Developer Guide

Working with OpenAdapt

Experiencing difficulty? Get support on our Discord.

Installation

OpenAdapt requires Python3.10, you can download it here or try our installer, which will download Python for you.

Poetry (Recommended)

First, clone & navigate to the repository:

git clone https://github.com/OpenAdaptAI/OpenAdapt.git
cd OpenAdapt

If poetry is not installed, you can use pip to install it:

pip3 install poetry

Lastly, run these lines to setup the environment:

poetry install
poetry shell
poetry run postinstall
cd openadapt && alembic upgrade head && cd ..

Git

Scripted Installation

Permissions(IMPORTANT)

See how to set up system permissions on macOS here.

Good to know: After installing OpenAdapt, run pytest to verify that the installation was successful.

Usage

Shell

Run this in every new terminal window once (while inside the OpenAdapt root directory) before running any openadapt commands below. This is required to activate the pythom environment:

poetry shell

You should see the something like this:

% poetry shell
Using python3.10 (3.10.13)
...
(openadapt-py3.10) %

Notice the environment prefix (openadapt-py3.10).

Run app locally for testing

python -m openadapt.entrypoint

CLI would look something like this after the command

After it runs successfully, you will see an app tray at top right of screen

Make your first recording

Create a new recording by running the following command:

To make your first recording, try the following command:

python -m openadapt.record "testing out openadapt"

Wait until all three event writers have started:

| INFO     | __mp_main__:write_events:230 - event_type='screen' starting
| INFO     | __mp_main__:write_events:230 - event_type='action' starting
| INFO     | __mp_main__:write_events:230 - event_type='window' starting

To stop recording, focus the terminal and send CTRL+C (SIGINT). The recording is finished saving when you see something like this:

| INFO     | __main__:record:637 - joining...
| INFO     | __mp_main__:write_events:221 - event_type='window' done
| INFO     | __mp_main__:write_events:221 - event_type='action' done
| INFO     | __main__:read_screen_events:356 - done
| INFO     | __main__:process_events:126 - done
| INFO     | __mp_main__:write_events:221 - event_type='screen' done
| INFO     | __main__:record:652 - saved recording_timestamp=1686243306.393652
| INFO     | __mp_main__:performance_stats_writer:433 - performance stats writer done

Current limitations:

• Recording should be short (i.e. under a minute), as they are somewhat memory intensive, and there is currently an open issue describing a possible memory leak.

• The only touchpad and trackpad gestures currently supported are pointing the cursor and left or right clicking, as described in this open issue

Visualize

• Quickly visualize the latest recording you created by running the following command:

python -m openadapt.visualize

This will generate an HTML file and open a tab in your browser that looks something like this:

• For a more powerful dashboard, run:

python -m openadapt.app.dashboard.run

• For a desktop app-based visualization, run:

python -m openadapt.app.visualize

Read more on openadapt.visualize.

This will open a scrollable window that looks something like this:

Playback

You can play back the recording using the following command:

python -m openadapt.replay NaiveReplayStrategy

Other replay strategies include:

• StatefulReplayStrategy: Early proof-of-concept which uses the OpenAI GPT-4 API with prompts constructed via OS-level window data.

• (*)VisualReplayStrategy: Uses Fast Segment Anything Model (FastSAM) to segment active window.

• (*)VanillaReplayStrategy: Assumes the model is capable of directly reasoning on states and actions accurately. With future frontier models, we hope that this script will suddenly work a lot better.

Good to know: The (*) prefix indicates strategies which accept an "instructions" parameter that is used to modify the recording. e.g.: python -m openadapt.replay VanillaReplayStrategy --instructions "calculate 9-8"

See strategy list for a complete list. More ReplayStrategies coming soon!

Features

System Tray Icon and Client GUI App(work-in-progress)

python3 -m openadapt.start

Detailed performance monitoring via pympler and tracemalloc: