Auto-populate the register from devices already on your network.

Most asset registers fail because nobody wants to type

The fastest way to kill an asset register is to ask IT to type every laptop into a form. The slow way to kill it is to ask them to maintain CSV imports forever. The discovery agent removes both jobs: it watches the network, enrols what it finds, and only asks a human to confirm the assets it can't match automatically.

Deploy it once with a single command (`curl ... | sh` for Unix, one-liner PowerShell for Windows). It heartbeats every 60 seconds, reports the host inventory, and quietly populates the discovery queue with anything new.

What the agent sees

• Hostname, primary MAC, manufacturer, model, serial number
• Operating system and version
• RAM, CPU, disk size, BIOS / firmware version
• Logged-in user (matched to employees by username)
• Active subnet and IP — used to scope discovery to specific VLANs

Matching, not duplication

Every discovered device runs through a match pipeline: exact serial match → exact MAC match → fuzzy hostname match. Confirmed matches update the existing asset's last-seen-on-network timestamp and any drifted fields. Unmatched devices land in a review queue — IT picks them off in batches, with one click to enrol or one click to ignore.

Microsoft Intune, Entra ID and Jamf integrations follow the same pipeline: each managed device is matched to an existing asset by serial first, then offered as a discovery if no match exists. You end up with one register, not three.

Subnet templates for multi-site

MSPs and multi-site teams use subnet templates to scope discovery — head office on 10.0.0.0/16, branch A on 10.1.0.0/16, customer X on 192.168.50.0/24. Each subnet has its own discovery cadence and its own auto-enrol rules, so a noisy guest VLAN doesn't pollute the corporate register.