Drop your document into the zone below and click any page to open it full size. Type text anywhere, drop in a logo or a photograph of your signature, draw freehand, highlight a paragraph, cover something up, or turn an area into a working hyperlink. Everything is written as real PDF objects, so the text stays selectable and the file stays a document rather than becoming a stack of pictures. Nothing is uploaded.
Your files are read by JavaScript running on your own machine. There is no upload, no queue, and no copy on our server to delete later.
Drag a card to reorder it, or use the arrows under each card. The order here is the order in the output. Each card is a preview of the finished page, so what you see is what the PDF will contain.
Esc close ← → move R rotate
Starting
This is your processor doing the work, so the time it takes depends on your machine and not on how many other people are on the site.
The tool above is ready to use. The guide below starts with the limitation nobody else states up front: you cannot click a paragraph in a PDF and retype it. It explains exactly why not, what the professional editors are really doing when they appear to, the two step method that gets you the same result here, and which of the six tools to reach for in which situation.
Three minutes, nothing to install, and no account needed for your first run. This free PDF editor executes inside the page you are reading, so all four steps happen on your own machine.
Click the drop zone above or drag the file anywhere onto this page. Every page appears as a tile, and the badge on each one counts what you have added to it as you work.
The page opens at full size with a toolbar beside it. Zoom in as far as you need, because placing a signature accurately on a thumbnail is not possible and this is where the actual work happens.
Text types where you click. Image drops a logo or a signature you can drag and size. Draw is freehand. Highlight marks a passage without hiding it. Cover up paints over something. Link turns a rectangle into a working hyperlink.
The arrows at the bottom walk the document without closing the editor, so a signature on page 1 and initials on page 14 are one session. Press Save changes and the file comes back with everything written into it as real PDF objects.
Undo and redo work throughout, including Ctrl+Z and Ctrl+Shift+Z. Click any object you have added to select it, drag it to move it, pull a corner handle to size it, and press Delete to remove it. Nothing you add is committed to the file until you press Save changes, so experiment freely.
This is the first thing anybody wants from a PDF editor and it is worth understanding why it is hard, because it explains what every tool including the expensive ones is actually doing when it appears to manage it.
A PDF does not contain paragraphs. It contains instructions to draw glyphs at coordinates. A line of text in a PDF is closer to "move to x=72 y=684, select font F1 at 11 points, draw these character codes" than it is to a sentence in a document. There is no paragraph object, no line wrapping rule, no concept of what belongs with what. The format was designed to reproduce a page exactly, and it succeeded by discarding everything about how the page was constructed.
To edit a sentence properly, a tool has to reconstruct all of that from the coordinates:
Acrobat does attempt this, and it is genuinely impressive on a simple document produced by Word. It is also why editing text in Acrobat frequently produces a subtly different font, a line that reflows oddly, or an error saying the font is not available. On a scan it does nothing. There is no text to edit, only a photograph of text.
We could ship a partial version of that reconstruction. We would rather tell you plainly what this tool does not do than ship something that works on the test file and mangles the contract.
If the source document still exists, edit that. Five minutes in Word and a re-export beats any amount of PDF surgery, and the result is a document instead of a repair. This tool is for the very common case where the source is gone, was never yours, or was a piece of paper.
Two steps, and it is what every browser based editor is doing whether or not it says so: cover the old text, then draw the new text on top. Done carefully the result is indistinguishable from an edit, and it works identically on a born-digital PDF and on a scan.
| Job | Works? | Notes |
|---|---|---|
| Correcting a date, a price or a reference number | Yes, easily | Short strings on a plain background are the ideal case. This is most real edits. |
| Fixing a name or a spelling mistake | Yes | Match the typeface and check the baseline. Nobody will see it. |
| Filling in a form that has no form fields | Yes | Text boxes on the lines. Extremely common with scanned or flattened forms. |
| Adding a clause to a paragraph | Poorly | The text after it will not reflow, so you are typesetting by hand. Go back to the source. |
| Editing text over a photograph or a pattern | Poorly | A solid cover box is visible against a busy background. There is no honest way around this. |
| Rewriting a page | No | Recreate the page in a word processor and replace it with Organize PDF instead. |
Covering text does not remove it. The words under a white box are still in the file and still come out of any text extraction, and the box can be deleted by anybody with an editor. For a price correction that is harmless. For a passport number, a home address or a bank account it is a serious disclosure, and the tool you need is Redact PDF, which destroys the content rather than the view of it.
