AI & Documents · 2026

How to Generate PDFs with Claude & ChatGPT

8 min read Last updated: August 2026

Neither Claude nor ChatGPT can generate PDFs on their own. But when connected to DocuQueue via MCP, they get tools to create invoices, contracts, certificates, and any other document right from chat. One config entry, no code, live in 2 minutes.

TL;DR: Sign up at DocuQueue, add one config entry to your AI assistant, restart, and ask it to generate a document. The MCP connector handles templates, rendering, and download links. Works on Claude Desktop, Claude Code, ChatGPT, and Gemini.

You ask your AI
→
MCP / Tool Use
→
API renders PDF
→
Download link
A
Written by Arun — Founder, DocuQueue
10+ years in document automation • Last updated August 2026
Overview

What is the DocuQueue MCP connector?

MCP (Model Context Protocol) is an open standard that lets AI assistants call external tools. DocuQueue's connector gives Claude, ChatGPT, and Gemini the ability to generate PDF documents directly from chat.

When you add DocuQueue to your AI assistant, it gains a set of document tools: generate from templates, fill existing forms, batch-produce certificates, merge files, translate, and more. You type what you want in plain English, the assistant calls the right tool, and DocuQueue renders the PDF.

Setup takes about 5 minutes. No code required for the MCP path. For a step-by-step walkthrough, see our guide to connecting Claude to document generation. The MCP server runs on DocuQueue's infrastructure, so there's nothing to install or maintain on your machine. DocuQueue offers a free tier with 25 credits per month after email verification, which is enough to test and run small workflows.

MCP is available on Claude Desktop, Claude Code, ChatGPT, and Gemini. All four platforms get the same set of tools through the same connector. For a broader look at document automation, see our document workflow automation guide.

Decision Guide

Which AI assistants are supported?

Four platforms, same document generation tools. Pick the one you already use.

Claude Desktop

Best for interactive chats and ad hoc documents. One config entry, restart, and you're live.

Setup: Config file edit, restart

Claude Code

Best for developers who work in the terminal. One CLI command to add the MCP server.

Setup: One CLI command

ChatGPT

Best for broadest access. Add DocuQueue via Settings or MCP config. Works on free tier.

Setup: Settings or MCP config

Gemini

Google's AI assistant supports MCP and connects to DocuQueue the same way. Add the server config and restart.

Setup: MCP config
Platform Setup method Best for Tools available
Claude Desktop Config file edit Interactive chats, prototyping generate, fill, list, batch, merge, translate
Claude Code One CLI command Developers, terminal workflow generate, fill, list, batch, merge, translate
ChatGPT Settings or MCP config Broadest access, free tier generate, fill, list, batch, merge, translate
Gemini MCP config Google ecosystem, multimodal generate, fill, list, batch, merge, translate
Demo

What it looks like

Once connected, you ask your AI assistant to generate documents in plain English.

Claude Desktop
You: Generate an invoice for Acme Corp, $5,500, due in 30 days
Claude: Creating document...
Template: Invoice
Status: completed
Document: https://docuqueue.com/api/v1/download/...
Invoice generated and ready to send.
ChatGPT
You: Generate an invoice for Acme Corp, $5,500, due in 30 days
ChatGPT: Creating document...
Template: Invoice
Status: completed
Document: https://docuqueue.com/api/v1/download/...
Invoice generated and ready to send.

Your AI assistant calls the generate_document tool behind the scenes. You don't need to write HTML, manage templates, or handle PDF rendering. The assistant fills the structured fields, the API renders the PDF, and you get a download link.

Setup

How do I connect DocuQueue to my AI assistant?

Add the MCP endpoint to your client config, then restart. You need an API key (format wp2_...) from your dashboard.

01

Get your API key

Sign up at docuqueue.com/register and copy your API key from the dashboard. The free tier includes 25 credits per month after email verification.

02

Claude Desktop

Edit claude_desktop_config.json (Settings → Developer → Edit Config) and add an mcpServers entry:

claude_desktop_config.json
{
  "mcpServers": {
    "docuqueue": {
      "url": "https://docuqueue.com/mcp/v1",
      "transport": "streamable-http"
    }
  }
}
03

Claude Code

Claude Code connects to remote MCP servers from the command line:

Terminal
# Add DocuQueue MCP server
claude mcp add docuqueue https://docuqueue.com/mcp/v1 \
  --transport streamable-http

# Verify connection
claude mcp list

Run claude mcp list to confirm the connection.

04

ChatGPT

ChatGPT supports MCP via Settings or custom MCP server config:

Option A: Settings
# 1. Open ChatGPT Settings
# 2. Go to Connected apps (or Data controls)
# 3. Search for DocuQueue
# 4. Connect and authorize with your API key
Option B: MCP config
{
  "mcpServers": {
    "docuqueue": {
      "url": "https://docuqueue.com/mcp/v1",
      "transport": "streamable-http"
    }
  }
}

After connecting, DocuQueue tools appear in your ChatGPT conversations automatically.

Security

Your API key stays in your local config. DocuQueue never stores chat messages, and the MCP server runs over HTTPS. Keep your key out of shared configs, and use a scoped key for MCP setups so a leaked config cannot delete your templates.

Pricing

Do I need a paid plan?

No. The MCP connector is available on DocuQueue's free tier. You get 25 credits per month after email verification at no cost, no credit card required. The MCP path requires no code, no SDK, and no server infrastructure. Sign up, grab your API key, add the config entry, and start generating.

Paid plans unlock higher document limits, priority rendering, and access to premium templates. But for testing and small workflows, the free tier covers most use cases.

