mcp

MCP server from newrelic/mcp-server

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio newrelic-mcp-server npx -y newrelic-mcp-server \
  --env AUTH_TOKEN="your authentication token or API key for New Relic" \
  --env NRQL_ENDPOINT="https://api.newrelic.com/v2/alerts_rules.json (or appropriate NRQL/api endpoint)"

How to use

The New Relic MCP server exposes natural language interfaces to query and analyze your New Relic observability data. It connects AI agents and tools to NRQL queries, alerts, incidents, logs, and performance metrics to help you investigate issues, assess deployment impact, and generate insights from your monitoring infrastructure. You can use it to perform tasks like retrieving recent incidents, analyzing alert policies, running NRQL queries via natural language, and exploring performance data across services.

To use it, configure your MCP client to point at the server endpoint provided by your deployment. Authenticate with the required credentials (API key or OAuth token) as documented. Once connected, you can issue natural language commands such as asking for the latest incident trends, querying metrics for a service, or analyzing deployment impact. The server translates these requests into the appropriate NRQL queries and other tool calls, returning structured results and insights you can act on.

How to install

Prerequisites:

Node.js v14 or later and npm/yarn installed on your machine
Access to a New Relic account and the necessary API keys or OAuth credentials

Installation steps:

Install Node.js and npm from https://nodejs.org
Verify installation:
- node -v
- npm -v
Install or run the MCP server via npx (as defined in the configuration):
- npx -y newrelic-mcp-server This will fetch and start the New Relic MCP server package.
Obtain your authentication credentials and configure environment variables as needed (see mcp_config for placeholders).
Point your MCP client to the server endpoint and optionally configure any NRQL data sources or alerting endpoints as required by your environment.

Additional notes

Tips:

Ensure your API keys or tokens have the required scope to query NRQL data, alerts, and incidents in New Relic.
If you run behind a firewall or proxy, configure environment variables or client settings to allow outbound API access to New Relic endpoints.
The server supports natural language queries; consider providing examples to calibrate responses, such as 'Show me the incidents from last 24 hours' or 'What is the NRQL query behind the latest error rate spike?'.
Review rate limits and plan accordingly for high-volume queries.
Keep the MCP server version updated to leverage new NR integrations and security fixes.