New Relic AI Model Context Protocol (MCP)

Model Context Protocol (MCP) is a standard way for AI models or agents to receive relevant information (context) about a specific system or environment. This context helps the AI understand the current situation, perform tasks more accurately, and provide more useful responses. Historically, connecting AI agents to various data sources like observability platforms meant building custom, often complex, integrations for each AI agent to talk to each platform. This fragmented approach leads to significant development and maintenance overhead, a form of "integration sprawl" for AI-driven insights.

The New Relic AI MCP server directly addresses this challenge. It is New Relic's implementation of this protocol, acting as a single, centralized bridge. Instead of building many individual integrations, your AI agents within supported development tools now have one consistent way to connect to and retrieve rich context and powerful tools from New Relic's observability platform.

You can ask questions and investigate data using natural language directly within your AI development environment. The MCP server translates your requests, retrieves relevant metrics or logs, and provides clearer answers, and summaries.

How it helps you

The MCP server offers practical benefits for engineers and operations teams and helps you and your AI in the following ways:

• Direct integration: Access New Relic's platform and some AI features directly from your development environment. This avoids switching between different tools.
• Faster troubleshooting: Quickly find and fix issues, analyze error logs, and determine probable causes with AI-provided insights.
• Automated operations: Perform tasks like checking alert statuses, generating incident reports, and analyzing how deployments affect your systems.
• Simple data queries: Formulate questions in plain English, and the server converts them into New Relic Query Language (NRQL) to get results, even if you are not familiar with NRQL.

Supported AI development tools

New Relic MCP works with these AI development environments:

• Claude Code: Command-line interface for Claude
• Claude Desktop: Desktop application for interactive AI development
• VS Code: Integrated development environment with MCP support
• Windsurf: Cloud-based AI development platform
• Gemini CLI: Command-line interface for Gemini

Each environment offers unique advantages depending on your workflow preferences. See our setup guide for platform-specific configuration instructions.

Prerequisites

To use New Relic MCP, you need:

• A New Relic account with API access
• Node.js installed (required for some authentication methods)
• One of the MCP-supported clients installed

Authentication options

• User API Key:

  1. Log in to New Relic.
  2. Select user menu -> API keys.
  3. Create new user key or copy existing one.
  4. Key format: NRAK-XXXXXXXXXXXXXXXXXXXXX.

• OAuth: New Relic provides pre-configured OAuth endpoints:

  • Client ID: pUWGgnjsQ0bydqCbavTPpw==
  • Authorization URL: https://login.newrelic.com/login
  • Token URL: https://mcp.newrelic.com/oauth2/token (US) and https://mcp.eu.newrelic.com/oauth2/token (EU)
  • Scopes: ["openid"]

Public preview access

Navigate to the Previews & Trials page in the New Relic UI and enable the New Relic AI MCP Server preview.

Next steps

Ready to connect your AI tools to New Relic observability data?

1. Choose your AI tool from our supported environments.
2. Follow our setup guide for step-by-step configuration.
3. Explore available tools in our tool reference.
4. Start querying your New Relic data using natural language.
5. Troubleshoot common issues.