Capabilities

What can I ask my AI assistant to do with DocuQueue?

Once connected, your AI assistant has access to these tools.

Capability What it does
Generate documents Create invoices, contracts, certificates, and letters from templates
Fill templates Fill existing PDF forms with specific data like names, dates, and amounts
List templates See what templates are available in your DocuQueue account
Batch generate Create multiple documents from a CSV or list in one operation
Edit documents Modify existing PDFs, replace text, update dates across all instances
Translate Translate documents to other languages while preserving layout
Merge Combine multiple documents into one PDF
Download Get a direct download link for any generated document
Try it

What are some useful things to try?

Eight prompts to get you started. Copy, paste, and see what your AI assistant does.

  1. Generate a standard document
    Prompt: "Generate a payment invoice: Acme LLC, $15,000, software development services, March 2026."

    Your AI fills the invoice template with the line items, company details, and payment terms. The API renders the PDF with your branding.

  2. Fill a template for a specific client
    Prompt: "Fill the NDA template: Acme Corp (disclosing), Beta Inc (receiving), effective January 2027."

    The AI calls fill_template with the exact fields. No copy-pasting, no manual form filling.

  3. Batch-generate personalized proposals
    Prompt: "Here's my proposal template. Generate individual PDFs for each of these 10 clients: [paste table]."

    The AI calls batch_generate, iterating through the list and producing 10 personalized proposals in one operation.

  4. Create a contract
    Prompt: "Draft a service agreement between us and GlobalTech for $50,000/year, starting April 2026."

    Your AI assembles the contract from your template, fills in the parties, scope, and terms, then renders a downloadable PDF.

  5. Generate certificates
    Prompt: "Create completion certificates for these 50 employees: [paste names]."

    The batch tool produces 50 personalized certificates in one go. Each one gets the correct name, date, and program details.

  6. Prepare a deal package
    Prompt: "Merge the contract, Exhibit A, and the invoice into one PDF and compress it."

    Your AI assembles multiple documents into a consolidated package, ready for review or signature.

  7. Edit existing documents
    Prompt: "In the contract, replace 'Old Corp' with 'New Corp' everywhere and update the date."

    The AI modifies the existing document, applying changes across all instances without manual find-and-replace.

  8. Translate a document
    Prompt: "Translate this NDA from English to Spanish, keeping the legal structure."

    Your AI produces a translated version while preserving the document layout, formatting, and legal structure.

Decision

Which option should you choose?

Pick by where your AI runs and who owns the render.

MCP for interactive chats

One config entry, no code. Your AI gets document tools in every conversation. Best for prototyping and ad hoc work.

Tool Use for production apps

Define generate_document in your code, run the loop server-side, keep API keys secure. Best for shipping a product feature.

Not sure yet?

Start with MCP to feel out how your AI writes the data, then port the same call into a tool definition when you ship.

FAQ

Common questions

Can Claude generate PDF documents?
Yes. Claude cannot emit binary files directly, but when connected to a document generation API via MCP, it can call tools to create PDFs, DOCX, and other formats. Claude writes the content, the API renders it.
Can ChatGPT generate PDF documents?
Yes. ChatGPT supports MCP and can connect to document generation APIs. Once connected, it has the same tools as Claude. The setup is similar: add the MCP server config and restart.
What is MCP for document generation?
MCP (Model Context Protocol) is an open standard that lets AI assistants connect to external tools. A document generation MCP server exposes tools like generate_document, fill_template, and list_templates that Claude, ChatGPT, and Gemini can call directly from chat.
Do I need to install anything to use MCP with Claude or ChatGPT?
No. For Claude Desktop, add a config entry and restart. For Claude Code, run one command. For ChatGPT, add the MCP server via Settings. No SDK, no npm install, no Docker. The MCP server runs on DocuQueue's infrastructure.
Do I need a paid plan for ChatGPT?
No. MCP works on ChatGPT's free tier. You can connect DocuQueue and start generating documents without a paid subscription.
Which is better for document generation, Claude or ChatGPT?
Both work equally well for document generation via MCP. Pick the assistant you are most comfortable with. Claude Desktop is popular with developers; ChatGPT offers broad accessibility. The document generation tools are the same regardless of which assistant you choose.
Can my AI assistant access other users' documents?
No. The MCP connector is strictly scoped to your account. Your AI assistant can only access templates and documents that belong to your DocuQueue account. There is no cross-account access.
Are my documents safe?
Yes. DocuQueue uses TLS 1.2+ for data in transit and encryption at rest. Your documents are not used to train any AI models. The MCP server runs over HTTPS, and API keys are scoped to limit access.

About this guide: This guide is maintained by the DocuQueue team. Setup instructions are verified as of September 2026.

Try it with Claude
Create a client proposal for Meridian Labs with our enterprise pricing
✓ Template Client Proposal found
✓ Filling fields: client name, pricing tier, terms
✓ PDF rendered — meridian-proposal.pdf
C
Done. Your proposal is ready: meridian-proposal.pdf

Start generating documents with Claude or ChatGPT

25 credits per month after email verification. No credit card required. Connect in 2 minutes.

Get Started Free
A
Arun — Founder, DocuQueue
10+ years building document automation tools. Has tested MCP integrations with Claude, ChatGPT, Gemini, and Cursor. Built DocuQueue's MCP server with 12 tools.

How This Guide Was Created

This guide was written by Arun, who has personally tested MCP integrations with Claude, ChatGPT, Gemini, and Cursor. We set up each platform, ran test generations, and documented the exact steps. The MCP setup instructions were verified against the official MCP specification (modelcontextprotocol.io). All code examples were tested in production.

References