PDF annotations are interactive objects stored in a page's Annots array, and Remove PDF Annotations deletes that array while leaving painted page content in place. When a PDF page carries comments, hyperlinks, popups, markup, stamps, file attachments, or form widgets, those objects live in a separate key in the page dictionary, not in the visual content stream that draws text, images, and shapes. The tool targets exactly that key. It inspects each page's Annots entry, reports how many annotation objects exist and how many pages are affected, and only writes a new copy when at least one entry is found. Anything that was already flattened or painted into a page stays visible afterward, because it is no longer controlled by the annotation layer. A highlight that was merged into page artwork before you opened the file remains as colored pixels. A completed form value that was painted onto the page stays on the page even after its interactive widget is removed. Understanding this single distinction, annotation versus painted page content, decides whether the result matches your expectation, and it is the main thing worth absorbing before you start.
What counts as a PDF annotation
The PDF specification describes annotations as objects associated with a page, distinct from the page's content stream. Common subtypes you will see in a typical document include link annotations (the rectangles around clickable addresses and cross-references), text comments and popups (sticky notes and reviewer comments), markup annotations such as highlight, underline, strikeout, and squiggly underline, stamp annotations, file-attachment annotations, and widget annotations that render form fields and accept input. All of these share the same storage pattern: they are children of a page dictionary under the Annots key, with each entry pointing to an annotation dictionary that describes appearance, position, and behavior.
Because annotations live in a per-page list, the total count is a per-page roll-up. That structure is why the tool reports both an annotation count and a page count, and why a PDF can carry many annotations on a single page or just one stretched across every page. The PDF Association's annotations resource documents this organization in detail if you want the formal reference.
Why a local browser tool fits a first attempt
If this is your first time removing annotations, the safer place to start is a tool that does not move your file off your device. Remove PDF Annotations runs entirely in the browser, so file loading, structural inspection, deletion of the Annots key, and serialization of the new PDF all happen locally. The document is never uploaded, which keeps the workflow private and lets you iterate without a network round trip.
Two other things make the starting experience predictable. The tool accepts a non-empty PDF up to 25 MiB, and it counts annotation objects before it offers a download. If your file does not contain any annotations, the tool tells you so and does not create a meaningless replacement. Empty selections, non-PDF files, files over 25 MiB, encrypted PDFs, and damaged cross-reference data all produce a visible error rather than a partial or silent result.
The output is a new file: the original is preserved untouched. The download filename adds -annotations-removed to the original base name, so you can compare the two side by side in your file manager. Selecting another file clears the previous inspection and download state, and any stale asynchronous job cannot replace a newer selection, so the workflow does not get confused by out-of-order results.
How to remove PDF annotations step by step
Once you have a file you are ready to work on, the actual workflow is short and verifiable at every step:
- Choose a non-empty PDF no larger than 25 MiB. Open Remove PDF Annotations in your browser and pick your file from the file picker. If your selection is empty, encrypted, or larger than 25 MiB, you will see a visible error and no download will be produced.
- Review the detected annotation count and number of affected pages. The tool inspects each page's Annots entry and reports how many annotation objects exist in total and on how many pages at least one entry was found. Use this to confirm the document has the kind of interactive objects you expected, before you commit to changing the file.
- Remove the page annotations, download the new PDF, and inspect its visible and interactive behavior. Click the remove action, then save the new file. The original is untouched. Open the downloaded file in the viewer or workflow that matters to you and check that links no longer respond, comments are gone, and form widgets are no longer editable, while any expected painted content is still on the page.
What stays on the page after removal
The removal rule is auditable and easy to summarize: the tool deletes the Annots key from every page dictionary, then serializes a new copy. Everything else about the page stays where it was. That means the page count, page order, page boxes, rotation, and existing content streams remain in place. Text, images, and vector shapes that were already drawn as normal page content are not erased merely because they look like a comment, underline, highlight, signature, or form value.
| Layer | Example | After removal |
|---|---|---|
| Annots array | Link, comment, popup, markup, stamp, file attachment, widget | Deleted from every page dictionary |
| Painted page content | Flattened highlight, completed form value, visible watermark, signature appearance | Unchanged |
| Document catalog and other structures | AcroForm definition, metadata, embedded files, scripts | Unchanged |
| Page structure | Page count, page order, page boxes, rotation, content streams | Unchanged |
This matters most for two cases you are likely to meet early on. The first is a flattened highlight: someone exported the document with marks already merged into page artwork, so the yellow band is now part of the page content, not an annotation. The annotation array may be empty even though the document still looks marked up. The second is a completed form value: the value a user typed has already been painted onto the page during flatten or save, so the visible number or name persists after you remove the widget that used to receive input.
Conversely, removing a link annotation can leave its blue underlined text visible while removing the clickable region. The text was always painted content; only the interactivity lived in the annotation array.
What this tool will not do
The scope is narrow and worth stating plainly so you do not lean on the wrong tool for the wrong job. The tool is not a redaction or sanitization product: deleting page references does not prove that referenced objects, prior revisions, metadata, attachments, scripts, hidden layers, embedded files, or recoverable source information have been purged from every byte. It does not flatten form values, edit metadata, strip embedded images, remove painted watermarks, or clean malware. It does not bypass password protection, repair damaged PDFs, or claim archival conformance. Different PDF viewers can render interactive structures differently, so the downloaded copy should be reviewed in the viewer or workflow that matters.
| What you want to do | Right starting tool |
|---|---|
| Strip links and comments from a normal PDF | Remove PDF Annotations |
| Preserve current field values as page content, then clear widgets | Flatten PDF, then Remove PDF Annotations |
| Permanently purge confidential text and prove it is gone | A purpose-built redaction workflow |
| Change page order, remove pages, or split a document | Rearrange PDF Pages, Delete PDF Pages, or Split PDF |
If your document is an AcroForm whose field definitions should remain intact, do not use it as a form structure tool: the document catalog can still contain form-related structures after annotation removal, and the tool makes no claim that it repairs, flattens, or fully deletes an AcroForm definition. Use a dedicated form editor when field structure must remain valid, or run Flatten PDF first if the goal is to preserve current field appearances as page content.
How to verify the downloaded PDF
After the download, open the new file in the viewer you actually use and check the behavior rather than only the appearance. For link annotations, confirm that the clickable area no longer responds. For comments, confirm that the comments pane is empty and that popups no longer open. For form widgets, confirm that fields are no longer editable. Then compare against the original to confirm the visual page content, including text, images, and any flattened marks, is preserved.
If something in the result does not match your expectation, the most common reason is that what you wanted to remove was actually painted page content rather than an annotation object. In that case, the marks were flattened into the document before you opened it, and removing annotation arrays cannot reach them. A flatten-then-remove workflow or a dedicated form editor is the right path for those cases. For a deeper checklist on what to look for after the download, the result-checking walkthrough walks through each annotation type one by one.
For the underlying load and save mechanics this tool relies on, the pdf-lib PDFDocument API documents how PDF files are loaded into memory and re-serialized without altering unrelated objects, which is the structural reason a narrowed Annots deletion leaves the rest of the document intact.
Related reading: How Do I Make Sure I Remove PDF Annotations Correctly.