Click and type. The typeface, size, colour and weight are all in the panel, and the box resizes to what you typed instead of wrapping, so what you see on the page is exactly what gets written. Text is drawn as real text using an embedded font reference, which means it stays selectable, searchable and copyable in the finished PDF. It is not a picture of words.
The three typefaces are Helvetica, Times and Courier, which are three of the fourteen that every PDF reader is required to carry. That is why adding text does not measurably increase your file size. Type a character outside Western European Latin, such as a Polish or Greek letter, and a real font file is fetched and a subset embedded automatically, containing only the glyphs you actually used.
Choose a file and it lands on the page ready to be dragged and sized. The obvious use is a logo. The more common one is a signature: photograph your signature on white paper, remove the background in any image editor, save it as a PNG, and drop it on the line. A PNG with a transparent background keeps its transparency all the way into the PDF as a proper soft mask, so the paper shows through instead of a white rectangle sitting on the page.
PNG and JPEG are embedded directly. WebP, AVIF, GIF and BMP are converted once through a canvas and embedded as PNG, which preserves transparency and costs nothing in quality. The opacity control is there for stamps and seals that should not obscure what is under them.
Freehand, with a colour and a width. A signature drawn with a finger on a phone or with a trackpad is the main use, followed closely by circling the paragraph somebody needs to read. Strokes are written as a series of real line segments with round caps rather than as an image, so they stay perfectly sharp at any zoom and add almost nothing to the file size. Points are thinned as you draw, because a signature captured at full pointer resolution is several thousand segments and none of them are visible.
Drag a band over a passage. It is written with a multiply blend mode, which is what a real highlighter does: the yellow multiplies with the page underneath so the words stay completely readable. A plain opaque rectangle at reduced opacity is the wrong mechanism and it always looks like a stain. Any colour works, though yellow and green are the ones that stay legible over black text.
A solid rectangle, white by default, drawn over whatever is underneath. Use it to hide a stale page number, a header, a scanner artifact or text you are about to replace. It paints over content and never deletes it. The characters under the box stay in the file, still come out of any text extraction, and the box itself can be selected and deleted by anybody with a PDF editor. If the point is that nobody can ever recover what was there, Redact PDF is the tool that actually destroys it.
Drag a rectangle and give it a web address. That area becomes clickable in every PDF reader, because it is written as a proper link annotation instead of painted text, so it works in Acrobat, in browsers, on phones, and in the reader built into every email client. Two good uses: turning a printed URL in a scanned document into something people can actually click, and making a logo link to your site. Both take about five seconds.
Two ways to get a link onto the page. They solve different problems. Leave the anchor text empty and the rectangle stays invisible, which is what you want over a logo, a printed web address or a photograph already on the page. Fill the anchor text in and the words are drawn into the page for real, underlined, inside their own click target, which is what you want when there is nothing there yet. Any text you add with the Text tool can carry a link address of its own, so a sentence you type becomes the anchor.
No PDF editor can turn a phrase that was already in the document into a link, ours included. Text in a finished PDF is drawing instructions, not a selectable range, so there is nothing for an annotation to attach itself to. The workaround is to cover the old phrase and type it again as an object you control. That is it. Type a bare domain into the address field and it gets the https prefix added, while an email address becomes a mailto link, because a bare domain in a link annotation opens nothing at all.
There are two ways to write an edited PDF and they produce very different files. The one this tool uses is the one that leaves you with a document.
Your document is parsed and modified in place. Every existing page object is kept exactly as it was, with its content streams, its resources, its annotations and its structure tree untouched. Each edit appends a short set of drawing operations to the page it belongs to, and each link becomes a new annotation object attached to that page. Fonts are embedded once for the whole document, and images once each however many times they appear.
What survives, therefore:
The shortcut is to render every page into a bitmap, composite the edits onto it, and build a new PDF around the pictures. It is far easier to implement. It also destroys everything in that list in one pass: the text becomes pixels, the links become nothing, and a 400 KB contract becomes 14 MB of JPEG. If you have ever edited a PDF somewhere and found afterwards that you could no longer select any of the text, that is what happened to it.
Everything you place is stored as a proportion of the page as displayed, never as pixels and never as absolute points. That is what makes an object land in the same relative place regardless of your zoom, your screen or your device pixel ratio, and it is what lets a document with an A3 foldout in the middle behave sensibly.
Page rotation is mapped explicitly. A page stored sideways with a rotation flag is displayed upright, so an object placed by naive arithmetic ends up ninety degrees away from where it was dropped. Here the position is converted from the page as you see it into the file coordinate system, and the object itself is rotated to compensate, so a signature on a rotated page lands on the line you put it on and reads the right way up.
