Inventory Policies ​

Inventory policies decide what an agent gathers on its scheduled inventory sweep — software, services, certificates, and so on — and how often. They are the long-running, low-overhead counterpart to interactive assessments: every match agent runs the policy on its interval and ships the resulting facts back to InfraScout.

Inventory Policies tab with the system-owned Default policy listing priority, interval, on-startup behavior, and the categories collected

Layout ​

The page header carries the section title and the Admin Access chip; the tab strip puts you on Inventory Policies. Two header buttons sit above the list: New Policy and Refresh. A free-text Search box filters by name or description.

The body is a list of policy cards sorted by priority. A System chip marks the system-owned default; it can be edited only through this UI.

What a policy controls ​

Every policy has the same four levers:

  • Priority — lower numbers win. The system default is priority 0.
  • Interval — how often the agent runs the sweep. The default is 6 hours; common values range from 1 hour (frequent change tracking) to 24 hours (low-overhead).
  • On startup — whether to run a sweep the moment the agent comes online. The default is No; flip to Yes if you want fresh inventory immediately after a reboot.
  • Categories — which inventory categories to collect. The system default collects system, software, services, and certificates; new policies typically narrow this set.

The example screenshot shows the system-owned Default policy: priority 0, every 6 hours, no startup run, collecting system / software / services / certificates.

How policies match ​

Like update policies, inventory policies target agent groups and the lowest-priority match wins. Each agent runs exactly one inventory policy at a time. To carve out a specific behavior — say, hourly certificate-only sweeps on the PKI hosts — create a new policy with a lower priority number, target the relevant group, and tick only the certificates category.

What the data is used for ​

The collected facts feed three things: the Inventory tools that AI playbooks query, the change-tracking timeline on each agent detail page, and the search index that powers fleet-wide queries like "every host with OpenSSL older than 3.x".

Inventory data is stored on the InfraScout server, not on the agent. Reducing the category set or extending the interval is the right lever when you need to lower the storage or network footprint.