winedarksea/AutoTS

Automated Time Series Forecasting

About the project

AutoTS is a Python package that automatically searches forecasting models, data transformations, and ensembles for time-series datasets. It is used by developers and analysts to build forecasts across one or many series, including multivariate and probabilistic forecasts. The catalogue MCP server and instruction help coding agents use AutoTS in their workflows.

Latest release 1.0.4 · 20 Aug 2026

1 file for Codex and OpenCode: AutoTS AGENTS.md — 846 tokens loaded in every session.

1,434Stars on the repository
1Files it configures its agents with
846Tokens loaded in every session
2Agents configured

Instructions

These files are winedarksea/AutoTS's own configuration — they tell Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.