Agent skills
Skills you can use with AI coding agents, indexed from public GitHub repositories.
-
warp-primitives
Warp-level programming and SIMD optimization. Use warp shuffle instructions, voting functions, cooperative groups, warp-synchronous algorithms, and minimize warp divergence for optimal GPU performance.
a5c-ai/babysitter 514
-
tensorrt-optimization
NVIDIA TensorRT model optimization and deployment. Convert models to TensorRT engines, configure optimization profiles and precision modes, apply INT8 calibration, analyze kernel fusion, generate custom plugins, and profile inference performance.
a5c-ai/babysitter 514
-
parallel-patterns
GPU parallel algorithm design patterns and implementations. Implement parallel reduction, scan/prefix sum, histogram, parallel sort algorithms, stream compaction, and work-efficient patterns optimized for specific GPU architectures.
a5c-ai/babysitter 514
-
nsight-profiler
Expert skill for NVIDIA Nsight Systems and Nsight Compute profiling tools. Configure profiling sessions, analyze kernel reports, interpret occupancy metrics, roofline model data, memory bandwidth bottlenecks, and warp execution efficiency.
a5c-ai/babysitter 514
-
hip-rocm
AMD HIP and ROCm ecosystem for cross-platform GPU development. Execute hipify conversion tools, generate HIP-compatible kernel code, handle CUDA/HIP API differences, configure ROCm toolchain, and profile with rocprof.
a5c-ai/babysitter 514
-
cuda-graphs
Expert skill for CUDA Graph capture and optimization for reduced launch overhead. Capture CUDA operations into graphs, instantiate and execute graph instances, update graph node parameters, profile graph vs stream execution, design graph-friendly kernel patterns, and optimize launch latency for inference.
a5c-ai/babysitter 514
-
unified-memory
Expert skill for CUDA Unified Memory and memory prefetching optimization. Configure managed memory allocations, implement memory prefetch strategies, handle page fault analysis, configure memory hints and advise, profile unified memory migration, optimize for oversubscription scenarios, and compare managed vs explicit memory.
a5c-ai/babysitter 514
-
opencl-runtime
Cross-vendor OpenCL runtime management and kernel development. Query platforms/devices, generate portable OpenCL C kernel code, handle vendor-specific extensions, manage contexts and command queues, compile and cache programs.
a5c-ai/babysitter 514
-
cublas-cudnn
Expert integration with NVIDIA GPU-accelerated math libraries. Configure cuBLAS tensor core operations, generate cuBLAS GEMM calls, integrate cuDNN layers, handle algorithm selection, and support mixed-precision operations.
a5c-ai/babysitter 514
-
vulkan-compute
Vulkan compute shader development and pipeline configuration. Generate GLSL/HLSL compute shaders, compile to SPIR-V, configure compute pipelines, manage descriptor sets and resource bindings, implement memory barriers and synchronization.
a5c-ai/babysitter 514
-
cuda-toolkit
Deep integration with NVIDIA CUDA toolkit for kernel development, compilation, and debugging. Execute nvcc compilation with optimization flags analysis, generate and validate CUDA kernel code, analyze PTX/SASS assembly output, and configure execution parameters.
a5c-ai/babysitter 514
-
nccl-communication
NVIDIA Collective Communications Library integration for multi-GPU operations. Initialize NCCL communicators, execute collective operations, configure communication topologies, profile collective performance, and support RCCL for AMD compatibility.
a5c-ai/babysitter 514
-
cuda-debugging
Expert skill for GPU debugging using CUDA-GDB and NVIDIA Compute Sanitizer. Detect memory errors, race conditions, uninitialized memory access, validate atomic operations, analyze kernel synchronization issues, and generate debugging reports with recommendations.
a5c-ai/babysitter 514
-
gpu-benchmarking
Expert skill for automated GPU performance benchmarking and regression detection. Design micro-benchmarks, measure kernel execution time with CUDA events, calculate achieved vs theoretical performance, generate comparison reports, detect regressions in CI/CD, and profile power/thermal characteristics.
a5c-ai/babysitter 514
-
FFI Design
Expert skill for designing and implementing foreign function interfaces to native code
a5c-ai/babysitter 514
-
Lexer Generator
Expert skill for generating and hand-writing lexers using DFA-based, table-driven, and recursive approaches
a5c-ai/babysitter 514
-
generics-implementation
Expert skill for implementing parametric polymorphism including type parameter bounds, monomorphization, type erasure, variance, higher-kinded types, and associated types.
a5c-ai/babysitter 514
-
AST Manipulation
Expert skill for abstract syntax tree design, traversal, transformation, and manipulation patterns
a5c-ai/babysitter 514
-
pattern-matching
Expert skill for implementing pattern matching including exhaustiveness checking, decision tree compilation, and efficient match dispatch code generation.
a5c-ai/babysitter 514
-
Grammar Design
Expert skill for formal grammar design including disambiguation, precedence, and validation
a5c-ai/babysitter 514
-
Memory Allocator
Expert skill for custom memory allocator design optimized for language runtime needs
a5c-ai/babysitter 514
-
Error Messages
Expert skill for designing and implementing high-quality compiler error messages
a5c-ai/babysitter 514
-
Concurrency Primitives
Expert skill for implementing language-level concurrency support including threads, channels, and synchronization
a5c-ai/babysitter 514
-
Language Server Protocol
Expert skill for implementing Language Server Protocol servers with full IDE feature support
a5c-ai/babysitter 514