MCP servers that can list alerts
3 verified servers expose a tool that can list alerts
Alerts are the signals that something crossed a threshold, and an agent triaging system health reads them to see what is firing. Listing alerts returns what is currently raised, so the agent can summarize on-call status or decide what needs a response.
These verified servers let an agent list the alerts in a monitoring tool.
Grafana
Grafana Labs
Grafana Labs' official MCP server: query dashboards, Prometheus, Loki, incidents, alerts, and OnCall from your agent.
list_alert_groups
Grafana's list_alert_groups returns OnCall alert groups with filtering, the grouped view of what is firing for teams running Grafana OnCall.
SigNoz
SigNoz
SigNoz's official MCP server gives agents full access to traces, logs, metrics, dashboards, and alerts in your OpenTelemetry-native stack.
signoz_list_alerts
On SigNoz, signoz_list_alerts returns the alerts in its store, for teams using it as a single open-source observability backend.
MongoDB
MongoDB
MongoDB's official MCP server: query and manage databases, plus Atlas cluster administration.
atlas-list-alerts
atlas-list-alerts returns the alerts on a MongoDB Atlas deployment, surfacing database-level conditions an agent should know about.
What to know
An alert list sits one step before incidents: alerts fire on a condition, and some of them become incidents a team coordinates around. An agent reading the list sees what is active, a latency breach, a failing check, a database alert, and can group, summarize, or escalate from there. The tools differ in their backend: Grafana returns OnCall alert groups, SigNoz the alerts in its store, MongoDB Atlas the alerts on a deployment. The read is the same: surface what is firing so the agent does not have to poll each metric itself.
Across a triage session, remembering which alerts an agent already looked at keeps a second pass focused. An agent that re-reads the full alert list every run, with no memory of what it triaged or dismissed, re-processes the same noise. Holding which alerts it handled lets a follow-up attend to what is new.
Questions
- What is the difference between an alert and an incident?
- An alert fires when a condition crosses a threshold; an incident is the coordinated response some alerts become. Many alerts never escalate. Listing alerts shows what is currently firing, which is the input an agent uses to decide what, if anything, warrants declaring an incident.
- Does this cover database alerts too?
- Yes. MongoDB Atlas exposes alerts on a deployment, alongside the application-level alerts from Grafana OnCall and SigNoz. An agent triaging health reads across them to see both the service conditions and the database ones in one pass.