LLMWhisperer

ProductivityCredentials form

Extract layout-preserving text from PDFs, scans, images and office documents for LLMs with Unstract's LLMWhisperer API.

12 actions

Action	Effect	Data handled
llmwhisperer_extract_documentUploads a document (PDF, scan, image, Word or spreadsheet) and starts text extraction	write	Personal data
llmwhisperer_extract_from_urlStarts text extraction of a document LLMWhisperer downloads from a public URL	write	Personal data
llmwhisperer_get_statusReturns the status of an extraction: accepted, processing, processed, error or retrieved	read	General
llmwhisperer_retrieve_textReturns the extracted text of a processed document, with confidence metadata unless textOnly is set	read	Personal data
llmwhisperer_get_detailReturns the details recorded for one extraction, such as its mode, page counts, tag and timing	read	General
llmwhisperer_get_highlightsReturns the page position of extracted lines, for highlighting text in the original document	read	General
llmwhisperer_get_usage_infoReturns the account's plan, quotas and pages processed today and this month, per mode	read	General
llmwhisperer_get_usage_by_tagReturns the pages processed per mode for one tag, optionally within a date range	read	General
llmwhisperer_register_webhookRegisters a named webhook that LLMWhisperer calls with the extracted text when an extraction started with that webhook name finishes	write	General
llmwhisperer_get_webhookReturns the URL and token of a registered webhook	read	General
llmwhisperer_update_webhookChanges the URL or token of a registered webhook	write	General
llmwhisperer_delete_webhookDeletes a registered webhook	destructive	General

Sign-in

Also included

dexby.createSession({
  userId: customer.id,
  connectors: ['llmwhisperer']
})