{"slug": "skforecast-ai-agentic-time-series-forecasting-in-python", "title": "Skforecast-AI – Agentic time series forecasting in Python", "summary": "Skforecast-AI, a new Python package from the skforecast team, pairs a deterministic forecasting engine with an optional LLM reasoning layer to automate time series forecasting, returning both predictions and the runnable skforecast script that produced them. The package, requiring Python ≥ 3.10, profiles data, selects models, and evaluates performance in one call, supporting statistical and foundation models like Chronos-2, TimesFM, and Moirai.", "body_md": "| Package |\n|\n\n**skforecast-ai** is an **AI forecasting assistant** that pairs a deterministic engine, powered by [ skforecast](https://skforecast.org), with an\n\n**LLM reasoning layer**. Simply provide a time series, and the assistant automatically profiles the data, selects a model using established best practices, and evaluates its performance. It returns both the final forecast and the runnable skforecast script that produced it.\n\n[Why skforecast-ai?](#-why-skforecast-ai)[Installation](#-installation)[Quickstart (Python)](#-quickstart-python)[Quickstart (CLI)](#-quickstart-cli)[How it works](#-how-it-works)[Documentation](#-documentation)[Contributing](#-contributing)[Citation](#-citation)[License](#-license)\n\n- 🎯\n**Deterministic by design**: built as a strict rule-based engine to guarantee absolute consistency, same input always means the same output. - 🔍\n**Code you can inspect**: the script you see is the code that ran. Inspect it, version it, or run it standalone with plain** skforecast**. - ⚡\n**From data to forecast in one call**: automatic data profiling, model and estimator selection, lag/feature engineering, and backtest evaluation. - 💻\n**Python or terminal**: drive the full pipeline from a few lines of Python or from the command line. - 💬\n**LLM reasoning layer**: explains the engine's decisions in plain language, helps you improve the configuration, and lets you ask for advice. This layer is entirely optional; the core forecasting pipeline can run fully offline. - 🏗️\n**Built on skforecast**: recursive & direct forecasters, multi-series, statistical, and foundation models (Chronos-2, TimesFM, Moirai, and more).\n\nRequires Python ≥ 3.10.\n\n```\npip install skforecast-ai\n```\n\nTo enable the optional LLM reasoning layer:\n\n```\npip install \"skforecast-ai[llm]\"\n```\n\n## Install from source (for development)\n\n```\ngit clone https://github.com/skforecast/skforecast-ai.git\ncd skforecast-ai\npip install -e \".[dev]\"\n```\n\nFrom raw data to a validated forecast, and the code behind it, in a few lines:\n\n``` python\nimport pandas as pd\nfrom skforecast_ai import ForecastingAssistant\nfrom skforecast.datasets import load_demo_dataset\n\ndata = load_demo_dataset(verbose=False)\nassistant = ForecastingAssistant()\nresult = assistant.forecast(data=data, target=\"y\", steps=12)\n\nprint(result.predictions)   # forecast for the next 12 steps\nprint(result.metrics)       # evaluation metrics: MAE, MSE, MASE\nprint(result.code)          # the exact skforecast script that produced this result\n```\n\nThat single `forecast()`\n\ncall profiled the data, chose a forecaster and estimator, generated a `skforecast`\n\nscript, and executed it. `result.code`\n\nis the script that ran.\n\nThe returned `ForecastResult`\n\nexposes everything the pipeline produced:\n\n| Attribute | What it holds |\n|---|---|\n`result.predictions` |\nForecast for the requested horizon (includes interval columns when `interval` is requested) |\n`result.metrics` |\nBacktest evaluation metrics (MAE, MSE, MASE) |\n`result.code` |\nThe runnable `skforecast` script that produced the result |\n`result.profile` |\nWhat profiling detected about your data |\n`result.plan` |\nThe forecaster, estimator, lags, and metrics that were chosen |\n\n👉 New here? Walk through it step by step in ** Your first forecast**.\n\nThe same pipeline runs from the terminal. Point it at a CSV file or URL:\n\n```\n# End-to-end forecast (profile → plan → code → forecast)\nskforecast-ai forecast data.csv --target y --date-column datetime --steps 12\n\n# Just inspect the data\nskforecast-ai profile data.csv --target y --date-column datetime\n\n# Generate a standalone, runnable script without executing it\nskforecast-ai forecast-code data.csv --target y --date-column datetime --steps 12 --output forecast.py\n```\n\nRun `skforecast-ai --help`\n\nor `skforecast-ai <command> --help`\n\nfor inline documentation on any command.\n\n👉 Full command reference in ** CLI usage**.\n\n**skforecast-ai** supports two distinct workflows using the same underlying forecasting engine:\n\n-\n**The Fast Path:** Use this when you want a forecast or backtest result in a single call. The assistant profiles the data, builds the modeling plan, executes the workflow, and returns the results alongside the reproducible`skforecast`\n\ncode. -\n**The Step-by-Step Path:** Use this when you want granular control to inspect or adjust intermediate decisions. You can manually create a profile, build a plan, optionally refine it with the LLM, define a validation strategy, evaluate the model, and then generate the forecast.\n\nA useful mental model is that forecasting and validation are separate branches. Once you have a `profile`\n\nand a `plan`\n\n, you can use `forecast()`\n\nto produce future predictions directly, or `backtest()`\n\nto evaluate the model's performance on historical data. You can also use `compare()`\n\nto evaluate several candidate configurations under the same cross-validation strategy and obtain a ranked leaderboard, so the best configuration is chosen from measured performance rather than intuition.\n\nThe `ask()`\n\nmethod is available in both workflows. It can explain a profile, plan, validation setup, backtest result, comparison result, or answer general forecasting questions, but it will never execute the workflow or modify your parameters without explicit instruction.\n\nRead more in ** Introduction to agentic forecasting**.\n\nExplore the full capabilities of **skforecast-ai** with our comprehensive documentation:\n\n| Documentation | |\n|---|---|\n| 🚀\n|\n\n[Introduction to agentic forecasting](https://ai.skforecast.org/stable/user-guides/agentic-forecasting.html)[API Reference](https://ai.skforecast.org/stable/api/assistant.html)[Releases](https://ai.skforecast.org/stable/releases/releases.html)[More](https://ai.skforecast.org/stable/more/about-skforecast-ai.html)Contributions are welcome, whether it's a bug report, a feature idea, or a pull request. Please see the [Contributing Guide](/skforecast/skforecast-ai/blob/main/CONTRIBUTING.md) and our [Code of Conduct](/skforecast/skforecast-ai/blob/main/CODE_OF_CONDUCT.md) to get started.\n\nIf you use `skforecast-ai`\n\nin your work, please cite the underlying `skforecast`\n\nlibrary:\n\n**Zenodo**\n\n```\nAmat Rodrigo, Joaquin, & Escobar Ortiz, Javier. (2026). skforecast-ai (Version 0.2.0). Zenodo. https://doi.org/10.5281/zenodo.21338159\n```\n\n**APA**\n\n```\nAmat Rodrigo, J., & Escobar Ortiz, J. (2026). skforecast-ai (Version 0.2.0) [Computer software]. https://doi.org/10.5281/zenodo.21338159\n```\n\n**BibTeX**\n\n```\n@software{skforecast-ai,\n  author  = {Amat Rodrigo, Joaquin and Escobar Ortiz, Javier},\n  title   = {skforecast-ai},\n  version = {0.2.0},\n  month   = {8},\n  year    = {2026},\n  license = {Apache-2.0},\n  url     = {https://ai.skforecast.org/},\n  doi     = {10.5281/zenodo.21338159}\n}\n```\n\nView the [citation file](https://github.com/skforecast/skforecast-ai/blob/main/CITATION.cff).\n\nLicensed under the Apache License 2.0 (see [LICENSE](https://github.com/skforecast/skforecast-ai/blob/main/LICENSE) for details).\n\nBuilt with ❤️ on top of [skforecast](https://skforecast.org).", "url": "https://wpnews.pro/news/skforecast-ai-agentic-time-series-forecasting-in-python", "canonical_source": "https://github.com/skforecast/skforecast-ai", "published_at": "2026-09-01 16:10:31+00:00", "updated_at": "2026-09-01 16:23:16.577236+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-tools", "machine-learning"], "entities": ["skforecast", "skforecast-ai", "Chronos-2", "TimesFM", "Moirai"], "alternates": {"html": "https://wpnews.pro/news/skforecast-ai-agentic-time-series-forecasting-in-python", "markdown": "https://wpnews.pro/news/skforecast-ai-agentic-time-series-forecasting-in-python.md", "text": "https://wpnews.pro/news/skforecast-ai-agentic-time-series-forecasting-in-python.txt", "jsonld": "https://wpnews.pro/news/skforecast-ai-agentic-time-series-forecasting-in-python.jsonld"}}