AI agents call get-alerts to retrieve information from Alertmanager without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves alert information from Alertmanager without modifying state. It has no side effects and is consistent with other read-only operations on the server (get-alert-details, get-alert-groups, get-silences). The blast radius of misuse is minimal—an attacker could only view alert information already accessible within the Alertmanager system.
From the tool's definition Tool name 'get-alerts' indicates retrieval of alert data. Server description states it 'enables Claude AI to interact with Prometheus Alertmanager for alert retrieval' and lists sibling tools including 'get-alert-details' and 'get-alert-groups' which clearly…
Documented attack patterns abuse exactly the kind of access get-alerts gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Alertmanager, and nothing reaches the server without passing your rules. This is the rule we recommend for get-alerts:
{
"version": "1",
"default": "deny",
"tools": {
"get-alerts": {}
}
}get-alerts is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
get-alerts. It is categorised as a Read tool in the Alertmanager MCP Server, which means it retrieves data without modifying state.
Register the Alertmanager MCP server in PolicyLayer and add a rule for get-alerts: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Alertmanager. Nothing to install.
get-alerts is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get-alerts rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get-alerts. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get-alerts is provided by the Alertmanager MCP server (kaznak/alertmanager-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Alertmanager, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
6 Alertmanager tools catalogued and risk-classified — across an index of 43,000+ MCP servers.