Datadog AI (LLM Observability) MCP Server

Connect your Datadog account to any AI agent and take full control of your LLM observability and AI performance monitoring through natural conversation.

What you can do

• LLM Metrics Auditing — Query high-precision numeric telemetry targeting LLM Observability timeseries like token counts and latency
• Prompt & Span Search — Retrieve explicit APM payload contents capturing literal prompt logic and response traces limitlessly
• AI Monitor Management — List and create monitors to track when AI responses drop below SLI thresholds or plateau on requests
• Dashboard Insights — Enumerate widgets graphing global AI expenses across providers like OpenAI or Anthropic
• Incident Tracking — Monitor active outages and service disruptions blocking multi-agent orchestration dynamically
• Timeline Events — Pull pure textual deployment marks identifying exactly when dynamic LLM models were switched

How it works

1. Subscribe to this server
2. Enter your Datadog API Key, APP Key, and Site
3. Start monitoring your AI infrastructure from Claude, Cursor, or any MCP-compatible client

Who is this for?

• AI Engineers — monitor LLM latencies and token costs in real-time without leaving the dev environment
• MLOps Teams — audit prompt logs and trace AI model performance across different versions
• SREs — set up monitors for AI services and track incidents affecting agentic workflows
• FinOps — analyze dashboards graphing global AI infrastructure expenses and usage patterns