TL;DR

Connect your existing database + auth provider and launch an MCP server for your customers in a few minutes.

Watch me ship an MCP server and connect it to ChatGPT in 2.5 minutes

The Problem

A few days ago OpenAI announced Apps for ChatGPT — a new generation of apps you can chat with, right inside ChatGPT.

An MCP server is the foundation of every Apps SDK integration. It exposes tools that the model can call, enforces authentication, and packages the structured data plus component HTML that the ChatGPT client renders inline.

You can build an MCP server in one of two ways:

A)

Load all of your API endpoints into the MCP server as tools and have the client decide which aggregation of tools it should use to fulfil user requests.

This approach works well if you already have a highly composable API designed for data analysis — concretely the API expresses any combination of filters, sorts, and transformations in a structured way.

However, most SaaS APIs are designed for record management, not data analysis or flexible retrieval. These APIs break down when presented with the wide array of potential user queries which come with a natural language input.

B)

Build an AI agent attached to your database which takes in natural language and dynamically generates queries based on the input  — then attach the agent to your MCP as a tool.

This is hard to build because you need to handle conversation state, schema introspection, agent observability, multi-tenancy and query safety.

Luckily, with Inconvo we offer AI data agents as a service so you don’t need to worry about any of that.

Overall option B outperforms option A in power and flexibility, while eliminating the need for new API endpoints as user input cases grow.

Our Solution

With Inconvo you can build a powerful AI agent for data analysis on top of your database and make it available to your users in your own app or the AI apps they already use like ChatGPT.

How it works

1. Connect your existing database

2. Link your existing authentication and set user data permissions scopes

1. Launch a remote authenticated MCP server in a few clicks, which can be linked easily by your users to ChatGPT etc.

Inconvo handles the Details

We give you agent traces and logs for out of the box observability

And a semantic layer so you can improve the data agents performance through context engineering without needing to write any extra code.

Ready to ship MCP servers for your customers?

You can try inconvo today for free by visiting https://inconvo.com and creating an account.

Or book a call: https://inconvo.com/demo