MCP Datadog Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DATADOG_SITE | No | Datadog site (e.g., datadoghq.com, datadoghq.eu, us3.datadoghq.com). | datadoghq.com |
| DATADOG_API_KEY | Yes | Your Datadog API key. | |
| DATADOG_APP_KEY | Yes | Your Datadog application key. | |
| MCP_SLOW_TOOL_MS | No | Threshold in milliseconds to mark a tool call as slow. | 2000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| query_metricsA | Query Datadog metrics data for a specified time range. Returns time-series data with values aggregated over the specified period. |
| get_metric_metadataA | Retrieve metadata about a Datadog metric including units, description, tags, and integration information. |
| list_metricsA | List available Datadog metrics, optionally filtered by a search query. Useful for discovering what metrics are available in your environment. |
| search_logsB | Search Datadog logs with filters and a time range. Returns log entries matching the query, useful for debugging and log analysis. |
| get_log_detailsA | Get detailed information about a specific log entry. Returns full log data including all attributes and metadata. |
| aggregate_logsB | Aggregate log data for a time range using the specified aggregation type (count, avg, percentile, min, max, sum). Useful for statistical analysis of logs. |
| search_eventsA | Search for events in Datadog. Events can include monitor alerts, deployments, integrations, and custom events. Useful for understanding system changes and incidents. |
| get_event_detailsA | Get detailed information about a specific event. Returns full event data including timestamps, comments, and metadata. |
| list_monitorsB | List all Datadog monitors with optional filtering by status and tags. Useful for getting an overview of all monitors in your account. |
| get_monitor_statusA | Get detailed status information for a specific monitor, including alert status, downtime information, and historical state changes. |
| search_monitorsA | Search for monitors by name or other criteria. Useful for finding specific monitors when you don't remember the exact ID. |
| query_tracesA | Query Datadog APM traces for a service. Returns trace data with latency information and span details. Useful for performance debugging. |
| get_service_healthA | Get health metrics for a service including latency, error rate, and throughput. Useful for monitoring service performance and health status. |
| get_apm_service_dependenciesA | Get the APM service dependency map for a service over a time range, showing which services it calls and which services call it based on trace data. Useful for understanding service interactions. |
| get_service_dependenciesB | Get service dependencies for a given environment. Returns all services and their relationships in the specified environment. Useful for understanding service architecture and dependencies. |
| get_service_dependencies_multi_envA | Get service dependencies across multiple environments. Returns service relationships for each specified environment, useful for comparing architectures across prod, staging, dev, etc. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 16 tools
Most tools are distinct, but the three service dependency tools (get_apm_service_dependencies, get_service_dependencies, get_service_dependencies_multi_env) have highly overlapping purposes and could be easily misselected by an agent. Other tools like list_metrics vs query_metrics are clearly separated, but the dependency trio introduces ambiguity.
Tool names consistently use snake_case with a verb_noun pattern (get_, list_, search_, query_, aggregate_). The verbs are predictable and align with their actions. Minor deviation: get_ prefixes are used for both single items (get_metric_metadata) and collections (get_service_dependencies), but this is not overly confusing.
16 tools is on the higher end of the recommended range, but it covers multiple Datadog features (metrics, logs, events, monitors, APM, dependencies) reasonably. Each tool serves a distinct function within these domains, so the count is justified, though it feels slightly heavy for a single server.
The set provides broad read-only coverage for Datadog's main observability areas, but it lacks any write operations (create, update, delete monitors, events, etc.). There are also gaps like dashboards and trace detail. The service dependency trio is overrepresented while monitor management is incomplete, leaving agents unable to perform lifecycle actions.