Daniel Kossmann

Tools to Convert a Web Page into Markdown for AI (LLM)

in

Here is a list of tools that I tested to convert web pages to the Markdown format, mainly for using their content with AI (LLMs). To compare the results, I decided to convert a post I wrote that has subheadings, code, links, keyboard input, and an image.

Proprietary

pure.md

How to use it: Add the desired webpage URL after https://pure.md/

Example: https://pure.md/https://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/

---
url: https://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/
title: Post to test WordPress to Markdown conversion - Daniel Kossmann
description: This is the excerpt content.
access_date: 2025-05-25T14:30:18.000Z
current_date: 2025-05-25T14:30:18.458Z
---

# Post to test WordPress to Markdown conversion

May 25, 2025

—

by

Daniel Kossmann

in 📝 Others

This post is to test how tools handle converting website pages to Markdown, so I’m using a mix of WordPress blocks to check the results.

## h2: Styles

Here is **a text in bold**, _another in italic_, and one with ~~strikethrough~~.

Some more advanced styles are this highlight, 

I also like to display keyboard input like Ctrl+C.

### h3: code

I sometimes write some `inline code with 'special' or {} and [] and /`.

#### h4: Default code block

`// this is the default code block
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =`Code language: JavaScript (javascript)

#### h4: Block: Code Pro

```
// this is Code Pro for JSON
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =
```

JSON

Below is a separator

---

## h2: Images

Below is an image

![This is an alternative text](IMAGE) 

This is an image caption

But I can also have multiple images as a gallery

## Lists

### Numbered

1. Item 1  
   * Item 1.1  
   * Item 1.2  
         * Item 1.2.1 with **bold**  
                  * item 1.2.1.1 with _italic_  
         * Item 1.2.2 with a link inside
2. Item 2  
Second line of item 2
3. item 3

### Bullets

* Item 1  
   * Item 1.1  
   * Item 1.2  
         * Item 1.2.1 with **bold**  
                  * item 1.2.1.1 with _italic_  
         * Item 1.2.2 with a link inside
* Item 2  
Second line of item 2
* item 3

## Columns

Column 1

Column 2

Column 3

## Non-core blocks

WIP

---

Daniel Kossmann

I am passionate about continuous learning and sharing knowledge, which is why I advocate for open source and Creative Commons. I believe that sharing knowledge can drive innovation and improve the world.

---

## Comments

### Leave a Reply 

Your email address will not be published. Required fields are marked \*
Markdown
Expand

Here are a few notes about my experience using it:

  1. I love how the meta-info sits in the header with those --- lines. It just makes everything feel way more organized;
  2. It also captures non-article content like author bio and comments;
  3. I couldn’t figure out how to clear their cache if the content changes;
  4. Some things didn’t work: link, highlight, image URLs and gallery, keyboard input, core WP code block, and language descriptor from Code Pro block.

Jina AI Reader

How to use it: Add the desired webpage URL after https://r.jina.ai/

Example: https://r.jina.ai/https://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/

Title: Post to test WordPress to Markdown conversion

URL Source: http://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/

Published Time: 2025-05-25T14:23:56+00:00

Markdown Content:
This post is to [test how tools handle converting website pages to Markdown](https://www.danielkossmann.com/tools-to-convert-web-page-into-markdown/), so I’m using a mix of WordPress blocks to check the results.

h2: Styles
----------

Here is **a text in bold**, _another in italic_, and one with ~~strikethrough~~.

Some more advanced styles are this highlight,

I also like to display keyboard input like Ctrl+C.

### h3: code

I sometimes write some `inline code with 'special' or {} and [] and /`.

#### h4: Default code block

```
// this is the default code block
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =
```
Code language:JavaScript(javascript)
#### h4: Block: Code Pro

```
// this is Code Pro for JSON
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =
```

JSON

Below is a separator

* * *

h2: Images
----------

Below is an image

![Image 1: This is an alternative text](https://www.danielkossmann.com/wp-content/uploads/2025/05/improve-english-v1-1024x488.png)

This is an image caption

But I can also have multiple images as a gallery

![Image 2](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb1.png)![Image 3](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb2.png)![Image 4](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb3.png)

Lists
-----

### Numbered

1.   Item 1 
    *   Item 1.1
    *   Item 1.2 
        *   Item 1.2.1 with **bold**
            *   item 1.2.1.1 with _italic_

        *   Item 1.2.2 with [a link](https://www.danielkossmann.com/about/) inside

2.   Item 2

Second line of item 2
3.   item 3

### Bullets

*   Item 1 
    *   Item 1.1
    *   Item 1.2 
        *   Item 1.2.1 with **bold**
            *   item 1.2.1.1 with _italic_

        *   Item 1.2.2 with [a link](https://www.danielkossmann.com/about/) inside

*   Item 2

Second line of item 2
*   item 3

Columns
-------

Column 1

Column 2

Column 3

Non-core blocks
---------------

WIP
Markdown
Expand

Here are a few notes about my experience using it:

  1. Some title use # and other ---. I prefer to only use one of these styles.
  2. Numbered list was just numbered for the first level and there is an extra line break between some items;
  3. Some things didn’t work: highlight, keyboard input and language descriptor from core code block and Code Pro block.
  4. I couldn’t figure out how to clear their cache if the content changes

Open Source

MarkItDown

How to use it: install the library and pass the URL after the markitdown command.

