Document to Markdown — Word, PowerPoint, Excel, EPUB

Give it URLs of Word, PowerPoint, Excel, OpenDocument, EPUB, RTF or HTML files; it returns each as Markdown — headings, lists, tables and slide or sheet structure preserved. Use it when you hold a link to a .docx, .pptx, .xlsx, .odt, .odp, .ods, .epub, .rtf or .html file and need its content as text. Agents are routinely handed office files they cannot open: one call returns the content with its structure, sheets as Markdown tables and slides as sections, and pricing is per document, so a file costs a fraction of a cent.

Call it from an agent

Through Apify's MCP server (https://mcp.apify.com), tool call-actor:

{"actor": "hermeskd/document-to-markdown", "input": {"urls": ["https://raw.githubusercontent.com/python-openxml/python-docx/master/tests/test_files/test.docx"]}, "callOptions": {"maxTotalChargeUsd": 0.05, "timeout": 600}, "waitSecs": 30}

The call returns the run's status and its defaultDatasetId. Read the records with get-dataset-items:

{"datasetId": "<defaultDatasetId from the call>", "limit": 50}

As a dedicated tool, add https://mcp.apify.com?tools=hermeskd/document-to-markdown to your MCP client. A single file finishes well inside the 30-second wait, so one call-actor plus one get-dataset-items is the whole exchange. maxTotalChargeUsd caps what the run can charge; when it is reached the remaining URLs come back as charge_limit_reached records and are not charged.

Input

Field	Type	Required	Default	Meaning
urls	array of strings	yes	—	File URLs (.docx, .pptx, .xlsx, .odt, .odp, .ods, .epub, .rtf, .html), http or https, 1–50 per run, up to 50 MB each. The format is read from the file, not its name.
sheetRowLimit	integer	no	1000	Rows read from each spreadsheet sheet, the header row included (1–50,000).
includeNotes	boolean	no	true	Include each slide's speaker notes under a Notes heading.

Every limit above is enforced by the input schema and again in code.

Output

One record per input URL.

Field	Meaning
url	The input URL.
ok	true when the document converted; false with error.
format	The format read from the file's bytes: docx, pptx, xlsx, odt, odp, ods, epub, rtf or html.
markdown	The document as Markdown. Word and OpenDocument text: headings, nested and numbered lists, tables, bold and italic, links. PowerPoint and OpenDocument presentations: ## Slide N: title, text as bullets by level, tables, notes. Excel and OpenDocument spreadsheets: ## Sheet: name and a table, the first row as its header. EPUB: the chapters in reading order. RTF: the paragraphs (styling is not read). HTML: the whole page body. Images and document properties are left out.
sections	For slides and sheets: {index, title} per slide or sheet, in order; empty for other formats.
truncated	true when a sheet had more rows than sheetRowLimit, or the Markdown was cut at 6 MB to fit one dataset record.
error	{code, message} when ok is false.

Example record for the call above, as this Actor returned it in its deploy check on 2026-10-16:

{
  "url": "https://raw.githubusercontent.com/python-openxml/python-docx/master/tests/test_files/test.docx",
  "ok": true,
  "format": "docx",
  "markdown": "# python-docx was here!\n\npython-docx was here too!",
  "sections": [],
  "truncated": false
}

Events and prices

Pay per event; platform usage is included in the price, never passed through.

Event	Title	Charged when	FREE / BRONZE	SILVER	GOLD / PLATINUM / DIAMOND
document-converted	Document converted	One file converted to Markdown.	$0.002	$0.0018	$0.0016

Apify also charges its standard apify-actor-start event: $0.00005 per GB of run memory (2 events, $0.0001, at the default 2,048 MB). The minimum maximum-charge a caller can set is $0.01.

Worked example on the FREE tier: 10 files cost 10 × $0.002 = $0.02, plus $0.0001 for the start, $0.0201 in all. A file is one document whatever its length; nothing is charged for a failed file.

Limits and run time

• 1–50 URLs per run; files up to 50 MB; a zip-based file (Office, OpenDocument, EPUB) may unpack to at most 500 MB in at most 20,000 parts; up to 50,000 rows per sheet (1,000 by default).
• Each request — its redirects and its whole body included — must finish within 30 seconds (the per-request timeout; a slower file is download_failed); at most 2 requests a second go to one host.
• One record holds at most 6 MB of Markdown; a longer document is cut there and marked truncated.
• One document may take at most 120 seconds to convert; a file that needs longer is stopped and answered too_large, and the next document in the run starts fresh.
• Typical run time: a file converts in well under a second; a run of one file takes a few seconds at the default memory (2,048 MB). The default run timeout is 600 seconds.

Errors

Errors are records with ok: false and error.code; none is charged.

error.code	Meaning
download_failed	The URL did not return the file (status or network error).
unsupported_format	Not one of the listed formats (the message names the Hermes Actor for a PDF or an image; legacy .doc, .xls and .ppt are not read), a damaged file or one its reader cannot handle, or an archive over the unpacking limits.
encrypted	Password-protected, or an EPUB with DRM.
too_large	Larger than 50 MB, or too large or complex to convert within 120 seconds or within the converter's memory.
robots_disallowed	robots.txt forbids fetching the URL.
blocked_host	The host is on the deny-list, or resolves to a private or local address.
timeout	The document did not finish within the run's time budget.
internal_error	Unexpected failure; message carries the class.
charge_limit_reached	The run reached its maximum total charge before this URL; it was not processed.
invalid_input	The run input passed the input schema but cannot be used (for example no URL at all); the message names the field. One record per input given, or one with url: null when none was given; nothing is processed.

The run fails only when every input failed with an Actor-side code (internal_error, timeout), so a broken Actor is visible. Every other run succeeds — including one where nothing converted because of the caller's own inputs (bad URL, robots, deny-list, an unsupported or encrypted file, invalid input, the charge limit), or where those and Actor-side failures are mixed.

Sources and terms

The source is the caller's file. Process only documents you have the right to process. Fetching follows the Hermes fetch policy: robots.txt is obeyed for the user-agent HermesBot (and *), and a disallowed URL is refused without a request; platforms whose terms forbid automated access, and people-search sites, are on a deny-list and refused without a request; private and local addresses are refused, after DNS and on every redirect; the user-agent is HermesBot/1.0 (+https://apify.com/hermeskd); no login, cookies, credentials, CAPTCHA solving, evasion or proxies are used. Document properties — authors, last editor, comments (their text, author and date) and tracked-change authors — are never returned.

Not affiliated

This Actor is not affiliated with Microsoft, the Document Foundation, OASIS or the IDPF. mammoth, python-pptx, openpyxl, odfpy, markdownify, Beautiful Soup, lxml, striprtf and defusedxml are used under their open-source licences.

Related Hermes Actors

• pdf-to-markdown — PDF documents to Markdown, with OCR for scanned pages.
• image-to-text — OCR for images.
• audio-to-transcript — speech to text.
• video-frames — frames and metadata from a video.
• image-upscale — image upscaling.
• url-to-markdown — a web page to Markdown.
• audio-video-convert — audio and video format conversion.

Changelog

• 1.0 (2026-10-16) — comments in OpenDocument files are no longer read; a workbook with one far cell converts at once; a slide shape of unknown type is read as text; a file a reader cannot handle is unsupported_format, and one that cannot finish in 120 seconds is too_large.
• 1.0 (2026-10-16) — first release: nine formats, sheets as tables, slides as sections with notes, refusal of encrypted, damaged and oversized archives.