The newest open-source project out of Intel is the Intel Performance Skills project that is providing AI agent skills to help with CPU performance analysis and performance optimizations on Linux.

Intel Performance Skills began coming about in recent weeks to provide AI agents with structured, multi-step workflows for profiling, performance anti-patterns in source code, and more. The Intel Performance Skills can work with GitHub CLI / Copilot, Claude COde, OpenAI Codex, Gemini CLI, and OpenCode.

The MIT-licensed project can detect coding patterns and fixes around narrow SIMD usage, serial accumulator patterns, spinlock patterns, false sharing, and various other coding patterns and fixes for better performance.

There is skill integration around the Linux perf utility for profiling data collection to help in the analysis. In addition, Intel Performance Skills has native Phoronix Test Suite integration too for being able to run Phoronix benchmarks and analyze new optimizations in the software under test. In one of the benchmarks, for example, they found a 16x performance optimization possible with the software program being benchmarked.

Those curious about Intel Performance Skills for helping AI agents uncover new performance optimizations for Linux can find the project via intel/intel-performance-skills on GitHub.