Favicon of cpu-kernels

cpu-kernels Skill

AI Agent SkillPythonOpen source

Provides guidance for writing, optimizing, and benchmarking C++ CPU kernels with SIMD intrinsics (AVX2/AVX512) for the Hugging Face kernels ecosystem. Includes a two-phase workflow: Phase 1 correctness (generic → AVX2) and Phase 2 performance exploration (AVX512 with branching trial loop), runtime CPU dispatch, OpenMP Published by huggingface in kernels.

Decision snapshot

Is this a fit?

Best for

Writing, Includes SKILL.md, Reusable instructions

Works with

Compatibility not yet detected.

Access

Filesystem access, Shell access

Setup

Copy skill directory

Project health

1 month ago · Apache-2.0 license

Considerations

Access note: Local files. Access note: Shell access.

What is cpu-kernels Skill?

Provides guidance for writing, optimizing, and benchmarking C++ CPU kernels with SIMD intrinsics (AVX2/AVX512) for the Hugging Face kernels ecosystem. Includes a two-phase workflow: Phase 1 correctness (generic → AVX2) and Phase 2 performance exploration (AVX512 with branching trial loop), runtime CPU dispatch, OpenMP Published by huggingface in kernels. This profile combines repository metadata with install, compatibility, and usage signals so developers can quickly decide whether it fits their agent workflow before opening the source repository.

Trust signal
95/100
Maintenance signal
90/100
Adoption signal
71/100

Automated repository signals based on public metadata such as recency, license, installation evidence, and adoption. These are not a security audit or endorsement. See how SkillIndex evaluates profiles.

Key capabilities

  • Includes SKILL.md support
  • Reusable instructions support
  • Writing
  • Writing use cases

Declared skill metadata

  • Source file: kernel-builder/skills/cpu-kernels/SKILL.md

Allowed tools declared by source

ReadGrepGlobBash

These fields retain source and confidence evidence from the indexed SKILL.md.

Compatibility and setup

Copy skill directory
  • Install or run with Copy skill directory
  • Local files
  • Shell access

Requirements and access

Local filesShell access

Security and permissions

Review permissions before connecting any MCP server to an agent. Pay special attention to whether it can read local files, write data, call external services, or perform destructive actions.

Filesystem accessShell access

When to use cpu-kernels Skill

  • Use it for writing.

Built with

PythonCopy skill directory

Editorial notes

Source

  • Creator: huggingface
  • Repository: huggingface/kernels
  • Skill file: kernel-builder/skills/cpu-kernels/SKILL.md

What it does

Provides guidance for writing, optimizing, and benchmarking C++ CPU kernels with SIMD intrinsics (AVX2/AVX512) for the Hugging Face kernels ecosystem. Includes a two-phase workflow: Phase 1 correctness (generic → AVX2) and Phase 2 performance exploration (AVX512 with branching trial loop), runtime CPU dispatch, OpenMP

Skill instructions

CPU C++ Kernels for x86 Processors This skill provides patterns and guidance for developing optimized C++ kernels targeting x86 CPUs (Intel Xeon and compatible processors) with AVX2 and AVX512 intrinsics. Kernels are compiled via kernel-builder and distributed through the Hugging Face kernels ecosystem. Who runs these commands? You, the agent — not a human. This is an autonomous loop: you write/edit the C++ kernel, build it, then run the scripts below as tools (via Bash) to check correctness, benchmark, and profile. You read each result, record it with trialmanager.py, decide the next change from the Phase 2 decision tree, and repeat until you hit earlystopspeedup or run all maxtrials. Key Concepts (read before the Quick Start) The commands use a few names that mean different things. They are not interchangeable: | Name (example) | What it is | Used by | |----------------|-----------|---------| | baseline.py | The PyTorch reference implementation you optimize against. It is the ground

Verified compatibility and discovery

Frequently asked questions

What is cpu-kernels?

cpu-kernels is a open-source AI agent skill with Copy skill directory. Provides guidance for writing, optimizing, and benchmarking C++ CPU kernels with SIMD intrinsics (AVX2/AVX512) for the Hugging Face kernels ecosystem.

Who is cpu-kernels best for?

cpu-kernels is best for reusing agent instructions, scripts, and references, writing workflows.

How do I install cpu-kernels?

Install or run cpu-kernels using Copy skill directory. Check cpu-kernels for the latest setup command.

Is cpu-kernels actively maintained?

cpu-kernels may need a closer maintenance check before production use.

Share:

Stars
721
Forks
117
Last commit
1 month ago
Last verified
Aug 28, 2026
Metadata fetched
Aug 28, 2026
Repository age
2 years
License
Apache-2.0

Project health auto-fetched from the source repository.

Maintain this resource?

Review this source-backed profile, send a correction with evidence, or link to it from your documentation. Claims verify your relationship to the project; profile facts still require source evidence and editorial review.

Alternatives to cpu-kernels