New MCP server from groundcover redefines LLM observability
A new MCP server, faster than any other on the market, is launching today from groundcover, the eBPF-driven observability platform. Developers can now enhance their AI-driven workflows with deep system context, powered by groundcover’s granular access to logs, metrics, and traces via eBPF.
“AI is transforming software development and enabling teams to move faster than ever – but to ensure accuracy and reliability, AI needs the same deep context developers rely on,” said Orr Benjamin, Head of Product at groundcover. “By integrating with groundcover’s MCP server, tools like Cursor gain unparalleled visibility into the system and architecture, empowering AI to deliver code that’s not just fast but also production-ready.”
Developers are now deploying code 70% faster with AI assistance according to the latest DORA report. However, there are myriad day two problems that come with this tremendous boost in output, leaving platform engineering and security teams struggling to keep up. By plugging groundcover’s new MCP server directly into an IDE like Cursor, hallucinations will be reduced, and the quality of the code improved.
Developers working with AI applications can take advantage of all of the rich data already within groundcover including identified errors, problematic logs, cloud infrastructure issues, traces with specific customer IDs, the topology of communication of services, and more. The company’s new MCP server bakes all of this critical context into every piece of the software development lifecycle – from code generation to incident response – setting a new standard for LLM-driven observability that ensures quality, security, and performance.
“The same ultra-granular logs, metrics, and traces that drive groundcover’s observability now act as a force-multiplier for SecOps. The result is smarter threat detection, shorter incident response times, and a stronger security posture—without adding extra agents or overhead,” Benjamin told Help Net Security.