PRTG Network Monitor MCP server

What the model can do

PRTG Network Monitor on your own server: sensors that are down, warning or unusual (or any status) with their message and last value, devices with sensor counts per state, the log of state changes and events, tickets, one sensor's channels and its historic data; when you ask, can acknowledge an alarm, pause an object (indefinitely or for some minutes) and resume it, scan a sensor now, run an auto-discovery, rename an object, set its priority or tags, duplicate one, delete one, send a test notification, and add an object to a report.

Reads

• prtg_list_sensorsPRTG sensors with status, message, last value, priority, tags, last check and how long down or up.
• prtg_list_devicesPRTG devices with status, message, priority, tags, and how many of their sensors are up, down, warning and paused.
• prtg_list_logThe PRTG log, newest first: state changes, acknowledgements, pauses and system events, each with time, object, parent, type, status and message.
• prtg_list_ticketsPRTG tickets (to-dos, notification tickets and user tickets): id, last change, subject, status (open, resolved, closed), priority and type.
• prtg_sensor_channelsOne PRTG sensor's channels with their last values (e.g. a ping sensor's ping time and packet loss, a traffic sensor's in and out).
• prtg_historic_dataHistoric values for one PRTG sensor between two dates (server local time), raw or averaged over avg_seconds.

Acts, when you ask

These change something in PRTG Network Monitor. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• prtg_acknowledge_alarmAcknowledges the Down alarm on one PRTG sensor with your message, so it shows as Down (Acknowledged) and stops repeating its notifications.
• prtg_pausePauses monitoring of one PRTG object (probe, group, device, sensor, notification or user account) and everything under it, which stops its alerts.
• prtg_resumeResumes monitoring of one paused PRTG object.
• prtg_scan_nowAsks PRTG to scan one sensor (or every sensor under an object) now instead of waiting for its interval.
• prtg_run_auto_discoveryRuns an auto-discovery on one PRTG group or device, optionally with a device template file (e.g. "Linux SNMP.odt").
• prtg_renameRenames one PRTG object (group, device, sensor, map, report, library, notification template, schedule, user or user group).
• prtg_set_prioritySets one PRTG object's priority from 1 (lowest) to 5 (highest), which orders lists and can drive notifications.
• prtg_set_tagsReplaces one PRTG object's own tags with the list given (inherited tags are unaffected).
• prtg_duplicate_objectCopies one PRTG group, device or sensor into a target group or device under a new name (a device copy takes a new host).
• prtg_delete_objectDeletes one PRTG object and everything under it at once (deleting a group deletes its devices and sensors, with their history).
• prtg_test_notificationTriggers one PRTG notification template now as a test, which sends its real email, SMS, push or HTTP action to the people and systems it names.
• prtg_add_to_reportAdds one PRTG group, device or sensor to a report, so it appears in that report from its next run.

Connect it

First connect PRTG Network Monitor on Dashboard → Connections, then add Memtro to your client once. The PRTG Network Monitor tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared PRTG Network Monitor connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Which sensors are down right now?
• › Which devices have sensors in warning?
• › What was the ping time on the core switch yesterday?
• › What changed state in the last hour?
• › Which tickets are still open?

Questions

Is this an official PRTG Network Monitor MCP server?

No. It is Memtro's MCP server with Memtro's PRTG Network Monitor connector behind it: the connector reads PRTG Network Monitor through its API with the credentials you connect on Dashboard → Connections, and PRTG Network Monitor publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in PRTG Network Monitor?

Only when you ask, and only what it has been granted: acknowledge a Down alarm on a sensor; pause monitoring of an object (indefinitely or for some minutes); resume monitoring of a paused object; scan a sensor now; run an auto-discovery on a group or device (may add sensors); rename an object; set an object's priority; set an object's tags; duplicate a group, device or sensor; delete an object and everything under it; send a test of a notification template to its recipients; add an object to a report. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server