/site-overview

description: Overview of a Domotz site's network health
argument-hint: "<agent_id>"
arguments: [agent_id]

Site Overview

Generate a health overview for a Domotz-monitored site from collector status, device state, and network faults.

The overview is built from **device state**, not from alerts. Domotz's MCP surface exposes alert *configuration* only — there is no way to query fired alerts — so do not present this as an alert report.

Prerequisites

• Domotz MCP server connected with valid API credentials
• MCP tools `domotz_agents_get`, `domotz_devices_list`, `domotz_network_ip_conflicts`, and `domotz_devices_history` available

Steps

1. **Get agent details**

Call `domotz_agents_get` with the `agent_id` to get site name, status, license info, and last seen time. **If the collector is OFFLINE, stop and say so.** Everything below would be last-known data reported as if it were current, which is worse than no report.

2. **Get device summary**

Call `domotz_devices_list` for the agent. The full census arrives in one response. Aggregate:

• Total devices
• Online vs offline count
• Devices by type
• Top vendors

3. **Check for network faults**

Call `domotz_network_ip_conflicts` for the agent. Addressing collisions present as intermittent device faults and are easily misread as failing hardware.

4. **Qualify the offline devices**

For offline devices that matter, call `domotz_devices_history` to distinguish a clean outage from a device that has been flapping. A count of offline devices without this is not actionable.

5. **Build health report**

Present a structured overview:

• **Site Info** - Name, agent status, last seen, license utilization
• **Device Health** - Total devices, online/offline breakdown, notable offline infrastructure
• **Network Faults** - IP conflicts, if any
• **Overall Assessment** - Healthy / Warning / Critical, with the evidence behind it

State explicitly that this reflects device state observed by the collector, not what Domotz alerted on.

6. **Recommend actions**

Flag any issues needing attention: offline infrastructure, flapping devices, IP conflicts, or license capacity concerns.

7. **Optionally check monitoring coverage**

Call `domotz_alerts_profiles_list` and, for key devices, `domotz_alerts_device_list` to report which devices would actually generate a notification if they failed. Devices with no binding are silent on failure — usually the most useful finding in this report.

Parameters

| Parameter | Type | Required | Description | |-----------|------|----------|-------------| | agent_id | integer | Yes | The agent/site to review |

Examples

Get Site Overview

/site-overview --agent_id "12345"

Error Handling

• **Agent Not Found:** Verify the agent ID; call `domotz_agents_list` to find the correct one
• **Agent Offline:** Report that the collector is offline and that no site health assessment can be made from stale records
• **Asked for active alerts:** There is no fired-alert surface on this server. Say so and offer the coverage view in step 7 instead of synthesising alerts from device status
• **Authentication Error:** Verify API credentials and region

Related Commands

• `/device-inventory` - Detailed device list for the site
• `/device-lookup` - Find a specific device at the site