PageIndex

Primary document retrieval tool. After orienting with `get_folder_structure()` (when available) — and, for queries that may involve typed attributes, pairing it with `get_metadata_schema()` so you kno…

Get detailed information about a specific document by name. Requires: doc_name (string). Optional: wait_for_completion (boolean) to automatically wait up to 3 minutes if document is still processing. …

Retrieve an embedded image from a document. Requires image_path from get_page_content() response, format: <docName>/<imagePath> (e.g. MyDoc.pdf/figures/fig1.png). Returns image_base64 and content_type…

Extract the hierarchical structure of a completed document. Optional: wait_for_completion (boolean) to wait up to 3 minutes if still processing. Returns structured outline with headers, sections, and …

Orientation step: show the folder hierarchy as a tree (like `tree -d`). **Call this before browse_documents()** to plan targeted retrieval — folder names reveal content domains (e.g. "Research", "Fina…

List declared metadata fields (names, types, descriptions) so you can build `metadata_filter` / `sort` / `aggregates` / `group_by` for `search_documents(mode="metadata")`, or a `metadata_filter` that …

Extract specific page content from processed documents. Flexible page selection: single page ('5'), ranges ('3-7'), or multiple pages ('1,5,10'). Use tight, targeted ranges and split into multiple cal…

Upload and process PDF documents from public URLs. Only invoke with a valid, accessible PDF URL. If the user hasn't provided one, ask for it instead of guessing. Returns document name for subsequent o…

Permanently delete documents and all associated data. Only invoke when the user explicitly names the documents AND confirms deletion. Accepts an array of document names for batch deletion (maximum 10 …

Escalation tool — never the first discovery step. Always start with `browse_documents()` (after `get_folder_structure()` when visible). Pick a mode: • mode="query" — keyword precision when browse miss…