How to publish documents to a static site as Markdown
Leave feedback
On this page
Static site generators eat Markdown with YAML front matter and images on disk. That is exactly what convert_to_markdown produces when you ask for it.
Note
The commands and config snippets on this page are for the .NET build of the server — the only platform available today. Installation and client setup: MCP server for .NET. Other platforms will expose the same tools with their own launch command; everything else on this page applies unchanged.
The prompt
Convert these documents to Markdown with front matter, GitHub flavour, and images saved as files.
Three options doing three jobs:
frontMatter: true — YAML title, author, format, page count at the top. Hugo, Jekyll, MkDocs and Docusaurus all read it.
images: "file" — pictures written alongside the .md and referenced by relative path, which is how a site expects them.
flavor: "github" — GitHub-flavoured Markdown, the dialect most generators assume. Switch to commonmark if your toolchain is stricter.
Fitting your generator’s layout
The server writes files; arranging them is the agent’s job:
Put each converted file in a folder named after the document, with its images in an images subfolder next to it.
Most generators also want extra front-matter keys — weight, menu, date. Have the agent add them after conversion:
Add weight to the front matter of each file, numbered in the order I listed them.
Check before you publish
Two checks catch nearly everything:
Length against source. A 60-page document producing 3 KB of Markdown means a scan or the three-page evaluation limit.
Images actually present. Ask for a listing of the image folder; a .md referencing pictures that were never written renders as broken images on the site.
Tables and structure
Word and Excel sources convert with their structure intact — headings become headings, tables become Markdown tables. PDFs vary with how they were produced: a born-digital export converts cleanly; a scan does not convert at all in any useful sense.
Spot-check the messiest source before running a batch of two hundred.
Then the site build
At this point Markdown is Markdown — commit it, run your generator, and the documents are pages. The conversion stayed on your machine, which matters when the archive being published contains material that is public after review but not before.
Was this page helpful?
Any additional feedback you'd like to share with us?
Please tell us how we can improve this page.
Thank you for your feedback!
We value your opinion. Your feedback will help us improve our documentation.