Merge PDF files free, without uploading, and make your own PDF tool with AI

Your own PDF tool, on your own computer

You'll ask a free AI assistant for a small Python script that joins PDFs together. Put files in a folder and run it, or drop files straight onto it. Then add options to split a PDF and pull out just the pages you need.

• A script that merges every PDF in a folder, in name order
• Or just the files you give it, without touching the originals
• A split option that saves every page as its own file
• A page-range option, like "pages 1-3 and 7", and a way to turn pages round
• Drag and drop: drop files onto an icon on Windows, or right-click them on a Mac
• If you'd like one, a browser version in a single HTML file, like the tool above

Good for: anyone who sends forms, receipts or scanned letters as one file, people who'd rather not upload private documents to a website, and anyone who wants a gentle first script that's useful straight away.

What you need

Build it in six steps

Step 1 gives you a simple script that works, and each step after it adds one more feature. Your script won't look like the tool at the top of this page: it's a different kind of tool that runs on your own computer.

1. Install Python and pypdf

   Follow How to install Python first. Then ask the AI to walk you through the one add-on the script needs, pypdf, which reads and writes PDFs:

   Copy this promptCopy

   I've just installed Python on my [Windows / Mac] computer and I don't know how to code. In very simple steps, tell me how to:
   - check Python is working
   - install the "pypdf" add-on with pip
   - make a new folder called pdf-tools in my Documents folder, with a folder inside it called to-merge
   - open a terminal in the pdf-tools folder
   
   Tell me exactly what to type and what I should see if it worked.

2. Ask for the merge script

   Copy this prompt into your AI assistant:

   Copy this promptCopy

   I don't know how to code. Please write me a Python script called merge_pdfs.py that joins PDF files into one. It's for my own files.
   
   Requirements:
   - ONE complete Python file using the "pypdf" add-on, which I have installed.
   - If I run it with no file names, it merges every PDF in the "to-merge" folder next to the script, in name order. Sort numbers the way a person would, so "2.pdf" comes before "10.pdf".
   - If I give it file names, it merges just those, sorted by name in the same way.
   - Save the result as merged.pdf next to the script. If that file already exists, use merged-2.pdf, merged-3.pdf and so on, so it never overwrites anything.
   - Print each file name and its page count as it goes, then the total pages and where it saved the file.
   - If a file is password-protected or damaged, print a clear message saying which one, skip it, and carry on.
   - Never change or delete the original files.
   
   Please give me the complete file in one code block, then tell me in simple steps how to save it in my pdf-tools folder and run it.

   Copy the code, paste it into your text editor, and save it as merge_pdfs.py in your pdf-tools folder.

3. Run it

   Put two or three PDFs in the to-merge folder. If the order matters, start each name with a number: 01 cover, 02 letter, 03 receipts. Open a terminal in your pdf-tools folder and type python merge_pdfs.py (on a Mac, python3 merge_pdfs.py). You should see each file and its page count, and a new merged.pdf in the folder. Open it and check the pages are all there, in the right order.

   If anything looks wrong, paste what you saw straight back into the chat:

   Copy this promptCopy

   I ran merge_pdfs.py and got this:
   
   [paste everything the terminal printed]
   
   Please explain what it means in plain English. If something needs fixing, send me the complete fixed file.

4. Add split and page-range options

   The same script can take PDFs apart as well. Ask for a few simple commands:

   Copy this promptCopy

   Please add some options to merge_pdfs.py, keeping everything it does now:
   - Split: "python merge_pdfs.py split myfile.pdf" saves every page as its own PDF in a new folder named after the file, like myfile-page-01.pdf.
   - Pages: "python merge_pdfs.py pages myfile.pdf 1-3,7" saves just those pages, in that order, as a new file called myfile-pages.pdf.
   - Rotate: "python merge_pdfs.py rotate myfile.pdf 2,5 90" turns those pages a quarter turn clockwise and saves a copy called myfile-turned.pdf.
   - If I ask for a page that doesn't exist, tell me how many pages the file has instead of crashing.
   - "python merge_pdfs.py help" prints a short list of these commands with an example of each.
   
   Send me the complete updated file.

   Tip: to remove a page, use the pages option and leave that page out. "1-4, 6-12" skips page 5.

5. Make it drag and drop

   Typing commands gets old. On Windows, a .bat file is a tiny file that runs commands for you: drop PDFs onto it and Windows hands it their names. On a Mac, a Quick Action adds your own item to the menu you get when you right-click files in Finder. The Shortcuts app can make one in a few clicks.

   Copy this promptCopy

   I'd like to merge PDFs without opening a terminal. My merge_pdfs.py is in Documents/pdf-tools on my [Windows / Mac] computer.
   
   - On Windows: write me a file called "Merge PDFs.bat" that I can keep on my desktop. When I drop PDFs onto it, it runs merge_pdfs.py on those files, using the full path to the script, and keeps the window open at the end so I can read what happened.
   - On a Mac: give me click-by-click steps to make a Quick Action with the Shortcuts app (or Automator, if that's simpler), so I can select PDFs in Finder, right-click, and choose Quick Actions, then Merge PDFs. Use the full path to python3 and to the script.
   - Update merge_pdfs.py so that when it's given files, it saves merged.pdf in the same folder as the first file.
   
   Send me the complete updated merge_pdfs.py as well.

   Number your files if order matters. Windows doesn't always pass dropped files in the order you see them, which is why the script sorts them by name.

6. Optional: a browser version in one file

   Would you rather click than type? Ask for a single web page that works like the tool at the top of this page. It loads a free PDF library called pdf-lib from the internet, but your files are still only read on your own computer.

   Copy this promptCopy

   Please make me a single HTML file called pdf-merge.html that merges PDFs in my browser, with no uploads. Use the pdf-lib library from a CDN (jsDelivr or unpkg), with a fixed version number.
   
   - A big area I can drag PDFs onto, plus a "Choose files" button.
   - Show the files as a list with their page counts, with up and down buttons to change the order and a remove button for each.
   - A "Merge and download" button that saves merged.pdf.
   - If a PDF is password-protected, say so clearly instead of failing silently.
   - A clean, simple design that works on a phone too.
   
   Give me the complete file in one code block, and tell me how to save it and open it.

   Not sure how to save and open it? See How to save and open an HTML file.

Quick fixes

Twist it into something else

Once you have a script that opens PDFs, lots of small jobs are one prompt away. Ask the AI to add:

Some files shouldn't go to random websites

Online PDF sites are handy, but every file you merge there is uploaded to someone else's computer. Most say they delete files after an hour or two. You have to take their word for it. That's fine for a takeaway menu. It's less fine for a contract, a payslip, a bank statement or a letter from your doctor. Your own script runs on your computer, works without the internet, has no daily limits or watermarks, and never asks you to sign up. It's also a gentle first script: once you've made one thing that tidies up files for you, you'll spot lots of other jobs it could do.

What to build next