Example: markitdown https://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/

[![Daniel Kossmann](https://www.danielkossmann.com/wp-content/uploads/2018/12/2017-daniel-kossmann-ferraz-blog-logo.png)](https://www.danielkossmann.com/)

[Daniel Kossmann](https://www.danielkossmann.com)

* [꩜ About Me](https://www.danielkossmann.com/about/)
* [📝 Blog](https://www.danielkossmann.com/archive/)
* [🛠️ Projects](https://www.danielkossmann.com/projects/)
* [Português](https://www.danielkossmann.com/pt/)

# Post to test WordPress to Markdown conversion

[May 25, 2025](https://www.danielkossmann.com/post-to-test-wordpress-to-markdown-conversion/)

—

by

[Daniel Kossmann](https://www.danielkossmann.com/author/webmaster-site/)

in [📝 Others](https://www.danielkossmann.com/category/others/)

This post is to [test how tools handle converting website pages to Markdown](https://www.danielkossmann.com/tools-to-convert-web-page-into-markdown/), so I’m using a mix of WordPress blocks to check the results.

## h2: Styles

Here is **a text in bold**, *another in italic*, and one with ~~strikethrough~~.

Some more advanced styles are this highlight,

I also like to display keyboard input like `Ctrl`+`C`.

### h3: code

I sometimes write some `inline code with 'special' or {} and [] and /`.

#### h4: Default code block

```
// this is the default code block
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =Code language: JavaScript (javascript)
```

#### h4: Block: Code Pro

// this is Code Pro for JSON
‘single quote’ or { } and [ ] and “double quotes” with / and \
– and + and =

```
// this is Code Pro for JSON
'single quote' or { } and [ ] and "double quotes" with / and \
- and + and =
```

JSON

Below is a separator

---

## h2: Images

Below is an image

![This is an alternative text](https://www.danielkossmann.com/wp-content/uploads/2025/05/improve-english-v1-1024x488.png)

This is an image caption

But I can also have multiple images as a gallery

![](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb1.png)
![](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb2.png)
![](https://www.danielkossmann.com/wp-content/uploads/2024/10/thumb3.png)

## Lists

### Numbered

1. Item 1
   * Item 1.1
   * Item 1.2
     + Item 1.2.1 with **bold**
       - item 1.2.1.1 with *italic*
     + Item 1.2.2 with [a link](https://www.danielkossmann.com/about/) inside
2. Item 2
   Second line of item 2
3. item 3

### Bullets

* Item 1
  + Item 1.1
  + Item 1.2
    - Item 1.2.1 with **bold**
      * item 1.2.1.1 with *italic*
    - Item 1.2.2 with [a link](https://www.danielkossmann.com/about/) inside
* Item 2
  Second line of item 2
* item 3

## Columns

Column 1

Column 2

Column 3

## Non-core blocks

WIP

---

![](https://secure.gravatar.com/avatar/1137b1d7d0de77e667a5e6a1b5102e9dd341bc71178e92d0147f86db003f7628?s=48&d=monsterid&r=g)

Daniel Kossmann

I am passionate about continuous learning and sharing knowledge, which is why I advocate for open source and Creative Commons. I believe that sharing knowledge can drive innovation and improve the world.

---

## Comments

### Leave a Reply [Cancel reply](/post-to-test-wordpress-to-markdown-conversion/#respond)

Your email address will not be published. Required fields are marked \*

Comment \*

Name \*

Email \*

Website

Δ

←[Previous:  How to use ChatGPT Desktop with Visual Studio Code (VS Code) in MacOS](https://www.danielkossmann.com/how-to-use-chatgpt-desktop-with-visual-studio-code-vs-code-macos/)

Search

The articles on this site are licensed under the [Creative Commons Attribution-ShareAlike 4.0 International License](https://creativecommons.org/licenses/by-sa/4.0/deed.en).

Proudly powered by [WordPress](https://wordpress.org/) and hosted on [Yogh Cloud Host](https://www.yogh.com.br/yogh-cloud-host-afiliados/ref/19/).

##

##

Loading Comments...

Write a Comment...

Email (Required)

Name (Required)

Website

###
Markdown
Expand

Here are a few notes about my experience using it:

  1. It also captures non-article content like header and footer navigation and content, author bio and comments;
  2. Some things didn’t work: highlight, Code Pro block (content showed twice) and there were no language descriptor form the two code blocks;
  3. Numbered list was just numbered for the first level.

Next steps

As I try out new tools or update my test post, I’ll keep this content up to date—maybe even add a comparison table. My main goal is to document what I find and how these tools work.

If you know any other tools for converting web pages to Markdown, or have tips on using the ones listed, please leave a comment.

Daniel Kossmann

I am passionate about continuous learning and sharing knowledge, which is why I advocate for open source and Creative Commons. I believe that sharing knowledge can drive innovation and improve the world.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *