PDF Inspector & PDF to Markdown
Inspect up to 10 PDFs, recover scanned pages with OCR, and extract clean Markdown or separate table workbooks through a controlled processing queue.
One temporary API request per PDF · two process at a time · files are not stored
Extract PDF text, tables, and document structure
PDF Inspector rebuilds reading order, identifies OCR pages, extracts Markdown, and exports native-text tables without forcing unrelated tables into one CSV.
Powered by Firecrawl PDF Inspector v1.12.0
Firecrawl's open-source Rust parser handles native PDF classification, position-aware text extraction, reading order, layout signals, Markdown, and table detection. Tuko hosts the parser and adds automatic English and Malay OCR routing, API responses, and separate-sheet Excel export.
View the Firecrawl PDF Inspector project →Real-world business use cases
| Business use case | How PDF Inspector helps | Business impact |
|---|---|---|
| Finance reporting | Extracts tables from management reports into separate Excel worksheets for review and reconciliation. | Reduces manual rekeying and preserves each table's headers and columns. |
| Insurance operations | Separates benefit schedules, coverage tables, and definitions from native-text policy PDFs. | Speeds policy comparison while keeping unlike tables independent. |
| Procurement comparison | Extracts supplier pricing and specification tables into structured rows or Excel sheets. | Shortens quote comparison and reduces copy-paste errors. |
| Operations document intake | Detects scanned attachments inside mixed PDFs and runs English or Malay OCR only where needed. | Cuts manual data entry while preserving reliable native text. |
| Knowledge-base preparation | Converts selected manual, policy, or research pages into Markdown with layout and OCR diagnostics. | Accelerates searchable documentation and flags pages requiring human review. |
| Document-processing software | Provides multipart APIs for page selection, Markdown extraction, OCR routing, and table parsing. | Reduces the engineering work needed to add PDF ingestion to an application. |
Verify extracted values against the source PDF before making financial, legal, medical, or operational decisions.
Batch PDF processing without one oversized request
Select up to 10 PDFs and process two at a time. Each file receives its own API request, status, result, and retry path, so one unsupported document does not discard the rest of the batch. Download extracted outputs together as a ZIP; table downloads keep one workbook per source PDF and one worksheet per detected table.
Table extraction without broken columns
Run Extract tables to preview every native-text table. Download one table as CSV, or export all tables as an Excel workbook where each table receives its own worksheet. See the guide to extracting PDF tables to Excel without breaking their structure.
Automatic OCR for scanned PDF pages
Automatic mode sends only missing or unreliable pages through English, Malay, or combined OCR. It processes up to five flagged pages per request. For image files, use the Malay OCR tool.
Developer API
The extraction endpoint returns document metadata with Markdown or plain text. The table endpoint returns page numbers, headers, rows, Markdown, and CSV for every detected table. Use the interactive Swagger API reference, download the OpenAPI 3.0 JSON contract, or read the PDF extraction API guide.
curl -X POST \
-F "[email protected]" \
-F "format=markdown" \
-F "ocr=auto" \
-F "lang=eng" \
https://www.tuko.my/api/v1/pdf/extractTemporary server processing
PDF Inspector processes the selected file temporarily on the server and does not intentionally retain it. For local page organization, use the browser-based PDF organizer instead.
PDF extraction guides
PDF Inspector & PDF to Markdown FAQs
Common questions about using our PDF Inspector & PDF to Markdown tool.
What does PDF Inspector detect?
It classifies PDFs as text-based, scanned, image-based, or mixed and reports pages with tables, columns, encoding issues, or missing text layers.
Does this tool perform OCR?
Yes. Automatic OCR renders and recognizes only pages whose native text layer is missing or unreliable, then merges the recovered text into the result. OCR is capped at five pages per request.
Can I inspect multiple PDFs at once?
Yes. Select up to 10 PDFs of 15 MB each. The browser queues them, processes two separate API requests at a time, keeps failures isolated, and can download successful outputs or table workbooks together as ZIP files.
Are uploaded PDFs stored?
No. PDFs are processed for the current request and are not intentionally persisted by Tuko. Unlike Tuko's offline PDF tools, this tool currently requires a temporary server upload.
Can developers use an API?
Yes. POST a multipart PDF to /api/v1/pdf/extract for extraction and OCR, or /api/v1/pdf/tables for structured table rows and CSV data. Interactive Swagger documentation and an OpenAPI 3.0 JSON contract are available.
How are multiple PDF tables exported without breaking their structure?
Each detected table is written to a separate worksheet in one Excel workbook. Tables with different headers or column counts are never combined into a single CSV.
Related Tools
Need something else?
Explore All 50+ Tools