← All skills

pipeline-gpu-kernel

95 / 100 · Low risk

Fine to try.

By roohe/agentic-super-skills (GitHub) · other · 10 stars · scanned at commit f483e54

Apply software pipelining (double-buffering) to a tiled GPU kernel to overlap global memory loads with Tensor Core computation. Covers prologue/loop/epilogue restructuring, LDG-register vs cp.async (LDGSTS) variant selection based on compute/load ratio, shared memory budget verification against architecture-specific occupancy cliffs, and SASS-level verification of load/compute overlap.

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'pipeline-gpu-kernel' ] || [ -L ~/.agents/skills/'pipeline-gpu-kernel' ]; then echo 'A skill folder named pipeline-gpu-kernel already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/roohe/agentic-super-skills' "$d" && git -C "$d" checkout --quiet 'f483e54f3ad5f202a77d45f2e57f8d970b71195f' -- 'skills_library/pipeline-gpu-kernel' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills_library/pipeline-gpu-kernel' ~/.agents/skills/'pipeline-gpu-kernel' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'pipeline-gpu-kernel' ~/.claude/skills/'pipeline-gpu-kernel' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for roohe/agentic-super-skills shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/roohe/agentic-super-skills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/roohe/agentic-super-skills/)

Score history