Document Text Extractor: PDF, Word, Excel to Markdown
Pricing
from $4.00 / 1,000 document extracteds
Document Text Extractor: PDF, Word, Excel to Markdown
Turn PDF, DOCX, PPTX, XLSX, HTML, CSV and EPUB files into clean Markdown or plain text, with page counts, metadata and optional RAG-ready chunks. Failed documents are free.
Pricing
from $4.00 / 1,000 document extracteds
Rating
0.0
(0)
Developer
Pipwort Data
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Share
Turn documents into clean text your tools and AI models can use. Give it links to PDF, Word, PowerPoint, Excel, HTML, CSV or EPUB files, and get back Markdown or plain text, page counts, metadata and, if you want them, ready-made chunks for embeddings.
You pay only for documents that convert successfully. Failed downloads, unsupported files and broken documents are listed with a clear reason and cost nothing.
Supported files
PDF · DOCX · PPTX · XLSX · XLS · HTML · CSV · JSON · XML · TXT · MD · EPUB
The file type is detected from the file itself, not just the URL, so links without an extension still work.
What you get for each document
markdownand/ortext: headings, lists and tables kept as Markdown, or plain text if you preferpageCount(PDF pages or PowerPoint slides),sheetCount(Excel)title,author,createdAt,producerwhere the file includes themwordCount,charCount,fileSizeBytes,sha256(handy for spotting duplicates)chunks(optional): text split on paragraph boundaries at your chosen size, with optional overlappages(optional): the text of each PDF page separatelylikelyScanned: a flag for image-only PDFs, which need OCR (not included)
Input
| Field | What it does |
|---|---|
documentUrls | Links to the files (required). You can paste a list or upload a text file. |
outputFormat | markdown (default), text or both |
chunkSize, chunkOverlap | Turn on chunking for RAG (characters) |
includePages | Add per-page text for PDFs |
maxPagesPerDocument | Read only the first N pages of long PDFs |
maxFileSizeMb | Skip files above this size (default 50 MB) |
cleanLayoutNoise | Remove rotated margin text and stray dots from PDFs (on by default) |
Example output (shortened)
A failed document looks like this, and is free:
Pricing
- US$0.004 per document converted (includes up to 20 pages or slides)
- US$0.0002 per extra page beyond 20, so a 100-page PDF costs US$0.02
- A small start fee of US$0.005 per run
Use maxPagesPerDocument or the run's "Maximum cost" setting to keep costs predictable. The Actor stops cleanly at your limit.
Good to know
- Scanned PDFs (pictures of pages) contain no text layer, so they come back nearly empty with
likelyScanned: true. OCR is not included. - Password-protected PDFs are reported as
errorType: encrypted. - Text from complex PDF layouts (forms, multiple columns) is extracted in reading order as well as the file allows. It may need light tidying.
- Only process documents you have the right to use. Files are handled in memory during the run and are not stored anywhere else by this Actor.
Support
If a file converts badly, open an issue with the link (if it's public), and we'll improve the converter.
