Cases Skills About

timesfm-forecasting - Zero-Shot Time Series Forecasting with TimesFM

Forecast univariate time series without custom training using TimesFM, with point predictions, calibrated quantile intervals, and optional exogenous covariates.

Tags

Updated: 2026-10-04

time-series forecasting zero-shot forecasting foundation model prediction intervals quantile forecasting exogenous covariates

Capabilities

Run system preflight checks Validate dataset memory fit Forecast univariate time series Generate quantile prediction intervals

Typical Inputs

Univariate time series Forecast horizon CSV, DataFrame, or array data

Typical Outputs

Point forecasts Quantile forecasts Prediction intervals

What this skill does

Run system preflight checks
Validate dataset memory fit
Forecast univariate time series
Generate quantile prediction intervals
Forecast with exogenous covariates
Process batched time series

Inputs

Univariate time series
Forecast horizon
CSV, DataFrame, or array data
Dynamic numerical covariates
Dynamic categorical covariates
Static categorical covariates
Forecast configuration

Outputs

Point forecasts
Quantile forecasts
Prediction intervals
System preflight status
Dataset memory estimate

Requirements

Python 3.10 or newer
TimesFM installation
PyTorch or JAX/Flax backend
Sufficient RAM and disk space
Optional XReg dependencies
First-use model download access

Source

Spec: SKILL.md

ClawIndex

OpenClaw Skills & Use Case Index

ClawIndex is an ecosystem-driven index of OpenClaw skills and real-world use cases.

Index

Meta

© 2026 ClawIndex All Rights Reserved.