# dynatrace-managed-mcp [Health: Active]

**Category:** 📊 Monitoring  
**Repository:** https://github.com/dynatrace-oss/dynatrace-managed-mcp  
**GitHub Stars:** 33  
**npm Downloads (last month):** 2992  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/dynatrace-managed-mcp

## Description
MCP server for Dynatrace Managed to access logs, events, and metrics.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "dynatrace-managed-mcp": {
    "command": "npx",
    "args": ["-y","@dynatrace-oss/dynatrace-managed-mcp-server@latest"],
    "env": {
      "DT_CONFIG_FILE": "",
      "DT_ENVIRONMENT_CONFIGS": ""
    }
  }
}
```

**Requires environment variables:** `DT_CONFIG_FILE`, `DT_ENVIRONMENT_CONFIGS` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What dynatrace-managed-mcp MCP server does

The dynatrace-managed-mcp MCP server gives AI assistants access to observability data stored in self-hosted Dynatrace Managed environments. It can work with one environment or several, allowing the assistant to query data through MCP tools instead of requiring separate manual API requests.

Supported data areas include problem records, security problems and vulnerability details, monitored entities and their relationships, service-level objectives, system events, logs, and performance metrics. The server is intended for environments where Dynatrace Managed is the primary observability system. It can also provide historical data after an organization has moved newer observability workloads to Dynatrace SaaS.

## How it works

The server reads a configuration describing each Dynatrace Managed environment. Every entry identifies the cluster API URL, the managed environment ID, and a human-readable alias. Authentication uses a Dynatrace Managed API token with the scopes required for the requested data. Multiple entries let an assistant address several environments through the same MCP connection.

Two operating modes are documented. Local mode runs the server on the client machine, commonly through the npm package and a stdio MCP configuration. Remote mode communicates over HTTP/SSE and is intended for distributed or production-like arrangements. Natural-language requests are mapped to MCP tools, which in turn make the relevant Dynatrace API queries.

## Setup and configuration

To run the dynatrace-managed-mcp MCP server locally, provide a Dynatrace Managed API token, an environment configuration file, an MCP connection definition, and the `DT_CONFIG_FILE` variable pointing to the YAML or JSON configuration file. The file can contain tokens directly or reference runtime variables such as `DT_PROD_TOKEN`. An alternative configuration uses `DT_ENVIRONMENT_CONFIGS` with a JSON array, which is suitable for Docker or Kubernetes deployments.

The token needs `DataExport`, `entities.read`, `events.read`, `logs.read`, `metrics.read`, `problems.read`, `securityProblems.read`, and `slo.read` for full functionality. Optional HTTP and HTTPS proxy URLs can be set per environment, but the documented configuration says not to use both proxy parameters together.

## Tools and capabilities

Available capability areas include:

- Listing and retrieving problem details, including service-related problems.
- Listing security problems and retrieving vulnerability information.
- Inspecting monitored entities and relationship mappings.
- Listing and retrieving SLO details, evaluations, and error budgets.
- Listing and retrieving system events.
- Searching and filtering logs with content and time-based criteria.
- Querying and analyzing performance metrics through the Dynatrace V2 Metrics API.

## Limitations and notes

The dynatrace-managed-mcp MCP server is specifically for self-hosted Dynatrace Managed deployments. Dynatrace SaaS environments require the separate Dynatrace MCP project. The project is community-supported, with issue-based assistance referenced in the repository. Access also depends on the configured token scopes and the availability of the target Managed cluster.

_Full upstream README: https://allmcps.com/mcp/dynatrace-managed-mcp/readme_