iLovePDF is now in ChatGPT, and ready for your other AI tools too

You're in the middle of a ChatGPT conversation, drafting an email or preparing a report, and suddenly you need to do something with a PDF. Make it smaller. Combine it with two others. Turn it into a Word file so you can edit it.

Until now, that meant leaving the chat, opening another tab, uploading the file somewhere else and coming back.

Not anymore. The iLovePDF plugin for ChatGPT (you may also see it called an app) is now live, so you can handle your PDFs right inside the conversation. And if you work with other AI assistants, like Claude Desktop or Cursor, the new iLovePDF MCP server brings our PDF tools there too.

Key takeaways

• The iLovePDF plugin for ChatGPT is now live, with a dedicated interface right inside the chat.
• In ChatGPT, you can compress, merge, split and convert files, add watermarks and page numbers, and use OCR on scanned documents.
• There's nothing to install for ChatGPT. Connect iLovePDF, upload a file and ask in plain language.
• The iLovePDF MCP server gives Claude Desktop, Cursor and other MCP-compatible clients access to 10 iLovePDF tools.
• With the MCP server, files are read from and saved to your own computer, and they aren't sent to OpenAI or Anthropic.
• You connect the MCP server with your own iLovePDF API key, so the credentials, usage and control stay with you.

Why bring PDF tools into AI assistants?

AI assistants are great at understanding what you want. What they don't always have is a reliable way to actually do something with your documents.

Ask a chatbot to "shrink this PDF so I can email it" and, without the right tool, the best it can do is explain how you might do it yourself. With iLovePDF connected, it can just do it.

That's the idea behind both launches. You describe the task in plain language, iLovePDF handles the document work, and you stay in the tool you were already using.

Both are built on the same foundation. The iLovePDF MCP server makes our PDF tools available to AI assistants through a standard protocol, and the ChatGPT plugin builds on those same capabilities with a dedicated interface on top.

iLovePDF in ChatGPT

What is the iLovePDF plugin for ChatGPT?

It's a way to use iLovePDF's tools directly inside ChatGPT. Instead of clicking through menus, you describe what you want in your own words, and iLovePDF does the document work in the background.

Because it comes with its own interface inside the chat, you can follow what's happening as your file is processed and see the result clearly when it's ready.

It's designed for everyday tasks, the quick jobs that usually interrupt what you're doing. Resizing an attachment, putting a contract together, getting text out of a scan.

Do you need to install anything to use iLovePDF in ChatGPT? No. You just connect it in ChatGPT and start using it.

What you can do with iLovePDF in ChatGPT

Compress PDFs. Large files get bounced by email servers and rejected by upload portals. Ask ChatGPT to compress your PDF and you'll get a lighter version that's easier to share. Here's how PDF compression works in iLovePDF if you're curious.

Merge and split PDFs. Combine an invoice, a contract and its annex into one file, or pull out only the pages you need. It's a simple way to merge PDF files into one document without leaving the conversation.

Convert files to and from PDF. Turn a PDF into Word, Excel, PowerPoint or an image so you can edit or reuse its content. Or convert those formats back into PDF when something needs to look the same on every screen. For editable text, our PDF to Word converter shows what to expect.

Add watermarks. Mark a document as a draft, confidential or yours before it goes out. It's a quick way to watermark a PDF before sharing it, especially outside your team.

Add page numbers. Long reports, contracts and manuals are easier to review when every page is numbered.

Extract text from scanned documents with OCR. Scans often look like text but behave like images. OCR (Optical Character Recognition) recognizes the words on the page so you can search, copy and reuse them. See for yourself how OCR works in iLovePDF.

How to use iLovePDF in ChatGPT

Getting started takes about a minute.

1. Follow this link or open ChatGPT and find iLovePDF in the Plugins section.
2. Connect iLovePDF to your ChatGPT account.
3. Upload your file in the chat. It can be a PDF, a Word, Excel or PowerPoint document, or an image.
4. Ask for what you need in your own words.
5. Download the result directly from the conversation.

Because you're still in a chat, you can keep going. Ask for one change, look at the result, then ask for the next.

Example requests to try in ChatGPT

You don't need to know any tool names. Just describe the result you want:

"Compress this PDF so I can send it by email."

"Merge these three files into one PDF and add page numbers."

"Convert this PDF to Word so I can edit it."

"Add a 'Confidential' watermark to this contract."

"This is a scanned receipt. Extract the text with OCR."

You can also chain tasks in one request, like: "Convert these images to PDF, merge them, and compress the final file."

The iLovePDF MCP server for Claude, Cursor and more

ChatGPT is where most people will meet iLovePDF in an AI assistant first. But plenty of work happens in other AI tools, and that's where the iLovePDF MCP server comes in.

What is MCP?

MCP stands for Model Context Protocol, an open standard that lets AI applications connect to external tools in a consistent way. Instead of every AI app building its own integration with every service, one MCP server can work with many different clients.

