Manage Datadog monitors. Actions: list, get, search, create, update, delete, mute, unmute, top, history, preview, test_notification. Filters: name, tags, groupStates (alert/warn/ok/no data). get/create/update return the full options object so callers can safely read-then-patch. create/update acce...
Part of the Datadog server.
Free to start. No card required.
AI agents may call monitors to permanently remove or destroy resources in Datadog. Without a policy, an autonomous agent could delete critical data in a loop with no way to undo the damage. PolicyLayer blocks destructive tools by default and requires explicit human approval before enabling them.
Without a policy, an AI agent could call monitors in a loop, permanently destroying resources in Datadog. There is no undo for destructive operations. PolicyLayer blocks this tool by default and only allows it when a human explicitly approves the action.
Destructive tools permanently remove data. Block by default. Only enable with explicit approval workflows.
{
"version": "1",
"default": "deny",
"hide": [
"monitors"
]
}See the full Datadog policy for all 23 tools.
These attack patterns abuse exactly the kind of access monitors gives an agent. Each links to the full case and the policy that stops it:
Other destructive tools across the catalogue. The same approach applies to each: deny by default, or require human approval.
Manage Datadog monitors. Actions: list, get, search, create, update, delete, mute, unmute, top, history, preview, test_notification. Filters: name, tags, groupStates (alert/warn/ok/no data). get/create/update return the full options object so callers can safely read-then-patch. create/update accept a config object validated against a typed schema covering the documented Datadog Monitor fields: - Top-level: name, type, query, message, tags, priority (1-5, nullable), restrictedRoles, multi, options. - options.* validated keys grouped by category: - notification: notifyNoData, noDataTimeframe, notifyAudit, notificationPresetName. - evaluation/delay: newHostDelay, newGroupDelay, evaluationDelay, requireFullWindow, onMissingData. - renotification: renotifyInterval (nullable), renotifyOccurrences, renotifyStatuses, escalationMessage. - lifecycle: timeoutH (nullable), includeTags, locked, silenced (record of timestamps/null), groupRetentionDuration. - thresholds: thresholds (critical/warning/ok/criticalRecovery/warningRecovery/unknown), thresholdWindows. - scheduling: schedulingOptions. Unknown keys (top-level or under options) are forwarded to Datadog as-is and surfaced via an optional warnings array on the response, so the schema does not lag the API. snake_case aliases are accepted on input and normalized to camelCase before validation. Validation errors short-circuit before any HTTP call and surface as. It is categorised as a Destructive tool in the Datadog MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Datadog MCP server in PolicyLayer and add a rule for monitors: 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 Datadog. Nothing to install.
monitors is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the monitors 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 monitors. 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.
monitors is provided by the Datadog MCP server (datadog-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 23 Datadog tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.