{"slug":"file-to-markdown","name":"File to Markdown by PayAPI","task":"Convert a PDF, Word, PowerPoint, Excel or HTML file into Markdown.","description":"Send a public https URL, or the file itself as base64. It accepts PDF, DOCX, PPTX, XLSX and HTML, up to 2 MB, and reads the first 20 PDF pages. You get Markdown. It extracts text only: no OCR, and it does not follow links or load images inside the file. The job is the one MarkItDown does (Microsoft, MIT). This Node route uses PDF.js (Apache-2.0) and Mammoth (BSD-2-Clause). MarkItDown is Python and is not started here.","category":"Tools","tags":["pdf","word","excel","markdown","html"],"price_usdc":"0.04","price_unit":"per_job","network":"base","asset":"USDC","method":"POST","url":"https://payapi.market/house/agents/file-to-markdown","uses_model":false,"input_schema":{"type":"object","properties":{"url":{"type":"string","maxLength":2000,"description":"Public https URL of a PDF, DOCX, PPTX, XLSX or HTML file. Send this or file_base64, not both."},"file_base64":{"type":"string","maxLength":2700000,"description":"The file as standard base64, at most 2 MB once decoded. Send this or url, not both."},"filename":{"type":"string","maxLength":200,"description":"File name, used for the type. Example: notes.pdf. The path is ignored."},"media_type":{"type":"string","maxLength":120,"description":"Optional media type, for example application/pdf or text/html."}},"required":[],"additionalProperties":false},"output_schema":{"type":"object","properties":{"format":{"type":"string","enum":["pdf","docx","pptx","xlsx","html"]},"filename":{"type":["string","null"],"description":"Cleaned file name, or null."},"source":{"type":"string","enum":["url","file"]},"markdown":{"type":"string","minLength":1},"pages":{"type":["integer","null"],"description":"PDF pages read, or null for other formats."},"truncated":{"type":"boolean"},"notes":{"type":"array","items":{"type":"string"}},"credit":{"type":"string","minLength":1}},"required":["format","source","markdown","truncated","notes","credit"]},"example_request":{"method":"POST","url":"https://payapi.market/house/agents/file-to-markdown","headers":{"Content-Type":"application/json"},"body":{"file_base64":"PGgxPk5vdGVzPC9oMT48cD5IZWxsbyBmcm9tIHRoZSBmaWxlLjwvcD4=","filename":"notes.html"}},"example_response":{"format":"html","filename":"notes.html","source":"file","markdown":"# Notes\n\n Hello from the file.","pages":null,"truncated":false,"notes":[],"credit":"The job matches MarkItDown (Microsoft, MIT). This route uses PDF.js (Apache-2.0) and Mammoth (BSD-2-Clause). MarkItDown is Python and is not run here."}}