datadog-mcp

Query Datadog telemetry via MCP toolsets for observability workflows.

3|1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/Cogni-AI-OU/cogni-ai-agent-skills --skill datadog-mcp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: datadog-mcp
Source: https://github.com/Cogni-AI-OU/cogni-ai-agent-skills/tree/main/datadog-mcp
Command: npx skills add https://github.com/Cogni-AI-OU/cogni-ai-agent-skills --skill datadog-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Datadog telemetry retrieval often forces agents to switch context between tools or open UIs, wasting tokens and time while reducing reproducibility.

Core Features & Use Cases

  • Minimal toolset querying: Selects the narrowest Datadog MCP toolset needed (e.g., apm, alerting, llmobs) to reduce context window impact.
  • Telemetry domains coverage: Supports logs, metrics, traces, monitors, incidents, services, and more through task-appropriate MCP toolsets.
  • MCP-only discipline: Enforces MCP exclusivity for standard observability retrieval to avoid SSH, UI navigation, and raw API/curl-style fallbacks.

Quick Start

Use the datadog-mcp skill to analyze APM traces for a suspected performance regression by loading the apm toolset and querying spans relevant to the time window and service impacted.

Frequently Asked Questions about datadog-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I query Datadog logs and traces without context switching?▼

Query Datadog logs and traces without context switching by loading the minimal MCP toolset needed for your domain, executing targeted telemetry queries in-context, and synthesizing findings without UI navigation or raw API calls.

What is the best way to investigate APM trace spans for a performance regression?▼

Investigate APM trace spans for a performance regression by loading the `apm` MCP toolset and querying spans relevant to the impacted service and time window, synthesizing telemetry findings directly in-context.

Does Datadog MCP querying support monitors and incidents, or just logs and metrics?▼

Datadog MCP querying supports monitors and incidents alongside logs, metrics, traces, and services. It covers these telemetry domains through task-appropriate MCP toolsets like `apm` and `alerting` for comprehensive observability analysis.

How do I validate Datadog monitors using MCP queries?▼

Validate Datadog monitors using MCP queries by loading the `alerting` toolset to retrieve and analyze monitor states and configurations, enforcing MCP exclusivity for standard observability retrieval and audit-style analysis without UI navigation.

When should I not use raw API calls or SSH for Datadog telemetry retrieval?▼

Avoid raw API calls or SSH for Datadog telemetry retrieval when standard observability data is available via MCP. The MCP-only discipline restricts retrieval to MCP tools to avoid token waste and improve context window efficiency.