MetricsHub
MetricsHub Enterprise 2.1.00
-
Home
- Integrations
AI Agents Integration
You can configure AI-based assistants that support the MCP SSE transport, such as Claude[1] or OpenAI[2], to interact in real time with MetricsHub's internal knowledge and tools.
The following tools are currently available:
| Tool Name | Parameters | Description |
|---|---|---|
CheckHttpProtocol |
|
Checks if a host is reachable using the HTTP protocol. |
CheckIpmiProtocol |
|
Checks if a host is reachable using the IPMI protocol. |
CheckJdbcProtocol |
|
Checks if a host is reachable using the JDBC protocol. |
CheckJmxProtocol |
|
Checks if a host is reachable using the JMX protocol. |
CheckSnmpProtocol |
|
Checks if a host is reachable using the SNMP protocol. |
CheckSnmpV3Protocol |
|
Checks if a host is reachable using the SNMPv3 protocol. |
CheckSshProtocol |
|
Checks if a host is reachable using the SSH protocol. |
CheckWbemProtocol |
|
Checks if a host is reachable using the WBEM protocol. |
CheckWinrmProtocol |
|
Checks if a host is reachable using the WinRM protocol. |
CheckWmiProtocol |
|
Checks if a host is reachable using the WMI protocol. |
PingHost |
|
Checks if a host is reachable via ping and returns its response time and status. |
To get started, simply connect your AI assistant to the MetricsHub MCP Server using the Model Context Protocol (MCP)[3]. Once the connection is established, the available tools will be automatically published and ready for use.
Prerequisites
Before configuring the AI agents integration, make sure that:
- your AI assistant supports MCP SSE transport
- MetricsHub is installed and running
- you have network access to the machine where MetricsHub is running
- the
31888port is accessible from your machine or MCP client.
Important: Some MCP clients may require HTTPS with a valid TLS certificate. In such cases, you must enable HTTPS on your MetricsHub instance. One common approach is to place MetricsHub behind a reverse proxy (e.g., NGINX or Apache) with TLS termination, and ensure that the certificate is trusted by the client.
Configuring the integration
Configure your AI assistant to connect to http://<hostname>:31888/sse. Make sure to replace <hostname> with the actual hostname or IP address of the machine where MetricsHub is running.
Using the MCP tools
Once the connection is established, your AI assistant can call the PingHost tool as follows:
{
"method": "tools/call",
"params": {
"name": "PingHost",
"arguments": {
"arg0": "server-01"
}
}
}
The command will return this type of response:
{
"hostname": "server-01",
"responseTime": 27,
"reachable": true
}