Happy slice of mcp pizzaMCP.pizza

AutoTS

MCP.Pizza Chef: winedarksea

This tool automatically tests dozens of forecasting methods against your numbers and keeps whichever one predicts most accurately, whether you're projecting sales, website traffic, inventory needs, or any other figure that changes over time. It runs as Python code rather than through a chat app, so someone comfortable installing Python packages sets it up, and it can then produce forecasts for hundreds of related items — like every store in a chain — in a single run. No account or paid access is required to get started.

Data
Spreadsheets

Use This MCP server To

Predict next month's sales from past sales data Forecast website traffic for the weeks ahead Estimate future inventory needs before you run out Spot upcoming demand spikes before they happen Compare several forecasting methods to find the most accurate one Get a best-case and worst-case range for a forecast, not just one number

README

AutoTS

AutoTS is a time series package for Python designed for rapidly deploying high-accuracy forecasts at scale. Give it a try in your browser with the official demo app.

In 2023, AutoTS won in the M6 forecasting competition, delivering the highest performance investment decisions across 12 months of stock market forecasting.

There are dozens of forecasting models usable in the sklearn style of .fit() and .predict(). These includes naive, statistical, machine learning, and deep learning models. Additionally, there are over 30 time series specific transforms usable in the sklearn style of .fit(), .transform() and .inverse_transform(). All of these function directly on Pandas Dataframes, without the need for conversion to proprietary objects.

All models support forecasting multivariate (multiple time series) outputs and also support probabilistic (upper/lower bound) forecasts. Most models can readily scale to tens and even hundreds of thousands of input series. Many models also support passing in user-defined exogenous regressors.

These models are all designed for integration in an AutoML feature search which automatically finds the best models, preprocessing, and ensembling for a given dataset through genetic algorithms.

Horizontal and mosaic style ensembles are the flagship ensembling types, allowing each series to receive the most accurate possible models while still maintaining scalability.

A combination of metrics and cross-validation options, the ability to apply subsets and weighting, regressor generation tools, simulation forecasting mode, event risk forecasting, live datasets, template import and export, plotting, and a collection of data shaping parameters round out the available feature set.

Table of Contents

Installation

pip install autots

This includes dependencies for basic models, but additonal packages are required for some models and methods.

Be advised there are several other projects that have chosen similar names, so make sure you are on the right AutoTS code, papers, and documentation.

Basic Use

Input data for AutoTS is expected to come in either a long or a wide format:

AutoTS FAQ

Can I use this to predict my monthly sales?
Yes — feed it your past sales numbers and it tests many forecasting methods to find the one that predicts your future sales most accurately.
Can I use this to forecast demand for many products at once?
Yes — it can generate separate forecasts for hundreds or thousands of items in one run, so it scales well if you're tracking many products, stores, or accounts.
Do I need an account or API key to use this?
No account or API key is required for the forecasting itself; you would only need outside credentials if you choose to pull in optional external data.
Does this work inside Claude Desktop, Cursor, or ChatGPT?
The documentation doesn't describe a direct connection to chat apps like Claude Desktop, Cursor, or ChatGPT — it runs as Python code that a developer sets up, and the resulting forecasts can then be shared with anyone on the team.
How hard is it to set up?
It requires installing Python and running a short install command, so it's best handled by someone with basic coding experience rather than being a one-click setup.
What kind of data do I need to provide?
A simple table with dates and the numbers you want to predict, such as daily sales totals, weekly page views, or monthly inventory counts.
Can it tell me how confident it is in a forecast?
Yes — alongside its main prediction, it can give you an upper and lower range so you know how much the actual result might vary.
Will it work with data made up of many related time series, like sales by region?
Yes — it's built to handle many related series at once and can share patterns across them to improve accuracy.