Annotate PDF in C# — highlight, comment, reply, extract

The shortest honest answer to "how do I annotate a PDF in C#?" — a console application that highlights text, draws a box, attaches a comment with a reply thread, reads the markup back out and renders page images. It runs on Windows and Linux, with no Adobe Acrobat and no Microsoft Office installed, and the same calls work on DOCX, XLSX, PPTX and images as well as PDF.

dotnet run --project src/AnnotatePdf -- highlight samples/quarterly-review.pdf --text "Hosting costs"

Read this with

Resource	What it adds
Blog: How to annotate a PDF in C# — the 2026 edition	the walk-through, with what changed since the 2021 posts
Docs: Annotate a PDF in C#	the step-by-step reference page
This repository	the code that runs

Quick start

Requires the .NET SDK 8.0 or later.

git clone https://github.com/groupdocs-annotation/annotate-pdf-csharp-net.git
cd annotate-pdf-csharp-net

dotnet run --project src/AnnotatePdf -- info samples/quarterly-review.pdf
dotnet run --project src/AnnotatePdf -- highlight samples/quarterly-review.pdf --text "Hosting costs" --message "Check this against the invoice"
dotnet run --project src/AnnotatePdf -- list output/quarterly-review.annotated.pdf
dotnet run --project src/AnnotatePdf -- preview output/quarterly-review.annotated.pdf

The first command prints what GroupDocs.Annotation sees — file type, page count, page size and how many text lines it found per page. The rest add markup, read it back and render PNGs of the pages.

Commands

Command	What it does
info <document>	file type, pages, page size, text lines per page
highlight <document> --text "phrase"	finds the phrase and highlights every line it appears on
highlight <document> --page 0 --x --y --width --height	highlights an explicit rectangle
area <document> --page 0 --x --y --width --height	draws a box around a region
list <document>	prints annotations, authors and replies
remove <document> [--id N]	removes one annotation, or all of them
preview <document> [--no-annotations]	renders page images as PNG

Every command takes --out to choose where the result goes; the default is output/.

The five calls that matter

Everything the sample does lives in Annotations.cs.

Add a highlight with a reply.

using (Annotator annotator = new Annotator("input.pdf"))
{
    annotator.Add(new HighlightAnnotation
    {
        PageNumber = 0,                       // 0 = first page
        BackgroundColor = 0xFFE699,           // 0xRRGGBB
        Opacity = 0.5,
        Message = "Check this figure",
        User = new User { Id = 1, Name = "Reviewer", Role = Role.Editor },
        Points = new List<Point>              // top-left, top-right, bottom-left, bottom-right
        {
            new Point(72, 664), new Point(312, 664),
            new Point(72, 650), new Point(312, 650)
        },
        Replies = new List<Reply>
        {
            new Reply { Comment = "Confirmed with finance.", RepliedOn = DateTime.Now }
        }
    });

    annotator.Save("result.pdf");
}

Read the markup back.

using (Annotator annotator = new Annotator("result.pdf"))
    foreach (AnnotationBase a in annotator.Get())
        Console.WriteLine($"#{a.Id} {a.Type} page {a.PageNumber + 1}: {a.Message}");

Find text instead of measuring coordinates. GetDocumentInfo() hands back the text lines of every page with their boxes, so the phrase can be located instead of guessed:

IDocumentInfo info = annotator.Document.GetDocumentInfo();
foreach (PageInfo page in info.PagesInfo)
    foreach (TextLineInfo line in page.TextLines)
        if (line.Text.Contains(phrase))
            ...   // see HighlightText() for the conversion

Remove with annotator.Remove(id) and render with annotator.Document.GeneratePreview(...).

Two conventions that catch everybody

	Origin	Page numbering
PageInfo / TextLineInfo	top-left	PageNumber is 1-based
AreaAnnotation.Box	top-left	—
HighlightAnnotation.Points and other text markup	bottom-left	PageNumber is 0-based
PreviewOptions.PageNumbers	—	1-based

A line at TopIndent = t with height h is therefore pageHeight - t - h … pageHeight - t in a highlight, and the page index is PageInfo.PageNumber - 1. An annotation addressed to a page that does not exist is dropped without an exception — if markup "did not appear", check the page index first.

Linux

docker build -t annotate-pdf .
docker run --rm annotate-pdf info samples/quarterly-review.pdf
docker run --rm -v "$PWD/output:/work/output" annotate-pdf highlight samples/quarterly-review.pdf --text "Hosting costs"

The Dockerfile installs fontconfig and fonts-liberation: the .NET runtime image ships without fonts, and text rendering needs them.

Licensing

Without a licence the library runs in evaluation mode and the output carries evaluation marks. Set GROUPDOCS_ANNOTATION_LICENSE to the path of a .lic file to remove them — temporary licence.

Related samples

• ai-document-review-annotations-net — the same annotations, written from a model's findings
• browser-annotation-ui-net — drawing and answering them in a browser
• GroupDocs.Annotation-for-.NET — the full example set, one file per annotation type

License

MIT, see LICENSE. GroupDocs.Annotation itself is a commercial library with its own licence terms.