active workspace or category scope. Requires read access and does not restore or purge anything.

Return one document from trash to the active library; use batch_documents(action=restore) for multiple explicit IDs. This restores the document itself, unlike restore_document_version which changes content.

Irreversibly purge a document already in trash, including its version history. Use only on an explicit user request after checking list_trash.

Use for multiple explicitly selected documents that need the same action; for one document prefer the matching single-document tool. Accepts 1–25 IDs for automation compatibility. Supports move, set_category, add_tag, remove_tag, trash, restore, or refresh_index. Processes sequentially through the existing API with a permission check per document and a per-ID result; partial success is possible. Stops after authentication or rate-limit failure. Never expands a search result or permanently purges documents; version restore is separate.

Read one visible folder, category, or tag by UUID. Folder uses its direct API getter; category and tag lookup uses the complete permission-filtered lists. An absent or out-of-scope ID is reported as not found. Use list_workspace_structure to discover IDs first.