Formats
Markdown → DOCX
The default input type, and the best fit for text you already have — especially output from a language model.
Reading this with an AI agent?
The entire documentation lives in one Markdown file, built to be fetched and read by an LLM. Copy the prompt below and paste it into ChatGPT, Claude, Cursor or any agent with web access — it asks the agent to read the file before helping you.
Request
{
"input_type": "markdown",
"content": "# Quarterly Report\n\n## Summary\n\nRevenue grew **18%**.\n\n- North: 42k\n- South: 31k\n\n| Region | Total |\n|--------|-------|\n| North | 42000 |\n| South | 31000 |",
"response_format": "file",
"filename": "quarterly-report"
}input_type defaults to markdown, so you can leave it out entirely.
What carries over
| Markdown | In Word |
|---|---|
# … ###### | Heading 1 to 6, styled by the template |
**bold**, *italic* | Bold and italic runs |
- item / 1. item | Bulleted and numbered lists, nesting included |
| Pipe tables | A real Word table with a header row |
> quote | Block quote style |
`code` and fenced blocks | Monospace runs and code paragraphs |
[text](url) | Clickable hyperlink |
--- | Horizontal rule |
The newline trap
This is the single most common problem. content is a JSON string, so real line breaks must be escaped as \n:
"content": "# Title\n\nFirst paragraph.\n\nSecond paragraph."
"content": "# Title\\n\\nFirst paragraph."
If your document arrives as one long line, or you see \n printed inside the Word file, this is why. In no-code tools, pass the raw text through an expression and let the tool do the escaping.
Free plan watermark
Documents generated on the free plan get a Generated with DocxAPI line appended at the end. Paid plans produce clean documents. Nothing else about the output differs.