New Relic's new monitoring service monitors and optimizes MCP sources
New Relic's AI monitoring cloud now includes MCP sources in its analyses. This allows developers to optimize their performance and usage.
(Image: Gorodenkoff/Shutterstock.com)
- Manuel Masiero
The US web tracking and analysis company New Relic has added MCP (Model Context Protocol) support to its AI monitoring service. This allows developers to analyze the entire life cycle of an MCP request, including the tools used, call sequences and execution time. In addition to usage patterns, latencies and errors, the MCP optimization evaluates which tools the AI agents use for certain commands.
MCP monitoring is available from version 10.13.0 of the New Relic Python agent; support for other languages is planned for future versions.
(Image:Â New Relic)
MCP servers "often work as black boxes", explains the company in its announcement. They make it difficult for developers to monitor the performance of their AI agents. It is also difficult for MCP providers to identify performance bottlenecks and errors in their systems, which slows down their optimization and further development.
Since its release at the end of last year, Anthropic's open-source framework MCP has opened doors. The platform-independent standard for communication between AI systems and external data sources is being actively developed by the community. Security vulnerabilities have also emerged, such as the prompt injection attack on GitHub's MCP server.
Videos by heise
This has not dampened the success of MCP. Major AI providers such as OpenAI and Google DeepMind are now also relying on the Model Context Protocol. Microsoft has also jumped on the bandwagon and announced at its in-house developer fair Build 2025 that it will integrate agent-based AI into Windows 11.
(mma)
