Datadog MCP server
Datadog's official remote MCP server lets agents search logs, query metrics, pull APM traces, inspect monitors, and investigate incidents.
The Datadog MCP Server is the official remote server from Datadog that connects an AI agent to your observability data so it can investigate problems the way an on-call engineer would. The default core toolset covers the signals you reach for first: searching and analyzing logs, querying metrics and their metadata, pulling full APM traces and spans, listing hosts and services with their dependencies, searching RUM events, browsing dashboards and notebooks, and reading and triaging monitors and incidents. Beyond core, the server is organized into opt-in toolsets you enable with a toolsets query parameter, spanning alerting and SLOs, APM and Watchdog, dashboards and widgets, database monitoring, DDSQL, error tracking, feature flags, Kubernetes, networks, cases, security signals and findings, software delivery and CI, synthetics, workflows, and guided onboarding.
The canonical deployment is the hosted remote endpoint, which for the US1 site is https://mcp.datadoghq.com/api/unstable/mcp-server/mcp (swap the domain for your region, e.g. mcp.datadoghq.eu). It authenticates with OAuth 2.0 in interactive clients, and falls back to a scoped DD_API_KEY plus DD_APPLICATION_KEY pair sent as HTTP headers for headless agents. Datadog recommends a service-account key scoped to only the permissions an agent needs, since this server can read across your entire telemetry estate.
Quick install
Copy-paste configs are provided for all 8 supported clients. Pick your client below.
Add to ~/.claude.json
{
"mcpServers": {
"datadog": {
"type": "http",
"url": "https://mcp.datadoghq.com/api/unstable/mcp-server/mcp"
}
}
}claude mcp add --transport http datadog https://mcp.datadoghq.com/api/unstable/mcp-server/mcpHeads up
- First tool call opens a browser to authorize.
Available tools
| Tool | Description |
|---|---|
| search_datadog_logs | Searches logs with time, service, and query-string filters. |
| analyze_datadog_logs | Performs statistical analysis over logs using SQL-style queries. |
| get_datadog_metric | Queries historical and real-time metric data. |
| get_datadog_metric_context | Retrieves metric metadata, tags, and available tag values. |
| search_datadog_metrics | Lists available metrics with filtering. |
| get_datadog_trace | Fetches a complete APM trace by trace ID. |
| search_datadog_spans | Retrieves APM spans with filters. |
| search_datadog_hosts | Lists monitored hosts with filtering options. |
| search_datadog_services | Lists services in the Service Catalog. |
| search_datadog_service_dependencies | Shows upstream and downstream service relationships. |
| search_datadog_rum_events | Searches Real User Monitoring events with advanced query syntax. |
| search_datadog_monitors | Finds monitors by status, threshold, and alert conditions. |
| search_datadog_incidents | Retrieves incidents filtered by state, severity, and metadata. |
| get_datadog_incident | Retrieves detailed information about a single incident. |
| search_datadog_events | Searches monitor alerts, deployments, infrastructure changes, and service status events. |
| search_datadog_dashboards | Lists available dashboards with key details. |
| get_datadog_notebook | Retrieves a notebook by ID. |
| search_datadog_notebooks | Searches notebooks by author, tags, and content. |
| create_datadog_notebook | Creates a new notebook. |
| edit_datadog_notebook | Edits the contents of an existing notebook. |
| validate_datadog_monitor | Validates a monitor definition before creation (alerting toolset). |
| get_datadog_monitor_templates | Retrieves available monitor templates (alerting toolset). |
| search_datadog_monitor_groups | Searches monitor groups by name or criteria (alerting toolset). |
| search_datadog_slos | Finds SLOs by name, tags, or service (alerting toolset). |
| create_datadog_monitor | Creates a monitor in draft mode (alerting toolset). |
| get_monitor_coverage | Identifies monitoring gaps for services and hosts (alerting toolset). |
| apm_search_spans | Searches spans using APM query syntax (APM toolset). |
| apm_query_trace | Filters and aggregates spans within traces (APM toolset). |
| apm_discover_span_tags | Discovers available span tag keys (APM toolset). |
| apm_get_primary_tag_keys | Retrieves the organization's primary tag keys (APM toolset). |
| apm_search_watchdog_stories | Finds Watchdog anomaly stories for services (APM toolset). |
| apm_get_watchdog_story | Retrieves details of a specific Watchdog story (APM toolset). |
| apm_latency_bottleneck_summary | Analyzes latency bottlenecks across traces (APM toolset). |
| get_change_stories | Retrieves deployment and change events for APM services (APM toolset). |
| semantic_search_change_stories | Finds related change events using natural language (APM toolset). |
| apm_search_recommendations | Searches for APM optimization recommendations (APM toolset). |
| apm_get_recommendation | Retrieves the full details of a specific APM recommendation (APM toolset). |
| search_datadog_cases | Searches cases by status, priority, and project (cases toolset). |
