Connect New Relic AI Observability MCP for AI Agents
Monitor LLM Performance and Token Costs in Real-Time
We take care of the infrastructure, maintenance, security, and governance. Works with:
Claude 
ChatGPT 
Cursor 
Gemini 
Windsurf 
VS Code 
Vercel No credit card required. Experience the power of this integration risk-free.
Waiting for input…
What AI agents can do with New Relic AI Observability: 10 Tools for LLM Performance Analysis
Use these specific tools to run deep queries on error logs, token costs, latency data, and system components from your New Relic monitoring platform.
List alert policies
Checks internal arrays for specific automated alerts that might be triggering issues in your system.
List apm apps
Automates a validation check to see which core services or gateways are currently active and running.
Custom nrql
Runs complex, read-only queries using New Relic Query Language (NRQL) to extract detailed insights from your data.
List dashboards
Identifies which monitoring dashboards are currently active and available for viewing within the platform.
Query llm errors
Checks for specific error records related to LLM processing, helping you quickly spot failures in your generation process.
Query llm costs
Performs extraction of properties that calculate the precise dollar cost associated with running your AI models.
Query llm events
Gathers specific records detailing events related to LLM usage within the New Relic platform.
Query llm feedback
Retrieves explicit user feedback and rating scores provided by human reviewers for your model outputs.
Query llm latency
Provides data on how fast your LLM responses are, including key metrics like the 95th percentile latency.
Post custom event
Adds custom telemetry rows to track specific internal states or behavioral markers for deeper monitoring.
Frequently Asked Questions
How does New Relic AI Observability help me track my spending on LLMs? +
It calculates your total token cost by extracting the usage data from all your models. You get a clear, dollar-based report that shows exactly how much each model costs you over time.
Can New Relic AI Observability tell me if my model is running slowly? +
Yes, it measures performance by providing the 95th percentile (p95) latency. You can ask for this data to see if your responses are meeting their speed goals.
Is New Relic AI Observability better than just looking at dashboards? +
It's much faster because you don't have to click around. Instead, you ask a natural language question—like 'What went wrong?'—and it pulls the right data from multiple sources automatically.
Does New Relic AI Observability help me understand what users think of my model? +
Yes, it gathers and displays human feedback. You can retrieve rating scores and comments, letting you identify quality issues quickly after deployment.
Can I check my total AI token costs through my agent? +
Yes. Use the query_llm_costs tool. Your agent will execute a NRQL aggregation summing the tokenSpanCost property from your LLM events over the last 24 hours, faceted by model, to provide a clear financial breakdown.
How do I monitor the p95 latency of my LLM generations? +
The query_llm_latency tool retrieves the average duration and latency matrices for your AI providers. Your agent will report the results as a timesheet or summary, helping you identify performance bottlenecks instantly.
Can my agent run custom NRQL queries against my telemetry data? +
Absolutely. Use the custom_nrql tool to provide any valid read-only NRQL string. Your agent will query New Relic's NerdGraph API and return the resulting dataset, allowing for complete flexibility in how you analyze your AI operations.
Other MCPs in this category
Sentry MCP Integration
15 toolsMonitor errors and application health via Sentry. Query issues, inspect events, audit releases and check alert rules directly from any AI agent.
YAML Parser Engine MCP Integration
1 toolsThe YAML Parser Engine MCP allows your AI agents to reliably convert data between YAML and JSON. It's essential for anyone dealing with complex configuration files, like Kubernetes manifests or Docker Compose setups. This tool guarantees zero structural loss when parsing deeply nested, whitespace-sensitive configurations that generic AI models often break.
Prismic
10 toolsQuery and manage your Prismic headless CMS content. Search documents, list custom types, and retrieve specific content directly from any AI agent.
Linear MCP
11 toolsShip software faster with issue tracking built for modern teams that combines speed, keyboard shortcuts, and beautiful design.
Related MCPs
Firebase (REST & Admin APIs) MCP Integration
14 toolsManage Firebase Realtime Database, Firestore, and Auth via REST APIs. Query data, manage users, and send FCM messages directly from your AI agent.
Envoy MCP
10 toolsManage workplace operations via Envoy. Register visitors, book desks and rooms, track deliveries, and monitor office capacity directly from any AI agent.
ORCID MCP
13 toolsAccess and manage ORCID researcher records. Search the registry, fetch biographical data, and manage works or affiliations directly.
Fresha MCP
11 toolsQuery and analyze your Fresha salon data. List appointments, clients, locations, services, staff, and sales transactions directly from any AI agent.
Your AI, connected to everything.
Connect New Relic AI (LLM Observability) and 5,600+ more MCP servers to Claude, Cursor, or any AI you use.
No credit card required · Free tier available
