WrkFlo.Biz

You can perform zero-shot forecasting on any univariate time series — sales, sensor readings, stock prices, energy demand, patient vitals, weather, or scientific measurements — without training a custom model. It also supports advanced covariate forecasting (XReg) with dynamic and static exogenous variables, returning point forecasts with calibrated prediction intervals.

It targets data scientists, analysts, and engineers who need forecasts from CSV, DataFrame, or array inputs without building or training models. It suits anyone forecasting business metrics, IoT sensor data, financial series, or clinical vitals using Google's TimesFM foundation model.

First run the included preflight system checker script, which must be executed before initial use to verify your machine can load the model and handle your dataset. The skill then checks RAM/GPU, validates dataset fit, accepts CSV/DataFrame/array inputs, and returns forecasts with prediction intervals.

Yes. The skill is released under the Apache-2.0 license, permitting free use, modification, and distribution. It was authored by Clayton Young (@borealBytes) and is currently at version 1.0.0.

You need a machine with sufficient RAM or GPU capacity to load Google's TimesFM foundation model, verified by the mandatory preflight system checker before first use. Input data must be a univariate time series supplied as CSV, DataFrame, or array, optionally with exogenous covariates for XReg forecasting.