Soar

This is the code base for Soar, a cognitive architecture for developing systems that exhibit intelligent behavior. For more information about this project, please visit: The Soar home page The GitHub project page Note that the readme included with the Soar distribution for end-users is in the Release-Support repository. For binary builds of Soar you can get them in two places: Official Releases — multiplatform zip including bin/win x86-64 , bin/linux x86-64 , bin/mac x86-64 , and bin/mac ARM64 (Apple Silicon). Latest development builds: the CMake-based release pipeline ( cmake-multi-platform.yml ) produces per-platform artifacts including ARM64 macOS. The SCons CI ( build.yml ) is also kept current but does not produce an ARM64 macOS artifact — use cmake-multi-platform.yml or build from source on Apple Silicon. Some performance statistics are calculated automatically using the Factorization Stress Tests. You can see performance on a commit-by-commit basis in Performance.md. Disclaimer: These are worst case tests. Average performance is probably much higher. In addition, these show that even in worst case, Soar beats its goal of 50 msec reactivity (in these tests, the max is ~30msec per decision). Soar supports two build systems: CMake (see build with CMake) and scons (see build with scons). Both are kept in sync and produce binary-compatible artifacts that land in the same out/ and multi-platform install layout. The release zip is built by the CMake-based cmake-multi-platform.yml workflow. The following table compares supported build features for Soar between the two build systems as of 9.6.5. The instructions below are cursory and may be out of date; the most up-to-date instructions for compiling Soar from source will always be the CI build scripts. The CMake-based release pipeline is cmake-multi-platform.yml ; the SCons CI lives in build.yml . The project supports generating compile commands.json, which can be used by e.g. VSCode with the C/C++ plugin to provide IntelliSense. To generate this file, run scons with the cdb target: Note for M-series Mac users: you'll want to make sure you're compiling for ARM64, not x86 64. Sometimes users have Python installed in compatibility mode, leading to compiles for the wrong architecture. You can check which architecture your Python is built for using this: You can also check your clang 's default compile target using clang --version . Debug mode enables debugging, but also activates assertions, which are important for catching bugs early. --scu (single compilation unit) simplifies the debugging experience. The following prerequisites must be available: CMake ( = 3.21) Python 3, including pip , for the Conan package manager ( pip install conan ). A C/C++ toolchain (Visual Studio 2022 / Xcode command-line tools / GCC or Clang). For the Java debugger and SWIG-Java bindings: a JDK 11 or newer (Temurin recommended). For SWIG bindings: SWIG (Windows users can install via choco install swig ). Then pick a preset that matches what you want to build. The most common presets for everyday development and CI are: compile commands.json is generated automatically and picked up by VS Code's C/C++ extension, clangd, and other tools without further configuration. The VS Code CMake Tools extension integrates with these presets directly. See CMakePresets.json for the underlying definitions and build.sh / build.bat for one-shot wrapper scripts.