| get_datadog_case | Retrieves case details including its timeline (cases toolset). |
| create_datadog_case | Creates a new case with a title and project (cases toolset). |
| update_datadog_case | Updates case fields such as status and priority (cases toolset). |
| add_comment_to_datadog_case | Adds a markdown-formatted comment to a case (cases toolset). |
| link_jira_issue_to_datadog_case | Links a Jira issue to a case (cases toolset). |
| list_datadog_case_projects | Lists available case projects (cases toolset). |
| get_datadog_case_project | Retrieves details for a specific case project (cases toolset). |
| search_datadog_users | Searches users by email, name, or handle (cases toolset). |
| execute_code | Runs agent-authored TypeScript in a sandbox with the dd.* namespace for multi-signal investigations (code execution toolset). |
| get_datadog_dashboard | Retrieves a dashboard by ID with its widgets (dashboards toolset). |
| upsert_datadog_dashboard | Creates or updates a dashboard (dashboards toolset). |
| delete_datadog_dashboard | Permanently deletes a dashboard (dashboards toolset). |
| get_widget_reference | Returns widget schemas and building instructions (dashboards toolset). |
| validate_dashboard_widget | Validates a widget definition (dashboards toolset). |
| ask_widget_expert | Answers questions about widget configuration (dashboards toolset). |
| find_datadog_database_instances | Discovers and ranks database instances (database monitoring toolset). |
| get_datadog_database_calling_services | Identifies services that call a database's queries (database monitoring toolset). |
| get_datadog_database_explain_plans | Retrieves PostgreSQL execution plans (database monitoring toolset). |
| get_datadog_database_health_signals | Runs health checks for potential database issues (database monitoring toolset). |
| get_datadog_database_query_performance | Analyzes query performance metrics (database monitoring toolset). |
| get_datadog_database_query_statement | Retrieves the SQL statement text for a query (database monitoring toolset). |
| get_datadog_database_recommendations | Surfaces live database recommendations (database monitoring toolset). |
| get_datadog_database_schemas | Fetches database schema definitions (database monitoring toolset). |
| optimize_datadog_database_query | Analyzes a query for optimization opportunities (database monitoring toolset). |
| search_datadog_database_plans | Searches query execution plans (database monitoring toolset). |
| search_datadog_database_samples | Searches individual query executions (database monitoring toolset). |
| ddsql_get_spec | Returns DDSQL capability specifications (DDSQL toolset). |
| ddsql_schema_search_tables | Searches available datasets and tables (DDSQL toolset). |
| ddsql_schema_get_table_columns | Gets column definitions for a table (DDSQL toolset). |
| ddsql_schema_search_unstructured_fields | Searches fields in logs, RUM, and spans (DDSQL toolset). |
| ddsql_run_query | Executes a DDSQL query (DDSQL toolset). |
| ddsql_create_link | Generates a DDSQL Editor link (DDSQL toolset). |
| search_datadog_error_tracking_issues | Searches issues across RUM, logs, and traces (error tracking toolset). |
| get_datadog_error_tracking_issue | Retrieves issue details and impact (error tracking toolset). |
| update_datadog_error_tracking_issue | Updates an issue's state or assignee (error tracking toolset). |
| list_datadog_feature_flags | Lists feature flags with pagination (feature flags toolset). |
| get_datadog_feature_flag | Retrieves feature flag details (feature flags toolset). |
| create_datadog_feature_flag | Creates a new feature flag (feature flags toolset). |
| list_datadog_feature_flag_environments | Lists available feature flag environments (feature flags toolset). |
| list_datadog_feature_flag_allocations | Lists allocation rules for a flag (feature flags toolset). |
| update_datadog_feature_flag_environment | Updates a flag's environment configuration (feature flags toolset). |
| check_datadog_flag_implementation | Verifies a flag's code implementation (feature flags toolset). |
| sync_datadog_feature_flag_allocations | Syncs feature flag allocations (feature flags toolset). |
| search_datadog_k8s_resources | Searches Kubernetes resources across clusters (Kubernetes toolset). |
| describe_datadog_k8s_resource | Gets detailed information about a Kubernetes resource (Kubernetes toolset). |
| get_datadog_k8s_manifest | Retrieves the YAML manifest for a Kubernetes resource (Kubernetes toolset). |
| analyze_cloud_network_monitoring | Analyzes network flow data (networks toolset). |
| search_ndm_devices | Searches monitored network devices (networks toolset). |
| get_ndm_device | Retrieves details for a network device (networks toolset). |
| search_ndm_interfaces | Lists interfaces on network devices (networks toolset). |
| browser_onboarding | Guides Browser RUM setup (onboarding toolset). |
| devices_onboarding | Guides device monitoring setup (onboarding toolset). |
| kubernetes_onboarding | Guides Kubernetes cluster onboarding (onboarding toolset). |
| llm_observability_onboarding | Guides LLM observability setup (onboarding toolset). |
