New Relic Inc. is extending its observability capabilities for artificial intelligence agents with the launch today of a new Agentic AI Monitoring tool and its AI Model Context Protocol Server.

The first is meant to provide companies with more holistic visibility into their interconnected autonomous agents and the tools they use, to help with optimizing those agentic workforces. The latter provides a standardized way for companies to enable comprehensive observability for their AI agents.

New Relic explained that monitoring AI agents is not an easy thing, because the more sophisticated they become, the greater their complexity. The average AI agent deployed in the enterprise today relies on the outputs of multiple other agents, tools from multiple MCP servers and shared content and memory. It creates an intricate web that’s extremely difficult to debug when things go wrong.

For instance, if one agent starts hallucinating and sending false information to another, and then on to another, the faulty data might not be recognized for several steps. To solve this, New Relic aims to provide more comprehensive visibility into the health of every AI agent.

New Relic Agentic AI Monitoring does exactly that, providing insights into each agentic deployment and each tool they use. The idea is to help companies understand how their mesh of agents is communicating with each other, so that problems can be identified and troubleshooted more rapidly.

The capability delivers granular insights into tool utilization, performance and errors with a view that shows which agents and tools were called — in what order — and key performance data. It’s aided by an Agents Service Map that visualizes inter-agent interaction and enables their history to be fully traced. As a result, DevOps teams will be able to spot issues with their AI agents much faster, accelerate root cause analysis, solve problems and increase their performance.

As for the AI MCP Server, this brings New Relic’s basic observability capabilities to any AI agent, so developers can understand how their code is running in production. It allows engineers to retrieve deep, detailed data and insights from wherever they’re working, making their favorite AI agents instantly more powerful and productive. By eliminating context switching, teams can respond to incidents faster, shorten resolution times, improve uptime and accelerate new agent deployments.

In a related update, New Relic announced the general availability of Logs Intelligence, which is a series of new AI-enabled capabilities that can reduce the time and effort to extract critical intelligence about AI applications.

The new capabilities help to automate the way teams interact with log data, and can automatically analyze it to provide instant hypotheses of application issues. They also help eliminate the mundane and burdensome task of manually querying logs to complete periodic reports.

Image: SiliconANGLE/Meta AI