People arrive here wanting to sign a document, and the word covers two completely different things that are worth separating before you rely on either.
A picture of your signature placed on the page. This tool does it two ways: draw it with the Draw tool using a finger, a stylus or a trackpad, or photograph an ink signature, clean it up as a transparent PNG, and place it with the Image tool. For the overwhelming majority of everyday business documents this is exactly what "signed" means in practice and it is what the person on the other end is expecting.
Practical notes:
A cryptographic signature bound to a certificate, which proves who signed and proves the document has not changed by so much as a byte since. Different mechanism entirely. It requires a certificate from an authority, and it is what regulated processes mean when they say signature. This tool does not do it, and no browser tool that does not ask you for a certificate is doing it either, whatever the button says.
One consequence matters here regardless of which kind you need: if a document is already digitally signed, any edit you make invalidates that signature, including a single character of added text. That is the entire point of a signature and it is unavoidable in any software. Make your changes first. Sign afterwards, never the other way round.
The usual architecture for a free online PDF editor is an upload form, an editing session running against a copy held on a server you do not control, and a download link at the end. Your document crossed the internet, was written to a disk in a data center you have never seen, and sat there for the length of your editing session and however long afterwards a privacy page claims.
Think about what people actually edit: contracts before they are signed, invoices with bank details on them, offer letters, tenancy agreements, medical forms, passport scans. The category of document that needs editing is almost exactly the category that should not be uploaded to a stranger.
This tool has no upload endpoint, because it has no server-side step. Your PDF and any image you add are opened with the browser File API, the pages are rendered for you to work on by a renderer running as JavaScript on your processor, and the edited document is produced as a Blob your own browser downloads from itself. Every file involved exists on your disk and in your RAM, and nowhere else.
The single exception, named because it is the only request that can depend on what you typed: text containing a character outside Western European Latin causes a font file to be fetched from this domain so it can be embedded in your document. That request carries the name of a font and nothing else.
Stated in one place, so none of it is discovered halfway through a job. Every item here has a named alternative.
| Symptom | Cause | Fix |
|---|---|---|
| The cover box does not match the page | The page is not pure white. Scans are usually a light grey with a colour cast. | Zoom in, sample the page colour in any image editor, and set the cover colour to match. Or cover a whole band instead of a few words, so there is no edge to notice. |
| My new text does not match the old | The document uses a typeface that is not Helvetica, Times or Courier. | Those three cover Arial, Calibri, Times New Roman and most business documents by metric similarity. If the face is distinctive, replacing a whole line rather than a word hides the difference. |
| The signature has a white box around it | The image is a JPG, or a PNG saved without transparency. | Remove the background and re-export as PNG. JPEG has no alpha channel and never will. |
| My drawing looks jagged | Zoomed out while drawing. Points are thinned relative to the page, so a stroke drawn at 25 percent zoom has a quarter of the detail. | Zoom to at least 100 percent before drawing a signature. |
| The highlight hides the words | The colour is too dark for a multiply blend. Dark blue over black text leaves nothing. | Use yellow or a light green. Multiply darkens, so the lighter the highlight the more readable the text underneath. |
| Text I covered still shows up in search | Covering is paint, not deletion. The text is still in the file. | Use Redact PDF if it must be gone. This is the most consequential misunderstanding on this page. |
| Characters came out as question marks | Text contained a character outside Western European Latin and the font file could not be fetched, usually because of a blocker. | Allow this domain and run it again. The font request carries no document data. |
| The tab is slow on a long document | Page previews are held as bitmaps and the document is in memory. | Extract the pages you need first, edit those, and merge afterwards. Two operations beat one crashed tab. |
Editing is normally the middle of a job and not the end of one. These are the tools either side of it:
Drop the file into the zone above and click the page you want to change. It opens at full size with a toolbar beside it, and you can add text, drop in an image or a signature, draw freehand, highlight, cover something up or add a hyperlink. Press Save changes and the file comes back. It is free and there is no watermark on the result. No account is needed for your first run: an anonymous visitor gets one operation per 12 hours, a free CartHacker account gets two, and Premium removes the counter.
Not directly, and the reason explains every PDF editor on the market: a PDF does not contain paragraphs, it contains instructions to draw glyphs at coordinates. There is no sentence to click on. What works reliably is the two step method every browser editor actually uses: cover the old text with a box matching the page colour, then type the new text on top. For correcting a date, a price, a name or a reference number that is quick and the result is indistinguishable. For rewriting a paragraph, go back to the source document or convert to Word.
