dynatrace

The Dynatrace MCP Server brings Dynatrace observability data into your MCP-enabled assistants by exposing a local, AI-assisted interface for querying and interacting with your Dynatrace environment. It supports capabilities such as executing Dynatrace Query Language (DQL) against Grail-backed data, retrieving problems and vulnerabilities, and sending notifications or creating documents within the Dynatrace workspace. You can connect this MCP server to popular MCP clients (e.g., VSCode, Claude, Cursor, Amazon Q, Windsurf, ChatGPT, GitHub Copilot) and drive real-time observability insights through natural language prompts. Common workflows include asking for problem details, querying logs or events, and obtaining contextual debugging information with AI-generated explanations of DQL results.

Once configured, you can use the MCP server with a command like: npx -y @dynatrace-oss/dynatrace-mcp-server@latest, and provide your Dynatrace environment URL via DT_ENVIRONMENT. This will establish authenticated access to your Dynatrace environment and enable prompts such as "Show me error logs for service X" or "List recent security problems for cluster Y." The server also supports budget-tracking for Grail queries and guidance on cost-aware querying, so you can monitor usage while exploring data.

In addition to data retrieval, the server exposes capabilities such as:

• DQL execution to fetch logs, events, spans, and metrics
• Access to problem and vulnerability details
• Creation and retrieval of Dynatrace notebooks, dashboards, and other documents
• Notifications via Slack and automation engine workflows for proactive responses

Prerequisites:

• Node.js (with npm or npx) installed on your machine
• Access to a Dynatrace environment and appropriate permissions
• Internet access to install the MCP server package

Installation steps:

1. Install Node.js from https://nodejs.org/ (if not already installed).
2. Install or update npm to ensure npx is available.
3. Run the MCP server installation command from your preferred shell:

npx -y @dynatrace-oss/dynatrace-mcp-server@latest

4. Configure your environment variables, notably DT_ENVIRONMENT, to point to your Dynatrace environment. Example configuration (for VSCode MCP settings):

{
  "servers": {
    "npx-dynatrace-mcp-server": {
      "command": "npx",
      "args": ["-y", "@dynatrace-oss/dynatrace-mcp-server@latest"],
      "env": {
        "DT_ENVIRONMENT": "https://your-dynatrace-environment-url"
      }
    }
  }
}

5. Start using the MCP server with your MCP client by selecting the configured server entry and connecting. The DT_ENVIRONMENT value should point to your Dynatrace environment without using legacy Dynatrace URLs.

6. For local development, you can also pin a specific version by replacing @latest with a version tag, e.g., @dynatrace-oss/dynatrace-mcp-server@v1.2.3

Tips and important notes:

• DT_ENVIRONMENT must be set to your Dynatrace Platform URL (e.g., https://abc12345.apps.dynatrace.com). Do not use Dynatrace classic URLs.
• Authentication uses Authorization Code Flow in the browser; you do not need a Platform Token or OAuth Client to get started.
• The server includes Grail query budget tracking via DT_GRAIL_QUERY_BUDGET_GB (default is 1000 GB). Set this env var to control your query budget and monitor consumption.
• There are cost considerations for certain capabilities that query Grail data. Start with smaller timeframes and use built-in budgets to avoid unexpected charges.
• Capabilities include: DQL execution, retrieving problems and vulnerabilities, Slack notifications, Automation Engine workflows, and document management (Notebooks, Dashboards, Launchpads).
• If you encounter environment-related issues, ensure the DT_ENVIRONMENT URL is reachable from your network and that any required OAuth flows can complete in your browser.