Download the toolkit source code from https://github.com/abdmt-toolkit/abdmt/
Follow the instructions at https://github.com/abdmt-toolkit/abdmt/wiki/1.-Getting-Started
ABD-MT has three main modules: the Observers, the Ability Modeler, and the UI Adapter. Observers record input events and sensor data, currently for four types of abilities: touch, gesture, physical activity, and attention. Each Observer also provides ability information through calculation of human performance metrics. The Ability Modeler synthesizes captured interaction data and metrics from Observers to model a user's abilities, exporting methods for developers to inspect those abilities. For example, an app can query the Ability Modeler as to whether a user exhibits tremor, and whether this might be due to walking or running. Finally, the UI Adapter allows developers to manipulate UI widgets and layouts according to observed abilities and behaviors.
The full documentation can be found at the project wiki.
For a full list of Observer methods, check Observer Methods.
For a full list of Ability Modeler methods, check Ability Modeler Methods.
For a full list of UI Adapter methods, check UI Adapter Methods.