No. Your document is modified in place: every existing page object is kept as it was and each edit appends a short set of drawing operations to the page it belongs to. The original text layer stays searchable, existing links and bookmarks keep working, form fields stay fillable, and no existing image is decoded or recompressed. Tools that rasterize every page are far easier to build and they destroy all of that, which is why a document edited elsewhere sometimes comes back with no selectable text and ten times the file size.
Two ways. Use the Draw tool and sign with a finger, a stylus or a trackpad, which works well on a phone or tablet. Or photograph an ink signature on white paper, remove the background in any image editor, save it as a PNG with transparency, and place it with the Image tool. Zoom to at least 200 percent before positioning it, and add the date next to it as a text object. Note that this is a drawn signature, not a cryptographic digital signature bound to a certificate.
Real text, drawn with an embedded font reference. It stays selectable, searchable and copyable in the finished PDF. Helvetica, Times and Courier are three of the fourteen typefaces every PDF reader must carry, so adding text costs almost nothing in file size. If you type a character outside Western European Latin, such as a Polish or Greek letter, a real font file is fetched and only the glyphs you used are embedded.
No. Cover up paints a rectangle over the page. The content underneath is still in the file, still comes out of any text extraction, and the box can be deleted by anyone with an editor. For hiding a stale page number that is fine. For an address, an account number or a name it is a serious disclosure, and the tool you need is Redact PDF, which destroys the content instead of the view of it.
You can add anything you like to it: text, a signature, highlights, boxes and links all work exactly the same on a scan. What you cannot do is edit the words already on it. A scan is a photograph of a page, not text. Run OCR PDF first if you need the document to be searchable, and note that even after OCR the visible page is still the photograph and the text layer sits invisibly behind the picture.
You can place text on the lines, which prints and reads correctly and is the usual answer for a scanned or flattened form. That is not the same as filling a real form field: fields created by AcroForm are interactive objects rather than page content, and any PDF reader including your own browser will fill genuine ones. If the form has real fields, use your reader. If it is a picture of a form, use this.
No. Your PDF and any image you add are opened with the browser File API, the pages are rendered by JavaScript on your own processor, and the edited file is produced as a download your browser makes from itself. Load this page, disconnect from the internet entirely, and the whole editor still works. The only request that ever depends on your input is a font file fetch, and only when your text contains a character outside Western European Latin.
Yes. The arrows at the bottom of the editor walk through the whole document without closing it, so a signature on page 1 and initials on page 14 are one session and one operation. The tile for each page carries a badge counting what you have added to it, so you can see at a glance which pages you have touched.
Yes, and so does every other tool, because that is exactly what a digital signature is for. A signature proves the document has not changed by a single byte since it was signed, so any modification, including adding a single character of text or a highlight, correctly invalidates it. Make your changes first and sign afterwards. Any tool claiming to edit a signed PDF while preserving the signature is not preserving the signature.
The document limit is 100 MB and there is no page limit, but the practical ceiling is your own memory, since page previews are held as bitmaps in the tab. A few hundred pages is comfortable on a laptop; a very long high resolution scan is not comfortable on a phone. If a document is unwieldy, extract the pages you need, edit those, and merge afterwards. An operation is one save regardless of how many changes went into it.
All of these run the same way: your file is read by JavaScript on your own machine, the work happens on your processor, and the result is handed back to you as a download that never existed anywhere else.
Find the apps that slow your store down, by name and in milliseconds, then answer shoppers' doubts with trust badges under Add to cart. Both are free, with no card and no access to your store admin.
Paste any page and get its three Core Web Vitals, the apps costing time, like Klaviyo or Judge.me, and the Shopify or WooCommerce fix.
Show the promises that settle doubts, from free shipping to a money-back guarantee, right under the Add to cart button. One line in your theme.
Change the link after printing and count the scans. 5 tracked codes on a free account, static codes unlimited.
Anonymous visitors get one operation per 12 hours, per network address. That is not a sales tactic, it is the only lever we have: this tool runs on your processor, not ours, so the run itself earns us nothing. An account tells us who you are, and the ceiling goes up the moment you have one. Your allowance resets in 12 h.
No credit card, no trial period, no sales call. Join with Google, GitHub, Facebook or email.
Free accounts get two operations per 12 hours, and yours reset in 12 h. Waiting costs you nothing and is a completely reasonable choice. If PDF work is a weekly job rather than a one-off, the counter is the wrong tool for you and Premium takes it off the page.
Cancel anytime. Everything you have already produced stays yours either way.