Hey developers! Wouldn’t it be cool if your code editor already knew the answers to your project-specific questions? Imagine not having to leave your IDE to look up that one function in your team’s private API documentation.
Refer to our Cusror + CustomGPT.ai’s Hosted MCP server docs for more info.
By connecting CustomGPT.ai to Cursor, the AI-first code editor, you can make this happen.
What’s MCP? (The Simple Version)
Think of the Model Context Protocol (MCP) as a standard plug. It lets your tools, like the Cursor IDE, connect to data sources, like a CustomGPT.ai project full of your documentation. It’s a simple, secure way to make your tools smarter with your own data.
Why use CustomGPT.ai with Cursor?
- Find the right answer: Our system is top-ranked for accuracy, so you can trust the code snippets and explanations it gives you.
- No extra work: We manage the server, the security, and all the boring stuff.
- No extra cost: It’s included in your CustomGPT.ai plan.
For a more technical look, head over to our Hosted MCP launch blog post.
How Your Question Finds an Answer
With Cursor, the connection is direct and simple:
You, asking a question in Cursor (`@customgpt-mcp-server`)
│
│ (A direct, secure line over the internet)
▼
┌───────────────────────────────────────────┐
│ CustomGPT.ai Server │
│ (Instantly searches your private docs) │
└───────────────────────────────────────────┘
│
│ (Sends the answer straight back)
▼
The answer appears right in your Cursor IDE!What you’ll need first:
- A CustomGPT.ai account (free trial is fine).
- A project in CustomGPT.ai loaded with your documentation or knowledge base.
- The Cursor IDE.
Let’s Hook It Up: Step-by-Step
- Get your server URL: Log in to CustomGPT.ai. Go to your project’s Deploy ➞ MCP Server (Beta) tab. Generate a token if needed. You’ll see a JSON block; you just need the url from it.
- Open Cursor settings: In Cursor, go to Settings → Cursor Settings → MCP.
- Add your server: Click on + Add new global MCP server.
- Paste the config: Paste this little piece of code, but replace the url with the one you copied from your CustomGPT.ai dashboard.
{
"mcpServers": {
"customgpt-mcp-server": {
"url": "https://mcp.customgpt.ai/projects/<PROJECT_ID>/sse?token=<TOKEN>"
}
}
}- Save and you’re done!
Now, just hit ⌘ K in Cursor and type @customgpt-mcp-server to ask questions. Autocomplete and explanations will now be powered by your own private docs.
Watch it in action:
Your Turn to Build!
So there you have it. No more generic answers! Your favorite AI tool is now your personal expert, with a direct line to the documents and data you care about most. It’s a true game-changer for getting things done.
Why not give it a try? Head over to CustomGPT.ai to start a free trial and create your own private AI brain.
We can’t wait to see what you build with it!
P.S If are into MCP world, you will definitely live this MCP AMA that we did with Pinecone folks and other MCP folks (Santiago).
Find it here:
Further Reading
For those interested in exploring MCP in greater depth, check out the following resources:
- What Is Model Context Protocol (MCP)?
A comprehensive introduction to MCP’s origins, goals, and why it’s becoming the de facto “USB-C” for AI integrations. - Top 124 MCP Servers, MCP Clients & Tools (2025 Guide)
A curated catalog of production-ready MCP servers and clients—both open-source and proprietary—showcasing plug-and-play tools you can use today. - Problems MCP Model Context Protocol Solves
An exploration of the real-world pain points (API sprawl, ad-hoc integrations, security gaps) that MCP was designed to eliminate.
The Model Context Protocol (MCP) Architecture (2025)
A deep dive into MCP’s component layers, sequence diagrams, security considerations, and production-grade deployment patterns.
Frequently Asked Questions
Does CustomGPT.ai host the MCP server, or do I need to run my own infrastructure?
CustomGPT.ai positions this as a hosted MCP server integration, so the setup is intended to avoid running your own MCP server infrastructure. The page also states that server management and security are handled for you.
Can I ask questions in Cursor and use MCP to pull from my documentation?
Yes. The integration is described as connecting Cursor to a CustomGPT.ai project containing your documentation, so Cursor can answer project-specific questions without leaving the IDE.
Is using my own MCP server for internal tools different from this Cursor setup?
Yes. This setup is specifically about connecting Cursor to documentation stored in a CustomGPT.ai project via a hosted MCP server. If you want additional internal tools, that is a separate implementation choice beyond what is described here.
Why connect MCP to Cursor if I also need business-system actions later?
This integration is aimed at improving documentation lookup inside Cursor. If you also need operational actions in business systems, treat that as an additional integration layer, since action workflows are not the focus of this setup.
Can the same docs MCP connection work with Gemini, or is it Cursor-only?
MCP is described as a standard protocol for connecting tools to data sources, so compatibility depends on whether the client supports MCP connections. Cursor is the documented example on this page.
Do all uploaded file formats become available through MCP in Cursor?
The page confirms Cursor can query documentation in your CustomGPT.ai project, but it does not list file-format coverage details. A safe approach is to verify that your critical documents are correctly available in the project before relying on IDE answers.
How should teams handle strict enterprise or Azure-only environments for this integration?
The provided content does not include a detailed enterprise network architecture for Azure-only environments. It does emphasize managed server/security in the hosted model, so teams with strict requirements should validate policy fit with official security and implementation documentation before rollout.
Priyansh is Developer Relations Advocate who loves technology, writer about them, creates deeply researched content about them.