{"name":"PDF/Image to JSON Extraction API","description":"This API downloads a file from a Supabase path, converts it to images, and extracts financial document data using the BAML client. Supported formats: PDF and common image formats. Supported document types: bank statements, balance sheets, income statements, cash flow statements, and existing credit documents.","endpoints":{"/":"API info","/ocr":"POST endpoint accepting one or more Supabase paths to extract OCR data then CEL evaluation. Supports document_type parameter including 'existingcredit' for credit documents and 'corpdeck' for PowerPoint/PDF company decks. Automatically detects financial statement pages using RapidOCR - no need to specify page_range for balance_sheet, income_statement, or cash_flow_statement.","/ocr/financials":"POST endpoint accepting company_id to process all financial statements (balance_sheet, income_statement, cash_flow_statement) and existingcredit documents from uploads/{company_id}/202x_afs, uploads/{company_id}/afs, uploads/{company_id}/mfs, or uploads/{company_id}/existingcredit files. Auto-detects pages and overrides cache.","/ocr/xlsx":"POST endpoint accepting one or more Supabase paths to parse Excel files (XLS/XLSX) from storage. Supports document_type parameter for special parsing (e.g., 'aprap', 'business_customer').","/ocr/cache/{file_path}":"GET endpoint to retrieve cached OCR results","/ocr/xlsx/cache/{file_path}":"GET endpoint to retrieve cached XLSX results","/api/v1/statements/upload":"POST endpoint for bank statement upload with async processing via webhooks (v1 API, auth required)","/api/v1/statements/{subject_id}/summary":"GET endpoint to retrieve processed statement summary (v1 API, auth required)"}}