kongqi404/Timeseries-MCP

3.2

If you are the rightful owner of Timeseries-MCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

MCP server supporting multiple time series analysis models

Timeseries-MCP

English|

MCP server supporting multiple time series analysis models

[!Note]

Under Development

Introduction

Timeseries-MCP is a server that supports multiple time series analysis models, including forecasting, anomaly detection, and classification. It is designed to be flexible and extensible, allowing users to easily add new models and functionalities.

Features

Support for multiple time series analysis models
Easy to use and extend
Built-in minio for data storage

Supported Models

Forecasting (WIP)

✅ NLinear
PatchTST
iTransformer
Autoformer
Informer

Anomaly Detection (WIP)

NLinear
PatchTST
iTransformer

Classification (WIP)

NLinear
PatchTST
iTransformer

Installation

Install from Source

environment requirements: uv,podman,git

podman run -p 9000:9000 -p 9001:9001 quay.io/minio/minio server /data --console-address ":9001"

#upload your dataset to minio

git clone https://github.com/kongqi404/Timeseries-MCP.git

cd Timeseries-MCP

# create .env file

uvx . http [--port PORT] [--host HOST] [--env_file FILE_PATH] # http
uvx . stdio [--env_file FILE_PATH] # stdio