See what you’re importing.
1 trigger · 18 actions · Read-only preview
Before you run it.
Populate huntressaccountkey and huntressdevicegroup before enabling the trigger. The automation validates them, but an empty field means no install.
Double check that huntressdevicegroup matches the intended Huntress organization, especially for MSPs running many tenants under one account key.
Keep the Tag applied
HUNTRESS trigger disabled while you test against a small pilot group, then enable it once the branches behave as expected.
Deliver Full Disk Access, System Extension approval, and Network Filter approval to Macs through MDM configuration profiles ahead of or alongside the install, or the agent will be limited to persistence telemetry.
Review the imported shell and script actions against your own environment before enabling. Installer URLs and vendor scripts change over time.
Decide where installation alerts route so a failed branch is visible to a human rather than sitting unread.
More about this resource.
What it does
This automation packages Huntress agent deployment for Windows, macOS, and Linux into one workflow. It ships with a Tag applied: HUNTRESS trigger, which is disabled in the source automation, so you can review the logic and enable the trigger when you are ready to let tagging drive deployment.
Every operating system branch follows the same four-step pattern, with conditions on the actions so only the branch matching the endpoint runs:
- Check if installed. A shell action looks for an existing Huntress agent.
- Exit if already installed. An exit action stops the run so healthy endpoints are left alone.
- Confirm custom fields. A shell action validates that the deployment values are present before anything is downloaded.
- Install and alert. Windows and Linux download the installation script and start the install process; macOS runs the "Huntress Deployment (Mac)" script through a run script action. Each branch ends with a create alert action reporting the installation.
A separate alert action, labeled "Alert on CF", is wired to the custom field checks so a missing value surfaces as an alert instead of a silent failure.
Required custom fields
Two custom fields must be populated before this automation will install anything:
huntressaccountkeyhuntressdevicegroup(the organization key)
Populate these at the level that matches how you scope Huntress organizations so each endpoint registers where it belongs.
macOS permissions
Per Huntress documentation, the macOS agent requires Full Disk Access (FDA), System Extension approval, and Network Filter approval to provide full EDR visibility, file collection, threat hunting, and endpoint isolation. Without FDA, the agent is largely limited to persistence telemetry. macOS blocks the System Extension and Network Filter from executing until they are approved. Plan to deliver these approvals through your MDM configuration profiles, since they are not something this automation grants. Huntress publishes a configuration wizard for macOS at support.huntress.io.
The install logic lives in Level automations and scripting, so you can read every action before you enable the trigger.
When to use it
- Onboarding a new client or site: populate the two Huntress custom fields, apply the HUNTRESS tag, and let the matching operating system branch handle the install.
- Backfilling coverage across a mixed Windows, macOS, and Linux fleet without maintaining three separate deployment procedures.
- Safely re-running deployment across a group, since each branch exits early when the Huntress agent is already installed.
- Catching configuration mistakes early, where the custom field check alerts on an endpoint whose account key or organization key was never filled in.
- Giving technicians a per-endpoint installation alert they can work from instead of manually confirming each machine in the Huntress portal.
The problem it solves
Rolling out an EDR agent across a mixed fleet usually turns into three separate processes: one PowerShell path for Windows, a shell script for macOS, and another for Linux. Each needs the correct account key and organization key injected, each needs a guard so you do not reinstall over a healthy agent, and each needs some signal back to the technician when it finishes. Doing that by hand, or with three disconnected scripts, is where endpoints quietly get missed or land in the wrong Huntress organization. macOS adds a second problem: the installer completing is not the same as the agent working. Without Full Disk Access, System Extension approval, and Network Filter approval, the Huntress macOS agent is largely limited to persistence telemetry, so a "successful" install can still leave you without full EDR visibility.