DocsBolt

PDF to Markdown Converter

Turn a PDF into Markdown that keeps its structure — headings become #, bullets become -, code stays in fenced blocks — for notes apps, documentation repos and AI pipelines that want clean text with structure.

.pdf · up to 25 MB per file

  • PDF
  • up to 25 MB per file
  • Deleted after one hour

What PDF to Markdown does

Markdown is what documentation repos, note-taking apps and language models all read most easily, and a PDF is what most documents arrive as. Converting one to the other well means recovering structure the PDF only hints at: which lines are headings, which are bullets, which blocks are code. This tool infers that from font sizes and styles across the whole document — the body size sets the baseline, larger runs become headings by rank, bulleted lines become list items, monospace runs become code — and writes clean Markdown with bold and italic kept inline. The output opens in Obsidian, Notion, GitHub or a text editor, and pastes into a prompt without the noise a raw text dump carries.

Worked example

in · guide.pdf (page 3)
Installation            ← 18 pt bold
Run the installer, then:
• restart the terminal   ← bullet glyph
• run docsbolt --version
npm install docsbolt     ← Courier
out · guide.md
## Installation

Run the installer, then:

- restart the terminal
- run docsbolt --version

```
npm install docsbolt
```

Heading levels come from font size relative to the body text; bullets from the glyph; code from a monospace font. Page breaks are not marked unless you ask.

Formats and limits

Accepts
.pdf — a password-protected PDF needs its password, or Unlock PDF first
Produces
Markdown (.md)
Files per run
One
Size limit
25 MB per file

How it works

  1. 01

    Upload the PDF

    A document with a text layer.

  2. 02

    Convert

    Font sizes and styles are read to recover the structure.

  3. 03

    Download the .md

    Drop it into Obsidian, Notion, GitHub or a prompt.

Diagram: the text and picture blocks are lifted out of a page and delivered as a separate text file and image files.text.txtimg1.jpg

Limitations

  • Headings are inferred from font size: a document that sets headings by bold weight alone comes out as paragraphs.
  • Tables are recovered only when the PDF draws them with ruling lines; whitespace-aligned tables become plain lines.
  • Images are referenced by placeholder, not embedded — Markdown has no way to carry image bytes.

What happens to your file

Processed over an encrypted connection and deleted after one hour.

Full detail in the Privacy Policy.

Questions

How are headings detected?

By comparing each line's font size to the body text of the document. Larger runs become headings, with the biggest as #, then ##, and so on.

Are tables converted?

Tables come through as their text, row by row, not as Markdown pipe tables. For spreadsheet-ready tables use PDF to Excel.

Why do I want page markers?

If you will cite the source — 'see page 12' — turn them on; each page starts with an HTML comment naming it. For a clean reading copy, leave them off.

How is this different from Extract Text?

Extract Text gives you the words with line breaks and nothing else. PDF to Markdown gives you the words plus the document's structure — headings, lists, emphasis, code — expressed in Markdown syntax.

Will links be kept?

Visible URLs come through as text. Hyperlinks that show different text from their target are not currently converted into Markdown link syntax.

What happens with a two-column layout?

Columns are read in the order the PDF stores them, which is almost always left column then right. Where a file interleaves them, the Markdown will too; PDF to Word handles those layouts better.

Tools people use next

More PDF tools

Convert, compress, merge, split, edit, secure and extract from PDFs.

Browse PDF tools