# Dynare reference manual in markdown

**URL:** <https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367>\
**Category:** Dynare User Guide\
**Created:** [16 May 2026 21:04 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367 "2026-05-16T21:04:20Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![dh.murakami](https://forum.dynare.org/letter_avatar_proxy/v4/letter/d/58956e/32.png) [@dh.murakami](https://forum.dynare.org/u/dh.murakami)\
**Post date:** [16 May 2026 21:04 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367/1 "2026-05-16T21:04:20Z")

</div>

Hi Dynare team

Would it be possible to have the Dynare reference manual to be published online in markdown? It would allow LLMs to parse through support documentation a lot easier and more accurately.

Thanks!

---

<div class="post-metadata">

**Author:** ![fhamann](https://forum.dynare.org/letter_avatar_proxy/v4/letter/f/46a35a/32.png) [@fhamann](https://forum.dynare.org/u/fhamann)\
**Post date:** [17 May 2026 22:29 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367/2 "2026-05-17T22:29:50Z")

</div>

Check this out:

> **[GitHub - microsoft/markitdown: Python tool for converting files and office...](https://github.com/microsoft/markitdown)**
>
> Python tool for converting files and office documents to Markdown.

It’s free.

---

<div class="post-metadata">

**Author:** ![dh.murakami](https://forum.dynare.org/letter_avatar_proxy/v4/letter/d/58956e/32.png) [@dh.murakami](https://forum.dynare.org/u/dh.murakami)\
**Post date:** [18 May 2026 00:57 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367/3 "2026-05-18T00:57:35Z")

</div>

Thanks – I already used something similar to convert the PDF to a markdown file.

However, these methods often contain scan errors and mistakes that occur during the conversion. Moreover, you have to rescan and fix errors each time the manual is updated.

---

<div class="post-metadata">

**Author:** ![jpfeifer](https://forum.dynare.org/user_avatar/forum.dynare.org/jpfeifer/32/5044_2.png) [@jpfeifer](https://forum.dynare.org/u/jpfeifer)\
**Post date:** [18 May 2026 08:20 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367/4 "2026-05-18T08:20:01Z")

</div>

The manual PDF is built using the markdown files from [Making sure you're not a bot!](https://git.dynare.org/Dynare/dynare/-/tree/master/doc/manual/source?ref_type=heads)

---

<div class="post-metadata">

**Author:** ![sebastien](https://forum.dynare.org/user_avatar/forum.dynare.org/sebastien/32/19973_2.png) [@sebastien](https://forum.dynare.org/u/sebastien)\
**Post date:** [19 May 2026 13:00 UTC](https://forum.dynare.org/t/dynare-reference-manual-in-markdown/28367/5 "2026-05-19T13:00:49Z")

</div>

Actually the source of the reference manual is not in Markdown but in ReStructuredText. But that does not make a practical difference for an LLM since both are markup languages.

Moreover, since AI scraper bots are ingesting our GitLab instance on a daily basis, most major models have most likely already parsed the markup source of the manual.