DrowAI

User guide

DrowAI user guide

Product documentation for installing DrowAI, setting up Runner services, and understanding the main workspaces. Use it as a practical reference before running the application or preparing a task workflow.

Setup

Standalone installation

Download MD

Run DrowAI Management and one Runner on the same Linux host.

Standalone runs PostgreSQL, Management, and one Runner on the same Linux host. Task execution still goes through the Runner; Management does not execute tools directly.

1. Download DrowAI

git clone https://github.com/quareth/drowAI.git
cd drowAI

If you received the repository as an archive, extract it and open a terminal in the extracted drowAI directory instead.

2. Start the stack

Run this command from the repository root:

docker compose --project-directory . \
  -f deploy/compose/standalone.yml \
  up -d --build

The first build can take several minutes. Docker creates the database volume, generated configuration and secret volumes, Management services, and Runner. The Runner waits until the setup wizard publishes its enrollment, then connects automatically.

Check startup status:

docker compose --project-directory . \
  -f deploy/compose/standalone.yml \
  ps

3. Complete first-run setup

Open the following address in a browser, replacing <server-address> with the Linux host name or IP address:

http://<server-address>

For a browser on the same machine, use http://localhost. Standalone and Distributed deployments both use HTTP port 80 by default.

On a fresh installation, DrowAI opens the setup wizard automatically. If the wizard does not open, go to:

http://<server-address>/setup

The wizard asks for:

  • The database password DrowAI should use after bootstrap.
  • The initial administrator username, email, and password.
  • The display timezone.
  • The name of the default Runner Site.

Configure an AI provider and model after signing in under Settings -> API.

4. Confirm Runner readiness

Open Settings -> Runner Sites. The default site should change to Ready and show one connected Runner. It can take a few seconds after setup completes.

You can also inspect the Runner directly:

docker logs drowai-runner --tail=200

A working Runner logs its registration and connection state. Create a small test task only after the Runner Site shows ready. Kali task containers will appear on this same host because the Standalone Runner is local to Management.

5. Stop or update Standalone

Stop the services without deleting data:

docker compose --project-directory . \
  -f deploy/compose/standalone.yml \
  down

Update an existing checkout and rebuild it while preserving data:

git pull
docker compose --project-directory . \
  -f deploy/compose/standalone.yml \
  up -d --build

Reset Standalone completely

This permanently deletes the DrowAI database, generated secrets, Runner credentials, task workspaces, and setup state. Use it only when you deliberately want a new first-run installation.

docker compose --project-directory . \
  -f deploy/compose/standalone.yml \
  down --volumes --remove-orphans
sudo rm -rf /var/lib/drowai

Start the stack again with the command from step 2. A broad docker system prune is not required and may remove resources belonging to other projects.

Image preview