Built to be extended

Pushes the device inventory into NetBox as devices and sites, so the switch you adopted this morning is in your source of truth by lunchtime. Runs dry by default, so you can see exactly what would change first.

Creates Zabbix hosts for the devices FreeSDN finds, so a new switch is monitored without being typed into a second system. Dry by default and never deletes.

The same idea for LibreNMS. Adds what FreeSDN already knows about, so a new device does not have to be entered twice.

Creates asset records for hardware found on the network, so warranty, cost and ownership are tracked from the day a device appears rather than whenever somebody remembers.

Publishes network state into Home Assistant as sensors, so the automations you already have can react to the network. Estate totals, availability and per-site health.

Opens a Jira issue when an alert fires and closes it when the alert clears. One issue per alert, not one per firing, so a flapping rule does not fill your queue overnight.

The same for ServiceNow, with an explicit severity-to-urgency mapping rather than a guess: ServiceNow derives priority from a matrix every instance tunes.

Pages on-call when an alert fires and resolves the incident when it clears. Deduplicated by PagerDuty itself, and it defaults to critical-only, because this one wakes people up.

Compares your sites against a declared baseline and reports what has drifted: sites missing a kind of device, settings that no longer match, configuration that moved without a ticket.

Raises an alert when a device appears that is not on your allowlist. Matches by MAC, by OUI vendor prefix, or by device type.

Finds the switch-port problems nothing else looks for: ports enabled with no link for weeks, links that negotiated half duplex, error counters climbing quietly.

The worked example. Lists offline devices, raises an alert, emits an event, stores settings and exposes a REST route, in one readable file. Start here when writing your own.