Developer Tools & Devops · New relic + Claude
New Relic MCP with Claude: Faster Incident Triage and Performance Insight
The short answer
Claude works with New Relic through the Model Context Protocol (MCP), an open standard that lets Claude securely use other software. Once connected, Claude can fetch application metrics, query logs, send incident alerts and analyse system performance from natural-language requests. Access is controlled by the organisation, and consequential actions can require human approval.
What is New relic?
New Relic is a unified observability platform. It helps teams monitor applications, infrastructure and customer experience from one place. Engineering, site reliability, DevOps and IT operations teams use it to spot problems, understand why they happen and improve performance across their technology stack. It brings together data from many systems so teams can troubleshoot issues faster.
How Claude connects to New relic through MCP
The Model Context Protocol is an open standard that acts like a universal adapter between Claude and other software. It lets Claude read information from a tool and take actions in it, using permissions the organisation sets. For New Relic, the connection can use the New Relic MCP server or a direct API connection, depending on what suits the organisation.
Once connected with an API key, Claude can typically fetch application metrics, query logs, review system performance and draft or send incident alerts. A team member can ask a question such as which service slowed down overnight, and Claude can gather the relevant data and summarise it in plain English.
Permissions and approvals keep people in charge. Access is limited to the data and actions the organisation chooses, and the API key can be scoped and rotated. Alerts or changes that matter can be routed through a human approval step, and every action can be logged for audit.
What you can automate
Overnight incident summaries
Claude can review alerts and error trends from the previous night and produce a short summary for the on-call engineer. The team starts the day knowing which services were affected, what changed and where to look first, instead of scrolling through dashboards.
Plain-language log queries
Support and product staff often need answers from logs but cannot write query syntax. Claude can translate a question such as how many checkout errors happened after Tuesday's release into the right query, run it and explain the result in clear terms.
Performance review for release managers
Before and after a release, Claude can compare application response times, error rates and throughput. It can highlight regressions, so release managers can decide quickly whether to continue, roll back or investigate further.
Customer experience reporting
Customer experience leads can ask Claude for a weekly view of page load times or key transaction performance across regions. Claude can gather the metrics and draft a short report that the business team can read without needing observability expertise.
Alert triage and routing
When alerts arrive in volume, Claude can group related ones, suggest likely causes and draft an incident message for the right team. A person still reviews and approves anything that is sent, so the response stays accountable.
An end-to-end workflow: Overnight alert triage to morning briefing
TriggerNew high-severity alerts are raised in New Relic outside business hours.
Claude fetches the new alerts and the related application metrics and logs from New Relic.
Claude groups related alerts and identifies the likely affected service and time window.
Claude creates a draft incident summary and records the key details in the organisation's ticketing system.
The on-call engineer receives the summary in Microsoft Teams and approves or edits the proposed next steps.
Claude posts the approved update to the incident channel and prepares a morning briefing for the operations lead.
OutcomeTeams typically start the day with a clear picture of overnight incidents, faster first response and a complete audit trail of what was reported and approved.
How we set it up
Free automation audit
We review how your teams use New Relic today, which incidents and reports take the most time, and which workflows are worth automating first. There is no cost for this session.
Secure connection
We connect Claude to New Relic using an API key, the sign-in method New Relic supports for this set-up. The key is stored securely, scoped to the access your workflow needs and rotated on a schedule.
Workflow design
We design the end-to-end workflow across New Relic and your other systems, such as Microsoft 365, your CRM, ERP or ticketing tool. People approve consequential actions, and access is limited to what each workflow requires.
Testing with real cases
We test the workflow against past incidents and reports from your own environment. We check accuracy, adjust prompts and confirm that alerts and actions behave as intended before anything goes live.
Launch, monitoring and training
We launch with monitoring and audit logs in place, then train your engineers, operations staff and managers on how to use Claude with New Relic safely and effectively.
Who benefits
- Site reliability engineers
- Faster triage of alerts, with summaries that show likely causes and the right next step.
- DevOps and platform leads
- Quicker access to performance data across the stack, and clearer release reviews.
- IT operations managers
- A consistent incident record and audit trail, with less manual reporting.
- Product and customer experience teams
- Plain-language answers about performance and customer impact, without writing queries.
Security and control
- Use an API key scoped to only the data and actions each workflow needs, and rotate it on a schedule.
- Require human approval before alerts are sent or changes are made in production systems.
- Log every query and action Claude takes so that reviews and audits are possible.
- Test thoroughly against real past cases before launch, and monitor the workflow after go-live.
- Restrict who in the organisation can use the Claude workflow and what data it can reach.
For teams in the US, UK, India and the Middle East
Organisations in the United States and the United Kingdom can design Claude workflows around their own data protection and security requirements. In India, workflows can be designed to support obligations under the Digital Personal Data Protection Act, 2023. Organisations in the UAE and Saudi Arabia can align workflows with local data protection laws. Claude can be hosted in India, the United States, the United Kingdom or the European Union, and the hosting region is chosen to meet each organisation's requirements.
Learn more about Claude and MCP
Anthropic's AI model and assistant.
Secure links between Claude and business tools.
How engineers build and change software with Claude.
Hosted agents that run multi-step work.
The open standard Claude uses to connect to tools.
Frequently asked questions
What does the New Relic MCP server do?
The New Relic MCP server gives Claude a secure, standard way to reach New Relic data and actions. It can typically help fetch application metrics, query logs, analyse system performance and send incident alerts. The organisation decides what access the server has, so Claude only works within the permissions that have been set.
Is connecting Claude to New Relic secure?
Security depends on how the connection is designed. Typical safeguards include an API key with limited scope, regular key rotation, human approval for consequential actions and full audit logging. We design each connection around least-privilege access so that Claude only reaches the data a workflow genuinely needs.
What can Claude do with New Relic once connected?
Once connected, Claude can typically answer questions in plain language about application performance, logs and system health. It can also summarise incidents, compare metrics over time and draft or send alerts, subject to the approvals you set. The exact capabilities depend on the access granted.
How long does setup take?
A simple connection and single workflow can often be set up within a few weeks, including testing. Larger set-ups that involve several systems or custom applications take longer. We confirm the timeline after the free automation audit, once we understand your environment and priorities.
Which New Relic access is needed?
Connection is made with an API key. The level of access depends on what your workflow requires, and we recommend the narrowest scope that works. We confirm the exact requirements during the audit, based on your New Relic set-up and internal policies.