Which annotation types are exported?

Highlights, underlines, squiggles, strikeouts, sticky notes (Text), free text boxes, ink strokes, stamps, and links — every annotation subtype in the file. You can filter the export to just the types you want.

Does it include the text that was highlighted?

Yes, when the page has a text layer. The highlight's coordinates are mapped back onto the text, so the export quotes the marked passage. On scanned pages there is no text to recover — run OCR PDF first.

What does the Markdown output look like?

A heading per page (nested under chapter headings when the PDF has bookmarks), with each annotation as a list item: type, the quoted text as a blockquote, the comment, and author/date/color metadata.

What does the JSON output contain?

A structured object: source filename, export timestamp, chapter list, and per-page arrays of annotations with type, note, author, date, color, quote, and the annotation rectangle.

Does it change the original PDF?

No. The tool only reads the annotation layer — your file is never modified.

Can I export only highlights and skip the rest?

Yes. The type filter lists every annotation subtype found in the document; uncheck the ones you don't want and they are left out of the notebook.

Is my PDF uploaded to a server?

No. Everything runs in your browser — the file is parsed locally and the notebook is generated on your device.

Trình xuất chú thích

Trích xuất mọi đánh dấu, ghi chú và chú thích từ PDF thành sổ tay Markdown hoặc JSON — nhóm theo trang và chương.

1. Upload a marked-up PDF

Highlights, comments, underlines, and ink strokes are read straight from the file.

2. Filter and format

Pick which annotation types to include and choose Markdown or JSON.

3. Export the notebook

Annotations land grouped by page — and by chapter when the PDF has bookmarks.

About Annotation Exporter

Your Highlights, Out of the PDF

You marked up forty pages of a paper and now the notes are trapped inside it. Annotation Exporter reads the annotation layer directly — highlights, sticky notes, underlines, squiggles, strikeouts, free text, and ink strokes — and assembles them into a notebook you can actually use.

Each entry keeps its page number, the comment text, the author, the date, and the color. For highlights and other text markup, the tool also recovers the exact words that were marked, so your notes quote the source instead of pointing at it.

Built for Literature Reviews and Study Notes

The output is shaped for the tools researchers and students already use.

Markdown for Obsidian

One .md file with a heading per page, quotes as blockquotes, and comments inline — drops straight into an Obsidian vault or any notes app.

JSON for Pipelines

A structured .json export with type, page, rect, author, date, color, and quote per annotation — ready for scripts, spreadsheets, or a review-roundup dashboard.

Chapter-Aware Grouping

When the PDF has bookmarks, annotations group under their chapter headings instead of one long page list.

Quoted Text Recovery

Markup quadPoints are mapped back onto the text layer, so a highlight exports the sentence it covered — not just a yellow rectangle.

The One Caveat: Scanned Pages

A highlight drawn over a scanned page marks an image, not text — there is nothing to quote because the page has no text layer. The annotation still exports with its position and any attached note, but the quote field comes back empty.

If your PDF is a scan, run it through OCR PDF first to give the pages a text layer, then export. The same applies to markup on flattened or rasterized documents.

How It Compares

Most tools can create annotations or delete them. Fewer can read them back out into something useful.

Tính năng	Adobe Acrobat	Zotero / reference managers
Export highlights + comments	Paid	Library items only
Markdown output		Via plugin
JSON output
Quoted text under highlights
Chapter grouping from bookmarks
Works on any PDF file
No upload required	Desktop only

Read-Only and Fully Local

Exporting never modifies the source PDF — it is a read, not an edit. The file is parsed in your browser and the notebook is generated on your device, so a confidential review copy or an unpublished manuscript stays exactly where it started.

Pair it with Annotate PDF to mark up a document and Annotation Exporter to pull the notes back out — the full loop, without a server in the middle.