Prepare or review NVSHMEM CUDA kernels for TMA SMEM registration and direct-SMEM transfers. Do not use for unrelated CUDA tuning.
Install
npx skills add NVIDIA/skills --skill nvshmem-enable-tmanpx skills add NVIDIA/skills --skill nvshmem-enable-tma -a claude-code -gnpx skills add NVIDIA/skills --skill nvshmem-enable-tma -a codex -ggemini skills install https://github.com/NVIDIA/skills.git --path skills/nvshmem-enable-tmagh skill install NVIDIA/skills nvshmem-enable-tmagit clone --depth 1 https://github.com/NVIDIA/skills /tmp/NVIDIA__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/NVIDIA__skills/skills/nvshmem-enable-tma ~/.claude/skills/nvshmem-enable-tma # Claude Code
cp -R /tmp/NVIDIA__skills/skills/nvshmem-enable-tma ~/.agents/skills/nvshmem-enable-tma # Codex, Cursor, Gemini CLI, Copilot, OpenCode