Secure Datadog access for AI agents

Metrics, logs, traces, dashboards, incidents, monitors, and notebooks via Datadog's official MCP server.

Through Datadog, agents can query metrics, search logs and APM traces, read dashboards, incidents, monitors, and notebooks, and walk your service topology — every call runs through your policies and is logged for audit.

Server URL: https://mcp.datadoghq.com/v1/mcp

Setup

  1. Navigate to Resources and click Add Resource
  2. Select Datadog from the catalog
  3. Choose Per-org dynamic registration, the only mode Datadog offers — picking the card installs the resource with all tools pre-configured

Each user connects their own Datadog account via OAuth when they first use a Datadog tool. Navigate to Connections to manage linked accounts.

Dynamic client registration

This is the mode card labeled Per-org dynamic registration on the Datadog configure page. No app needs creating on Datadog's side and no client ID or secret to enter: the gateway registers an OAuth client with Datadog per organization at install time.

Available tools

All tools are read-only.

ToolDescription
aggregate_eventsAggregate events (counts, sums, percentiles) grouped by field/time
aggregate_rum_eventsAggregate RUM events (sessions, errors, load times)
aggregate_spansAggregate APM spans (request counts, durations, error rates)
analyze_datadog_logsRun SQL analysis against a virtual logs table
get_change_storiesGet change events (deploys, scaling, config) for an APM service
get_datadog_dashboardGet a dashboard's title, description, tags, and widgets
get_datadog_incidentGet incident details, timeline, and follow-ups
get_datadog_metricQuery metrics data (timeseries or scalar)
get_datadog_metric_contextGet a metric's metadata, tags, and related assets
get_datadog_notebookGet a notebook by ID
get_datadog_traceGet an APM trace's spans by trace ID
list_datadog_skillsList available Datadog skill guides
load_datadog_skillLoad a Datadog skill guide
search_datadog_dashboardsFind dashboards by title, widget, team, or author
search_datadog_eventsSearch raw events
search_datadog_hostsQuery the hosts inventory with SQL
search_datadog_incidentsSearch incidents by state, severity, team, or semantic query
search_datadog_logsSearch raw logs or log patterns
search_datadog_metricsList available metrics
search_datadog_monitorsFind monitors by title, status, team, or tag
search_datadog_notebooksFind notebooks by author, type, or content
search_datadog_rum_eventsSearch raw RUM events
search_datadog_service_dependenciesGet upstream/downstream service dependencies
search_datadog_servicesFind services by name or team
search_datadog_spansSearch raw APM spans

Policy examples

Rules are evaluated top to bottom and the first match wins. Allow rules on their own restrict nothing — your org starts with a seeded Allow all rule, so any tool your allows don't cover still falls through to it. Narrowing access needs a catch-all deny scoped to this resource, ordered below the allows. New rules are inserted at the top of the list, so create them in reverse order or drag them into place. See Policies.

  • Observability triage only. Allow search_datadog_logs, search_datadog_incidents, get_datadog_incident, and search_datadog_monitors, then deny * below them so no other Datadog tool falls through to Allow all.
  • Keep Datadog out of an agent's reach entirely. Deny * scoped to this resource and to that agent, above the seeded Allow all rule.

Since every Datadog tool is read-only, these examples limit what an agent can see rather than what it can change.

Next steps

  • Create a policy — start from the triage pattern in Policy examples above.
  • Policies — how rule order, scoping, and the seeded Allow all rule interact.
  • Connections — manage the Datadog accounts your users have linked.

On this page