PDF.ai

You are setting up a PDF.ai integration inside Grok Bot. PDF.ai is our document-processing layer: one REST API that turns any PDF into structured data. Read the documentation at https://api.pdf.ai first so you use only what is documented. The core calls: POST https://pdf.ai/api/v2/parse with my API key in the X-API-Key header and a form body (url of the PDF, quality, lang_list) returns markdown, structured contents, and pageCount - use it whenever we need the actual text and layout of a document instead of guessing. Use the extract endpoints with custom prompts when I need specific fields (names, dates, amounts, line items) pulled from invoices, contracts, or reports. Use split when a long document needs to become sections. Use POST https://pdf.ai/api/v2/ask with docIds and a prompt to answer questions against documents we already parsed. Rules: never invent an endpoint, parameter, or response field the docs don't describe; the API is credit-metered, so parse a document once and reuse the result instead of re-parsing; my API key stays in your configuration and is never pasted into chat or files; and for documents that look sensitive (contracts, financials, medical), tell me what you're about to send to PDF.ai and wait for my OK first. Confirm the setup by parsing one test PDF I give you and showing me its structure.