Draw, type or upload a signature and place it on the page a reader sees, at a width you can state in inches, on one page or every page. Transparent, so it does not hide the line you are signing on.
Initializing in your browser…
A scanned agreement needs your signature on the last page, and the scan carries a CropBox so the sheet the file holds is bigger than the page you see.
Input
agreement.pdf (CropBox 468 x 648 inside a 612 x 792 sheet) · drawn signature · 30% wide · the last page · dated line
Output
The mark centred where it was clicked to within 2 points, 140.4 pt wide, its own 2.5:1 shape kept, with the date printed under it
What a reader sees is the CropBox clipped to the MediaBox and turned by /Rotate, and none of that is what getSize() reports: placing against the sheet instead put the same mark 21.80 points left and 14.70 points up, and changed its shape from 2.494 to 2.656 because the preview and the page were two different rectangles. The signature is drawn on transparency rather than on white, so the signature line under it stays visible, and it is trimmed to its own ink so the width asked for is the width of the signature. After writing, the page is rendered back out of the file by a different implementation and shown, and poppler's pdfsig confirms what this is not: "does not contain any signatures", because a drawn mark is a picture on a page and not a certificate over the bytes.
Draw your signature, type it in a handwriting face, or upload a picture of it, then put it where you want on the page a reader actually sees. Choose one page, every page, the last page or a range, and the tool shows you the signed page read back out of the file it just wrote.
Sign a contract sent as a PDF and return it, without printing, signing and scanning.
Some agreements want a mark on every page. Choose Every page and it goes on all of them at the same place and size.
Sign a form and print the date under the signature in the same operation.
Place a signature or an approval stamp on the last page, which is usually where it belongs.
A scan often carries a CropBox or a rotation, so the sheet the file holds and the page you see are different. The mark goes where you put it either way.
Upload the lease, draw your signature, click the signature line on the last page, choose The last page, and send the signed file back.
Draw your initials, set the width to 8 percent, put them in a corner, choose Every page, and every page comes back with them in the same place.
This draws your signature onto the page as a picture. It is the same thing as signing a printed page with a pen and it carries the same weight: it shows intent, and it proves nothing about who drew it or whether the document changed afterwards. A cryptographic signature is a different object, a signature field whose value is a signature dictionary covering a byte range of the file (ISO 32000-1 clause 12.7.5.5) backed by a certificate, and nothing here writes one. Poppler's pdfsig reads those, and it reports "does not contain any signatures" for a file signed this way, which is the correct answer.
**Where the mark goes.** A PDF page is not the rectangle a library reports for it. `getSize()` gives the MediaBox's width and height with its POSITION thrown away, and what a reader sees is the CropBox clipped to the MediaBox (clause 14.11.2), turned by the page's own /Rotate entry. All three matter and all three are handled here: you point at the page as it is rendered, and the placement is resolved against that same box. On a page whose MediaBox begins at (20, 30) rather than the origin, a placement that ignores the box lands twenty points left and thirty points low; on a page cropped to 468 by 648 points inside a Letter sheet, it lands twenty-two points left and its shape changes as well, because the preview and the page are then two different rectangles.
**How big it is.** The width is a percentage of the page the reader sees, shown in points and in inches beside the slider, and the signature keeps its OWN aspect ratio: a square picture is placed square. A size given as a pixel count only means something relative to whatever preview canvas happened to be on screen, which is not a property of the document.
**What it is drawn on.** A drawn or typed signature is drawn on TRANSPARENCY, not on white. A signature made on a white canvas is an opaque rectangle with a squiggle in it, and it hides whatever it is placed over: a signature line, a printed name, a paragraph. The canvas here shows a checkerboard behind it so you can see that the ground really is transparent, and the mark is trimmed to its own ink before it is placed, so the width you ask for is the width of the signature and not of the canvas it was drawn on.
**And it reads its own output back.** After the file is written it is re-opened with a different implementation and the signed page is rendered and put on screen, so "the signature is on the page, there, at that size" is something visible rather than something asserted. An optional line under the signature, a date by default, is printed as real text in the font it is measured in, centred under the mark and turned with it.
It is a drawn signature, the same as signing a printed page with a pen, and its standing depends on your jurisdiction and on what the parties agreed. It is not a cryptographic signature with a certificate, and the page says so rather than leaving it to be assumed.
No. A drawn or typed signature is made on a transparent ground, so the signature line, the printed name and any text under it stay visible. An uploaded JPEG cannot be transparent, because the format has no alpha channel; upload a PNG for that.
The signature becomes part of the page content, so it cannot be dragged off in a viewer. It is not tamper proof: anything a picture on a page can be, this is.
It should be. If the page carries a CropBox or a rotation, what you see and what the file stores are different rectangles, and the placement is resolved against the one you see. The panel says which of those apply to the page you are on.
The width slider is a percentage of the page, and the points and inches are shown next to it. The panel also reports the resolution the signature will print at, so a small picture blown up is flagged before it goes on the page.
Yes. The drawing canvas takes touch input, so you can sign with a finger or a stylus.
PDF parsing and editing happen in your browser. Documents, and everything inside them, are never uploaded or stored remotely.