72 tagged cuda, measured the same way as everything else here.
Browse within: moe 63PyTorch 39LLM 37ai-coding 35gpu-kernels 35high-performance-computing 35inference 35kernel-optimization 35llm-agents 35ffmpeg 32llama 32onnx 32transformer 32blackwell 31
Skill Claude CodeCodex
Triage GitHub issues, review contributor pull requests, diagnose CI, and prepare explicitly requested releases for VoiceStudio's Electron desktop app and Python backend. Use for repository maintenance work, not ordinary VoiceStudio audio generation.
Skill Claude CodeCodex needs its repo
Use when choosing the fastest SGLang Diffusion flags for a model, GPU, and VRAM budget.
Skill Claude Code ✓ vendor
Launch and operate this repo's perf-optimize workflow, which iteratively APPLIES TensorRT-LLM serving optimizations — baseline benchmark at one concurrency or a Pareto curve of them (tok/s/user vs tok/s/gpu), analytical SOL projection on by default (via the internal-perf-sol-analysis skill) sizing the headroom the…
Skill Claude CodeCodex needs its repo
Use this guide when importing a new kernel-team drop or adapting FlashInfer to changes in that drop. VENDOR.md records the pinned source and export exceptions; TUNING.md covers tuning and measurements.
ForceInjection/AI-fundamentals
Skill Claude CodeCodex
Extract text from PDF files, translate it to a target language, and save the result as a Markdown file. Use this skill when the user wants to translate a PDF document or asks to "convert PDF to Chinese".
maxiaosong1124/ncu-cuda-profiling-skill
Skill Claude CodeCodex
Automated NCU (Nsight Compute) profiling workflow with full metrics collection and persistent storage.
Skill Claude CodeCodex
Add a tocstring function for a C++ enum type in the Celeritas codebase. Use when asked to add string conversion, tocstring, or enum-to-string functionality for any enum class. Follows the standard EnumStringMapper pattern used throughout the codebase.
Skill Claude CodeCodex
Guide the agent through implementing a single Triton kernel that computes y = rmsnorm(x + residual) while also writing back x + residual for the next transformer block's residual stream. This fusion is the dominant pattern in LLaMA, Mistral, Qwen, and similar decoder blocks: every attention sub-block and every MLP…
ForceInjection/cuda-code-skill
Skill Claude CodeCodex
CUDA kernel development, debugging, performance optimization, linear algebra, and multi-GPU communication for Claude Code. Use when writing, debugging, or optimizing CUDA code, GPU kernels, parallel algorithms, or CUDA library calls. Covers cuBLAS/cuBLASLt GEMM operations, CUDA Math API (half, bfloat16, FP8, FP6…
Skill Claude CodeCodex
NCU-driven iterative optimization workflow for CUDA/CUTLASS/Triton/CuTe DSL kernels. MANDATORY: every optimization MUST start with NCU profiling, followed by multi-dimensional analysis, then targeted code modification, then re-profiling to verify. Supports roofline, memory hierarchy, warp stalls, instruction mix…
staticroostermedia-arch/engram
Skill Claude CodeCodex
Engram geometric memory for AI agents — slash commands map decision moments to MCP rituals. You are the primary user. Lean highway + power escalation.
Skill Claude CodeCodex
Local speech-to-text using faster-whisper. 4-6x faster than OpenAI Whisper with identical accuracy; GPU acceleration enables 20x realtime transcription. SRT/VTT/TTML/CSV subtitles, speaker diarization, URL/YouTube input, batch processing with ETA, transcript search, chapter detection, per-file language map.
troycheng/cuda-kernel-optimizer
Skill Codex
Use when optimizing or diagnosing GPU workload performance with CUDA, CUTLASS, Triton, PyTorch, vLLM or TensorRT-LLM; interpreting NCU, Nsys or PyTorch Profiler evidence; or preparing the workload and correctness checks for such optimization. General coding, skill maintenance and conceptual GPU questions do not…
Skill Claude CodeCodex
Use when optimizing MLIR GPU performance, compiling PyTorch or linalg/tensor MLIR to PTX, writing Transform dialect schedules, tuning auxiliary pass pipelines, benchmarking MLIR kernels against PyTorch, or diagnosing linalg/torch-mlir GPU codegen.
Skill Claude CodeCodex
Profile inside a CuTe DSL (cutlass.cute) kernel with IKET (In-Kernel Event Tracing) and the run-iket profiler. Use when developing or tuning a CuTe DSL kernel and you need per-warp timelines of kernel phases (setup, mainloop, epilogue), producer/consumer (TMA/MMA) activity, or time spent in pipeline and mbarrier…
Skill Claude CodeCodex
Use only when the user explicitly mentions "rapids-first" or "RAPIDS first", to prefer the RAPIDS GPU stack (cudf, cuml, cupy, cupyx) and its zero-code accelerators over pandas, scikit-learn, numpy and scipy when writing Python data-science, ML, ETL or numerical code.
2233admin/performance-patterns-skill
Skill Codex
Detect, triage, and fix performance problems across Windows native C/C++ (clang-cl/MSVC; VTune, AMD uProf, ETW/WPA), WSL Linux, native Linux (perf/flamegraphs/GCC/Clang), and CUDA/NVIDIA environments (nvidia-smi, nvcc, Nsight Systems, Nsight Compute). Invoke when the user asks to optimize or debug performance, review…
Skill Codex
Create slides with the Prompt-as-Slide (PasS) protocol by defining each slide as one authoritative, self-contained Markdown prompt, generating one aligned slide image per prompt with an available image-generation capability, deriving synchronized speaker notes with target durations and bold visual anchors, and…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: