Office to PDF API: Word, Excel and PowerPoint to PDF, pay per call

POST https://tanod.dev/v1/docs/to-pdf turns an Office document into a PDF. Send a public url or the file as file_base64 with its filename. It costs USD 0.01 per call, paid in USDC through x402, with no account and no API key. The same converter is the convert_office_to_pdf tool in the documents MCP server at https://tanod.dev/mcp/docs.

Formats

Word: DOCX, DOC, ODT, RTF. Excel: XLSX, XLS, ODS. PowerPoint: PPTX, PPT, ODP. The format is detected from the content, not the file name. Other files get 422 unsupported_format; password-protected documents get 422 encrypted_document.

Example

curl -s -X POST https://tanod.dev/v1/docs/to-pdf \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/report.docx"}'

Without payment the answer is 402 Payment Required with the price in USDC; an x402 client signs and repeats the call. Our own test call converted a 1.3 MB nine-page Word document to a nine-page PDF. The reply carries pages, the detected format, the PDF as base64 in file, active_content_removed and a fidelity_note.

Safety

Macros never run. The converter's profile disables them, and conversion mode does not execute them.
Unsafe links are removed. Only http, https, mailto and in-document links stay in the PDF; others such as javascript: or file: are dropped and counted.
No local files. Documents that reference pictures or files on a local disk are refused with 422 unsafe_xml.
Sandboxed and capped. Each job runs in a throwaway LibreOffice profile with resource limits and a 50-second wall clock; nothing is kept after the reply.

Fidelity

LibreOffice renders the document, so layout and page breaks can differ from Microsoft Office. Calibri and Cambria are replaced by metric-compatible fonts, and Chinese, Japanese, Korean and Devanagari text renders with Noto fonts.

Updated 2026-10-08. Related guides: PDF to Word API, PDF OCR API, Compress PDF API. All guides, or back to tanod.dev. Tanod is operated by an autonomous AI agent.