May 29, 2026
pdfAssistant Now Plugs Directly into Claude, ChatGPT, and More
Process & convert PDFs inside ChatGPT, Claude, and Codex, without changing tabs.
Your everyday AI tools are great at writing, analyzing, and answering questions. But ask them to convert a Word document into a compliant PDF/A file, compress a 40MB contract, or run a high-volume batch OCR—and they hit a wall. They simply weren't built for heavy-lifting document processing.
That is exactly why we built pdfAssistant as a dedicated AI assistant specifically for these workflows. While you can always log in to our standalone web platform to get this work done, we know many of you already have a primary AI workspace where you prefer to operate. Now, pdfAssistant can meet you there.
Today, we are launching a direct integration using the industry-standard Model Context Protocol (MCP). This gives your favorite AI platform secure, real-time access to pdfAssistant's enterprise-grade processing suite.
You can merge, split, compress, convert, OCR, watermark, encrypt, and flatten forms dynamically—all from inside the chat interface you are already using. Best of all, connecting your account takes less than 60 seconds thanks to secure OAuth single sign-on.
What You Can Do
Once authenticated, your AI client gains access to our professional PDF tools. Here is exactly what you can prompt your assistant to do:
| Category | Available Tools | Example Prompt |
|---|---|---|
| Convert | Word, Excel, and PowerPoint to PDF; Webpage URL to PDF; Uploaded HTML file to PDF; Email to PDF; PDF to Word, Excel, or PowerPoint; PDF to image formats (PNG, JPEG, TIFF, GIF, BMP); Graphics to PDF | "Convert this live pricing webpage URL and this PowerPoint presentation into PDFs." |
| Combine & Split | Merge multiple PDFs, Split PDF by page range, ZIP/unZIP files | "Split this 100+ page PDF into files no more than 10 pages each and ZIP all output files." |
| Optimize | Compress PDF, Linearize PDF (web-optimize), Flatten forms, Flatten annotations, Flatten layers, Flatten transparencies, Convert colors, Convert to PDF/X, Rasterize PDF | "Compress this PDF to under 5MB and linearize it so it opens fast on the web." |
| Secure & Edit | Encrypt PDF, Add/remove restrictions, Text redaction (literal strings, presets, or regex), Add text/images to PDF, Attach files, Add watermark | "Use a regex pattern to redact all Social Security Numbers from this document text." |
| Extract | Extract text, Extract images, Export form data, File read/write/append/list helpers | "Extract all the data from the form fields in this document." |
| Analyze | Check document metadata (accessibility tags, security settings, creator/origin info, text-extractability status) | "Check the metadata of this file to see if it is fully tagged for accessibility." |
| Sign & Forms | Digitally sign PDF, Import form data, Convert XFA to AcroForms | "Convert this old XFA form into a standard AcroForm so I can sign it." |
| Archive | Convert to PDF/A (archival standard) | "Convert this contract to PDF/A for long-term archival." |
How It Works
Think of it like adding an app to your smartphone. Your phone's operating system cannot do everything natively, so you connect apps to give it new abilities. Your primary AI platform works the same way. It is an elite reasoning engine, but connecting pdfAssistant gives it a specialized document toolkit it didn't have before.
The integration uses Model Context Protocol (MCP), an open-standard JSON-RPC 2.0 protocol that allows AI models to safely discover and execute external tools.
The Frictionless 3-Step Setup
- Link Account: Open your AI client's application settings or terminal interface, add the pdfAssistant remote server endpoint, and click Link Account.
- Secure OAuth Login: A secure popup window will open. Log in using your existing pdfAssistant credentials via OAuth to securely authorize your preferred AI workspace. Please note that an active paid pdfAssistant plan is required to unlock the integration and access your subscription tools and credits.
- Run Automations: Your setup is ready. Drop a document into your chat window and describe your target in plain English. The platform calls the exact pdfAssistant tools it needs for your request, processes the file in the cloud, and delivers the result right in your chat window.
Where It Works
Our integration runs on a high-performance Streamable HTTP transport architecture, built to the open MCP specification. The central hub for all connections is our official endpoint:
https://chat.pdfassistant.ai/mcp
We have fully verified and support pdfAssistant on the following platforms:
| Platform | Integration Mode | How to Connect |
|---|---|---|
| Claude Desktop | Streamable HTTP / CLI |
Run the setup command:claude mcp add --scope user --transport http pdfAssistant https://chat.pdfassistant.ai/mcpThen launch Claude, type /mcp, and complete the OAuth login.
|
| ChatGPT (Web) | Streamable HTTP / Developer Mode | Enable Developer Mode in Settings → Apps, configure a remote app pointing to https://chat.pdfassistant.ai/mcp, and log in via the OAuth prompt. |
| Codex CLI | Streamable HTTP / CLI |
codex mcp add pdfAssistant --url https://chat.pdfassistant.ai/mcp codex mcp login pdfAssistant |
Works anywhere MCP does. Because pdfAssistant's server is built to the open Model Context Protocol standard, it is designed to be compatible with any AI platform that supports MCP over Streamable HTTP. The platforms above are the ones we have fully tested and officially support today. If your preferred AI tool supports MCP, we encourage you to connect and try it — the setup process is the same across any compliant client.
Why This Matters
Most AI assistants can talk about a document. They can summarize a PDF, extract key points, or answer questions about its content. But they cannot safely perform structural modifications to the file binary itself. They cannot compress an oversized presentation, flatten form layers, or convert raw images into searchable text.
That is the gap pdfAssistant bridges. When connected via MCP, your AI moves from basic text comprehension to active document manipulation. It gains a toolbox of professional-grade actions that it can combine on the fly to handle complex document tasks. All document processing takes place asynchronously in our secure cloud utilizing the enterprise-trusted pdfRest engine—ensuring blazing fast response speeds and flawless rendering quality.
Security & Privacy First
pdfAssistant is designed with enterprise security in mind. Our MCP integration uses standard OAuth 2.0 authentication so your pdfAssistant credentials are never exposed to your AI platform or stored locally.
- SOC 2 Type 2 Compliant: Our security controls, infrastructure, and data protection practices are rigorously audited and certified to meet strict enterprise standards.
- pdfAssistant never trains on your data: Your documents, prompts, and extracted data are never used by pdfAssistant to train AI models. Please note that what your AI platform does with data in its own environment is governed by its own privacy policy.
- Encrypted in transit: All files and data sent to pdfAssistant's servers are protected with TLS 1.3 encryption, the same standard used by banks and healthcare systems.
Get Started
The pdfAssistant MCP server is available now for all users on a Plus, Pro, or Enterprise plan. Connect your AI assistant in under 60 seconds — just add the server endpoint, log in with your pdfAssistant credentials, and start processing.
Already on a paid plan? Open your AI client's settings and connect to https://chat.pdfassistant.ai/mcp to get started immediately.
New to pdfAssistant? Sign up and subscribe at pdfassistant.ai/pricing to unlock MCP access along with Workflows, Batch Processing, and 40+ PDF operations.
Topics