In practice, that means one iLovePDF MCP server can give PDF capabilities to several AI tools you already use.

What the iLovePDF MCP server does

The iLovePDF MCP server gives any MCP client, such as Claude Desktop or Cursor, direct access to iLovePDF's PDF tools. It runs as a small local server on your computer and is open source under the MIT license.

So instead of uploading files to a chat, your AI assistant can pick up a file from a folder, process it with iLovePDF and save the result right back to your drive. It can also work with files from a web link. Document processing becomes part of your AI-powered workflow, without constantly switching between tools.

How it works, step by step

1. You make the request and point to the document, either a file on your computer or a web link.
2. Your AI client understands the request and picks the right iLovePDF tool.
3. The MCP server connects your AI client with iLovePDF's services.
4. iLovePDF processes the document in the cloud and returns the result, which is saved to your computer.

The 10 tools available through the MCP server

1. Compress PDF to reduce file size.
2. Merge PDF to combine multiple PDFs into one.
3. Split PDF by page range or into fixed chunks.
4. PDF to JPG to turn PDF pages into images.
5. Image to PDF to convert JPG, PNG or TIFF images into a PDF.
6. Office to PDF to convert Word, Excel or PowerPoint files.
7. Unlock PDF to remove passwords and restrictions from PDFs you have the right to open.
8. Watermark to add a text or image watermark.
9. Page numbers to number every page of a document.
10. OCR to extract text from scanned PDFs.

Which AI tools does it work with?

Does the iLovePDF MCP server only work with Claude? No.

It works with Claude Desktop, Cursor and any other client that can host an MCP server locally. As more AI tools adopt the protocol, the list of places you can use iLovePDF keeps growing.

One local MCP server, your favorite MCP client, and PDF tools ready to become part of how you already work.

How your files are handled with the MCP server

Are your files sent to OpenAI or Anthropic when you use the iLovePDF MCP server? No.

The server reads files from your local filesystem and writes the results back to it. Your documents go to iLovePDF for processing and nowhere else. Your AI assistant tells the server what to do, and the server takes care of the files.

File access stays within the folders you choose. The server works inside a working directory you define, plus any extra folders you explicitly allow. Attempts to reach files outside those locations are blocked.

Multi-file results come as one ZIP. Operations that generate several files, like splitting a PDF or converting pages to JPG, are saved as a single .zip archive, so your folders don't fill up with loose files.

You stay in control of your usage. The server connects with your own iLovePDF API key. That means your credentials, your usage and your control, with no third party in between. It's a setup that works for individuals and for companies that want to manage access themselves.

How to set up the iLovePDF MCP server

The full instructions, ready-to-copy configuration examples and optional settings are in the iLovePDF MCP setup guide on GitHub.

Setup takes a few minutes:

1. Install Node.js 18 or later on your computer, if you don't already have it.
2. Get your iLovePDF API key by creating a project in the iLoveAPI developer portal.
3. Add the iLovePDF MCP server to your client's configuration, for example, in Claude Desktop or Cursor, using your public key. The server runs with a single npx command, so there's nothing else to download.
4. Restart your client and ask it to work with a PDF on your computer.

Once it's connected, try something like "Compress every PDF in my Invoices folder" or "Split report.pdf into single pages."

ChatGPT plugin or MCP server: which one is right for you?

Both give your AI assistant access to iLovePDF. They just fit different ways of working.

Choose the ChatGPT plugin if you already use ChatGPT and want the fastest route. There's nothing to install. Upload a file, ask, and download the result. It's ideal for quick, everyday tasks.

Choose the MCP server if you work in Claude Desktop, Cursor or another MCP client, or want your assistant to work directly with files on your computer. It's especially handy for developers, power users, teams that want to manage their own API access and anyone processing documents in batches.

And you don't have to pick just one. Many people will use ChatGPT for a quick one-off and let the MCP server handle the heavier, recurring work.

Is it secure to use iLovePDF with AI assistants?

Whichever option you choose, your files are processed by iLovePDF with the same security standards that apply across our platform. It's the same iLovePDF that millions of people use for quick fixes and that international enterprises rely on for document management, PDF processing and e-signatures.

Keep in mind that files you upload in a ChatGPT conversation are also subject to ChatGPT's own data settings, which you can review in your account. With the MCP server, files stay on your computer and go only to iLovePDF for processing.

PDF tools, wherever your work happens

In summary: iLovePDF is now available right where a lot of your work happens. The ChatGPT plugin lets you compress, merge, split, convert, watermark, number and OCR files without leaving the conversation. The iLovePDF MCP server brings 10 iLovePDF tools to Claude Desktop, Cursor and other MCP clients, working with files on your own computer and your own API key.

This is just the start. We're already exploring new AI capabilities, platforms and integrations to keep expanding how you can work with documents.

Next time a PDF gets in the way of a task, don't stop and switch apps. Just ask your AI assistant to handle it with iLovePDF.

Your next PDF task is one message away

Try iLovePDF in ChatGPT