| test_optimization_onboarding | Guides test optimization setup (onboarding toolset). |
| serverless_onboarding | Guides AWS Lambda and GCP Cloud Run setup (onboarding toolset). |
| source_map_uploads | Guides source map uploading for RUM (onboarding toolset). |
| list_reference_tables | Lists and searches reference tables (reference tables toolset). |
| get_reference_table_rows | Retrieves reference table rows by primary key (reference tables toolset). |
| append_reference_table_rows | Adds rows to a reference table (reference tables toolset). |
| create_reference_table | Creates a reference table from cloud storage (reference tables toolset). |
| search_rum_applications | Lists RUM applications and resolves application IDs (RUM toolset). |
| get_rum_summary | Returns RUM vital metrics with period comparisons (RUM toolset). |
| get_rum_insight | Returns aggregated RUM view insights (RUM toolset). |
| search_rum_metrics | Explores available RUM metrics (RUM toolset). |
| search_rum_retention_filters | Lists RUM retention filters (RUM toolset). |
| datadog_secrets_scan | Scans for hardcoded secrets and credentials (security toolset). |
| search_datadog_security_signals | Searches Cloud SIEM and App & API Protection signals (security toolset). |
| analyze_datadog_security_signals | Analyzes security signals using SQL aggregations (security toolset). |
| get_datadog_security_signal | Retrieves details for a single security signal (security toolset). |
| security_findings_schema | Returns the available security finding fields (security toolset). |
| analyze_security_findings | Queries security findings with SQL (security toolset). |
| search_security_findings | Retrieves complete security finding objects (security toolset). |
| search_datadog_ci_pipeline_events | Searches CI pipeline events with filtering (software delivery toolset). |
| aggregate_datadog_ci_pipeline_events | Aggregates CI pipeline statistics (software delivery toolset). |
| get_datadog_flaky_tests | Searches flaky tests with triage details (software delivery toolset). |
| update_datadog_flaky_test_states | Sets the state of flaky tests (software delivery toolset). |
| aggregate_datadog_test_events | Aggregates test reliability trends (software delivery toolset). |
| search_datadog_test_events | Searches test events by branch or commit (software delivery toolset). |
| get_datadog_code_coverage_branch_summary | Fetches branch code-coverage metrics (software delivery toolset). |
| get_datadog_code_coverage_commit_summary | Fetches commit code-coverage metrics (software delivery toolset). |
| get_datadog_test_optimization_settings | Retrieves enabled test optimization features (software delivery toolset). |
| get_datadog_flaky_tests_management_policies | Retrieves auto-quarantine rules for flaky tests (software delivery toolset). |
| search_dora_deployments | Searches deployment events for DORA metrics (software delivery toolset). |
| aggregate_dora_deployments | Aggregates DORA deployment metrics (software delivery toolset). |
| get_synthetics_tests | Searches HTTP API synthetic tests (synthetics toolset). |
| edit_synthetics_tests | Edits synthetic test assertions and configuration (synthetics toolset). |
| synthetics_test_wizard | Creates HTTP API synthetic tests (synthetics toolset). |
| list_datadog_workflows | Lists and searches workflows (workflows toolset). |
| get_datadog_workflow | Retrieves workflow details and triggers (workflows toolset). |
| execute_datadog_workflow | Runs a published workflow (workflows toolset). |
| get_datadog_workflow_instance | Retrieves the status of a workflow execution (workflows toolset). |
| update_datadog_workflow_with_agent_trigger | Adds an agent trigger capability to a workflow (workflows toolset). |
Required configuration
- DD_API_KEYOptional
Datadog API key, sent as an HTTP header when OAuth is unavailable. Use a scoped service-account key.
- DD_APPLICATION_KEYOptional
Datadog application key, paired with the API key for header-based authentication.
What you can do with it
Investigate an incident end to end
The agent searches monitors and incidents, pulls the offending service's traces and logs, checks Watchdog stories, and correlates a recent deployment to find the root cause without you switching tabs.
Triage an alert from the editor
When a monitor fires, the agent queries the relevant metric, analyzes logs around the spike, and inspects upstream and downstream service dependencies to tell you what is actually broken.
FAQ
- Is it free?
- The MCP server itself is included with Datadog at no extra charge; you need a Datadog account and pay normal Datadog usage. Tools only expose data for products your plan already includes.
- Does it support remote/OAuth?
- Yes. It is a hosted remote server (US1: https://mcp.datadoghq.com/api/unstable/mcp-server/mcp, swap the domain for your region) that uses OAuth 2.0 in interactive clients and accepts a scoped DD_API_KEY plus DD_APPLICATION_KEY as HTTP headers for headless agents.
- How do I control which tools are exposed?
- Append a toolsets query parameter to the endpoint URL. The default is the core toolset; toolsets=all enables every generally available toolset, or you can list specific ones like apm, security, or synthetics.