wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Add a new cuTile GPU kernel operator to TileGym. Covers dispatch registration in ops.py, cuTile backend implementation, init.py exports, test creation, and benchmark in tests/benchmark. Use when adding, creating, or implementing a new cuTile operator/kernel in TileGym, or when asking how to register a new cuTile op.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through diagnosing and restructuring CUDA global memory access patterns to maximize effective memory bandwidth, covering coalescing requirements, vectorized loads, AoS vs SoA layouts, shared memory staging for non-coalesced patterns, and L2 cache behavior.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through choosing and tuning kernels for the prefill phase versus the decode phase of LLM inference. The two phases have fundamentally different arithmetic intensity, occupy different sides of the roofline, and respond to different optimizations. Continuous batching and speculative decoding shift the…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through designing and tuning shared memory tiling strategies for CUDA kernels, covering bank conflict analysis and elimination, tile shape selection, double buffering with async copy, occupancy tradeoffs from shared memory allocation, and the decision of when smem tiling is worth the complexity.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through the systematic process of choosing and tuning block size parameters in Triton kernels — BLOCKM, BLOCKN, BLOCKK for GEMM-style kernels; BLOCKSIZE for reduction and pointwise kernels; and the associated numwarps and numstages values that control parallelism and pipeline depth. This is an…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
How to optimize TileLang GPU kernels for better performance. Use this skill whenever the user has a working TileLang kernel that is too slow, wants to tune tile sizes or pipeline stages, needs to improve TFLOPS or bandwidth utilization, wants to use the AutoTuner, or asks questions like "how do I make my kernel…
wenyi-li/awesome-agent-kernel-skills
Skill Claude Code
Compile-time optimization guidance for HIP/C++ kernels using opus.hpp. Use when writing or reviewing OPUS kernels, analyzing compile time, reducing template instantiation overhead, or optimizing hipcc build performance.
wenyi-li/awesome-agent-kernel-skills
Skill Claude Code
Module-level JIT build-wall optimization for opus-based aiter modules. Use when an aiter JIT module's first-call build wall is a user-visible bottleneck or when adding a new module.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Performance analysis coordination workflow. Guides profiling delegation, bottleneck classification (compute/memory/launch/communication/sync), and structured report generation. Use when the user asks to analyze performance, profile a workload, check MFU/SOL, or diagnose bottlenecks.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Analyze host/CPU overhead in TensorRT-LLM inference from nsys traces. Detect whether host overhead is the bottleneck using GPU idle ratio, host prep exposed ratio, and per-phase evidence. For regressions, isolate forward steps via allreduce/NVTX patterns, compare host operation breakdowns across versions, and identify…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Profiles and optimizes TensorRT-LLM host/CPU overhead using lineprofiler (with nsys support planned). Runs iterative profile-analyze-optimize-validate rounds. Use when GPU utilization is low or optimizing PyExecutor throughput.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Analyze ncu (NVIDIA Nsight Compute) profiling output: SOL% bottleneck classification, roofline analysis, occupancy diagnosis, memory hierarchy analysis, warp stall analysis, metric interpretation, and programmatic .ncu-rep report analysis. NOT for kernel writing or code generation, Nsight Systems (nsys), host-side…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Nsight Systems (nsys) CLI for system-level timeline profiling. Use when the user wants to run nsys profile, analyze .nsys-rep reports, use nsys stats/analyze/recipe commands, diagnose GPU idle time from timeline traces, or profile distributed training with NCCL overlap analysis. NOT for kernel-level metrics like SOL%…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Performance optimization coordination playbook. Contains specialist routing table, TileIR two-step pipeline, kernel generation specialist selection, prioritization criteria, and safe modification workflow. Use when the user asks to apply optimizations, write kernels, or improve performance. Covers both user-specified…
wenyi-li/awesome-agent-kernel-skills
Skill Claude Code
Run accuracy benchmarks (FlagEval, when available) and performance benchmarks (vllm bench serve) against a served model. Covers 5 workload profiles: short/long prefill x short/long decode + high concurrency. Collects throughput, latency, TTFT, TPOT metrics.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Apply CUDA Graphs to PyTorch workloads — API selection (torch.compile, PyTorch makegraphedcallables, TE makegraphedcallables, MCore CudaGraphManager, FullCudaGraphWrapper, manual torch.cuda.graph), code compatibility, capture workflows, dynamic pattern handling, and troubleshooting. Triggers: CUDA graph…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Identify and eliminate host-device synchronizations in PyTorch code. Detects sync points (.item(), .cpu(), boolean indexing, torch.tensor on CUDA), classifies false vs true dependencies, provides sync-free alternatives. Triggers: sync-free, synchronization, .item(), .cpu(), host-device sync, eliminate syncs, CPU…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup, per-iteration statistics, and anti-pattern avoidance. Also covers NVTX…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through translating a CUDA kernel to AMD HIP for ROCm-compatible hardware (MI250, MI300, RDNA), preserving correctness and performance intent while adapting to the HIP execution model, memory model, and AMD-specific toolchain.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Guide the agent through systematically porting an existing CUDA kernel to Triton, mapping the CUDA execution model to Triton's tile-based program model, preserving numerical correctness, and identifying the patterns that do not translate directly.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
How to benchmark and profile TileLang GPU kernels for performance measurement. Use this skill whenever the user wants to measure kernel latency, compare TileLang performance against PyTorch/cuBLAS/Triton, use dobench or getprofiler, calculate TFLOPS or bandwidth, run Nsight Compute or Nsight Systems on a TileLang…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
CUDA kernel development, debugging, and performance optimization for Claude Code. Use when writing, debugging, or optimizing CUDA code, GPU kernels, or parallel algorithms. Covers non-interactive profiling with nsys/ncu, debugging with cuda-gdb/compute-sanitizer, binary inspection with cuobjdump, and performance…
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
This skill should be used when optimizing PyTorch models and kernels, including efficient tensor operations, torch.compile, custom autograd/CUDA/Triton extensions, mixed precision, memory and data pipeline tuning, model optimization techniques, CUDA graphs, and profiling.
wenyi-li/awesome-agent-kernel-skills
Skill Claude CodeCodex
Debug ROCm/HIP kernel crashes in SGLang and vLLM on AMD GPUs (MI300X/MI325X/MI355X). Adapts SGLang's @debugkernelapi kernel boundary logging to ROCm: captures input tensors before crash, tracks shapes/dtypes/values, dumps crash artifacts for offline analysis. Integrates with amdpilot executor failurereason field 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: