Datadog MCP
by Datadog
AI-assisted monitoring, metrics, and incident investigation with Datadog
devops Python Intermediate Self-hostable Verified
⭐ 200 stars 📅 Updated: 1mo ago
Description
Official Datadog MCP server that brings the full power of Datadog's monitoring and observability platform to AI assistants. Query metrics, search through logs, analyze distributed traces, and manage monitors and dashboards — all through natural language. Built for DevOps teams and SREs who want to speed up incident investigation, review infrastructure health, and create monitoring configurations without switching context. Supports the complete Datadog API surface for read and management operations across your entire observability stack.
✅ Best for
DevOps teams and SREs using Datadog who want conversational access to monitoring data
⏭️ Skip if
You don't use Datadog as your monitoring platform
💡 Use cases
- Investigating production incidents by querying metrics and logs conversationally
- Creating and managing monitors and alert configurations through AI
- Reviewing dashboard data and infrastructure health during on-call shifts
- Analyzing APM traces to identify performance bottlenecks
👍 Pros
- ✓ Official first-party server from the Datadog team
- ✓ Comprehensive coverage of metrics, logs, traces, monitors, and dashboards
- ✓ Python-based with straightforward pip installation
- ✓ Supports both read operations and management actions
👎 Cons
- ✗ Requires both API and APP keys for full functionality
- ✗ Datadog's pricing model means heavy query usage may impact costs
- ✗ Large metric query results can consume significant context window
🔧 Exposed tools (5 tools)
| Tool | Category | Description |
|---|---|---|
| list_monitors | management | List and filter configured monitors and their statuses |
| manage_dashboards | management | View, create, and update Datadog dashboards |
| query_metrics | query | Query and retrieve metrics data from Datadog |
| search_logs | query | Search and filter log entries with Datadog log queries |
| get_traces | query | Retrieve and analyze APM distributed traces |
⚡ Installation
Prerequisites:
- • python v3.10+
- • Datadog API key
- • Datadog APP key
- • API key required
Check Claude Code documentation to configure this MCP server.
💡 Tips & tricks
Set DD_API_KEY and DD_APP_KEY environment variables before starting. Start with list_monitors to get an overview of your alerting setup, then drill into specific metrics or logs for investigation.
🔄 Alternatives
Quick info
- Author
- Datadog
- License
- Apache-2.0
- Runtime
- Python 3.10+
- Transport
- stdio
- Category
- devops
- Difficulty
- Intermediate
- Self-hostable
- ✅
- Auth
- —
- Docker
- —
- Version
- latest
- Updated
- Feb 28, 2026
Client compatibility
- ✅ Claude Code
- ✅ Cursor
- ✅ VS Code Copilot
- ❓ Gemini CLI
- ✅ Windsurf
- ✅ Cline
- ❓ JetBrains AI
- ❓ Warp
Platforms
🍎 macOS 🐧 Linux 🪟 Windows