MCP server · Analytics

Disco Discovery Engine MCP server

Lets your AI scan a spreadsheet for hidden patterns and interactions, with p-values and citations.

Analytics For science and research Data Science Tools

Disco is a service that looks through a table of data for patterns you did not think to ask about. You give it a spreadsheet with one column you care about, and it finds combinations of other columns that shift that outcome. It is handy for analysts, researchers, and anyone staring at a dataset wondering where to start.

What is an MCP server? The 30-second version

On its own, your AI can only chat about what you paste into it. An MCP server is a small helper program that gives your AI a new skill or a connection to another service. This one connects your AI to Disco, so when you ask it to look for patterns in a data file, the AI can send that file to Disco and bring the findings back into your chat. You do not install anything on your computer for this; the connection runs over the web.

What this MCP server does

You ask your AI to find patterns in a data file you have. The AI passes your file and the column you want to understand to Disco through this helper. Disco runs its analysis in the cloud, which usually takes a few minutes. When it is done, the AI brings back a list of patterns, each with the exact conditions, an effect size, a p-value, and a note on whether the finding is new or already known. You can also ask for a link to a web report that shows everything visually.

Flow diagram: you ask your AI “Find patterns in patients.csv that affect readmission”, the Disco Discovery Engine MCP server works in steps: send your file, then disco scans it, then check the findings, and you get back patterns with p-values.

Click to zoom

What you can do with it

Upload a spreadsheet or data file to Disco without leaving your chat
Ask for patterns that shift a column you choose
Get p-values and effect sizes for each pattern found
See whether each pattern is new or already in the academic literature
Check how many credits a run will cost before starting it
Follow the progress of a run while it is processing
Open a web report with all patterns visualised

Try asking your AI

“Find patterns in my file patients.csv that affect the readmission column”
“Which combinations of columns in sales.xlsx are linked to a higher profit column?”
“Estimate the cost of running a private analysis on this 12 MB file”
“Show me the results of my last Disco run and summarise the novel patterns”

What it gives back to you

You get back a list of patterns, each described in plain words, with the exact ranges or categories involved, a p-value, an effect size, and a note on whether it is novel or already known. You also get a short written summary of the overall findings and a link to an interactive web report. Everything appears right in your chat, so you can ask follow-up questions about any pattern.

Before you start

What you need

A Disco account (you can sign up with just an email, no card needed)
An API key from disco.leap-labs.com/developers (a kind of password for apps; the site gives you one)
A data file in a flat table format like CSV, Excel, or Parquet, up to 5 GB

Good to know

Private runs cost credits and public runs are free but your data and results are published, so pick the visibility setting carefully before you start.

Install it with your AI

Add Disco Discovery Engine MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check Disco Discovery Engine MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.