Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

v0.1.0 - 2026-08-20

Compare with v0.0.1

Features

  • Tools for hosting simulator engines as fastapi applications on beam

  • Tools for creating n8n tools for simulator engine applications

  • CLI for managine simulator applications and n8n tools

  • Support for soil file input to crop models

  • Utilities for iterating over pydantic fields that checks for annotations that should be skipped (used for n8n forms and CLI)

  • Ability to create ApsimX models piecewise from components

Added

  • Added flag to prevent hang on apsimx termination due to failed connection (27da4bf by Meagan Lang).

  • Added option to force reinstallation of simulator (0f6846c by Meagan Lang).

  • Added dockerfile and pixi tasks for building/running docker image containing server (4128bbf by Meagan Lang).

  • Added output_vars & timestep fields to model engine for use with server and update how the resume methods works to allow for a variable timestep (0cb714e by Meagan Lang).

  • Added soil file base class and ISRIC class for getting data from ISRIC SoilGrids (the REST API is not stable atm so this should not be the default soil data source) Begin matching data model to server Move data classes into data module and add a BaseDataFile class for more uniform treatment of conversion between data files and caching (bbac3d7 by Meagan Lang).

  • Added apsimx conda dependencies to recipe (caa6fa2 by Meagan Lang).

  • Added summary job to conda build GHA workflow for use in rulesets Set python_min in recipe context to allow building outside conda-forge Lint (a5509c1 by Meagan Lang).

  • Added docs on release process and added badges to README.md (0ce6982 by Meagan Lang).

  • Added CHANGELOG (79a894d by Meagan Lang).

v0.0.1 - 2026-08-07

Compare with first commit

Initial release migrating from ApsimXGym.

Features included

  • Support for ApsimX simulator engine including installation from GitHub

  • Classes for running simulator engines interactively

  • Classes for running simulator engines as gymnasium RL environments

  • Ability to generated LLM prompts for simulator based RL environments for AgriManager

  • CLI for running simulators and managing simulatr configuration

  • pixi environments and tasks

  • Sphinx based documentation with GHA workflow to automatically updated

  • Conda recipe with GHA workflow to build recipe

  • Tests for all classes with GHA workflow to run tests