add-sgl-kernel Skill
Step-by-step tutorial for adding a heavyweight AOT CUDA/C++ kernel to sgl-kernel (including tests & benchmarks) Published by microsoft in llm-42.
Decision snapshot
Is this a fit?
Testing, Includes SKILL.md, Reusable instructions
Compatibility not yet detected.
Permission behavior not yet detected.
Copy skill directory
2 months ago · Apache-2.0 license
No specific cautions were detected. Review the source and requested permissions before installing.
What is add-sgl-kernel Skill?
Step-by-step tutorial for adding a heavyweight AOT CUDA/C++ kernel to sgl-kernel (including tests & benchmarks) Published by microsoft in llm-42. 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.
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
- Testing
- Testing use cases
Declared skill metadata
- Source file: .claude/skills/add-sgl-kernel/SKILL.md
These fields retain source and confidence evidence from the indexed SKILL.md.
Compatibility and setup
- Install or run with Copy skill directory
When to use add-sgl-kernel Skill
- Use it for testing.
Built with
Editorial notes
Source
- Creator: microsoft
- Repository: microsoft/llm-42
- Skill file: .claude/skills/add-sgl-kernel/SKILL.md
What it does
Step-by-step tutorial for adding a heavyweight AOT CUDA/C++ kernel to sgl-kernel (including tests & benchmarks)
Skill instructions
Tutorial: Adding a New Kernel to sgl-kernel (AOT / Heavyweight) This SKILL is a step-by-step guide for adding a heavyweight CUDA/C++ kernel to sgl-kernel/. Typical characteristics: - Depends on heavyweight components such as CUTLASS / FlashInfer / DeepGEMM / sgl-attn - Needs AOT build and distribution (wheel / torch extension), so build time, link flags, CUDA arch targets, and binary size matter - Exposed as a stable sglkernel API and used by higher-level code (including torch.compile) Two rules of thumb (must follow) 1. Heavyweight kernels go to sgl-kernel. If it depends on CUTLASS/FlashInfer/DeepGEMM (or similarly heavy stacks), implement it in sgl-kernel/. 2. Lightweight kernels go to python/sglang/jitkernel. If it is small, has few dependencies, and benefits from rapid iteration, implement it as a JIT kernel instead. In addition, every new kernel must ship with: - Tests (pytest) - A benchmark script (triton.testing) --- Goal Add a new kernel end-to-end, including: - CUDA/C++ implem
Verified compatibility and discovery
Frequently asked questions
What is add-sgl-kernel?
add-sgl-kernel is a open-source AI agent skill with Copy skill directory. Step-by-step tutorial for adding a heavyweight AOT CUDA/C++ kernel to sgl-kernel (including tests & benchmarks)
Who is add-sgl-kernel best for?
add-sgl-kernel is best for reusing agent instructions, scripts, and references, testing workflows.
How do I install add-sgl-kernel?
Install or run add-sgl-kernel using Copy skill directory. Check add-sgl-kernel for the latest setup command.
Is add-sgl-kernel actively maintained?
add-sgl-kernel may need a closer maintenance check before production use.
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.
