Convert Objective-C code to idiomatic Swift. Use when migrating Objective-C projects to Swift, translating Objective-C patterns to modern Swift, modernizing legacy codebases, or establishing gradual migration strategies with interop. Extends meta-convert-dev with Objective-C-to-Swift specific patterns including bridging and @objc attributes.
Convert Erlang code to idiomatic Roc. Use when migrating Erlang projects to Roc, translating BEAM/OTP patterns to functional patterns, or refactoring Erlang codebases. Extends meta-convert-dev with Erlang-to-Roc specific patterns.
Use when optimizing Swift code performance, reducing memory usage, improving runtime efficiency, dealing with COW, ARC overhead, generics specialization, or collection optimization
Update a Raycast extension to the latest version
Foundation layer for identity verification, noise filtering, and environment awareness.
Use this skill when working with the Crystal programming language. Crystal is a statically-typed, compiled language with Ruby-like syntax. It features type inference, null safety, macros, and compiles to efficient native code. Use for understanding Crystal's standard library, syntax, semantics, concurrency model, and FFI bindings.
Use when optimizing agent context, reducing token costs, implementing KV-cache optimization, or asking about "context optimization", "token reduction", "context limits", "observation masking", "context budgeting", "context partitioning"
Use when "training LLM", "finetuning", "RLHF", "distributed training", "DeepSpeed", "Accelerate", "PyTorch Lightning", "Ray Train", "TRL", "Unsloth", "LoRA training", "flash attention", "gradient checkpointing"
Optimize training performance for Kaggle competitions. Use PROACTIVELY when training is slow, GPU memory issues occur, or user asks about performance tuning. Keywords: 遅い, 最適化, OOM, CUDA, batch_size, num_workers, 高速化, パフォーマンス
Embedding backends (InsightFace/PyTorch+ONNXRuntime vs TensorRT). Use when optimizing embedding throughput or debugging drift/fallbacks.
Optimize CoreML models for iOS and macOS deployment. Covers quantization, palettization, pruning, Neural Engine targeting, compute unit selection, and performance profiling. Use when converting ML models to CoreML, optimizing model size/latency, debugging Neural Engine issues, or benchmarking on-device inference.
Computer vision library for PyTorch featuring pretrained models, advanced image transforms (v2), and utilities for handling complex data types like bounding boxes and masks. (torchvision, transforms, tvtensor, resnet, cutmix, mixup, pretrained models, vision transforms)
Rust performance optimization and zero-cost abstractions
Modern C++ (C++23/C++20) development specialist covering RAII, smart pointers, concepts, ranges, modules, and CMake. Use when developing high-performance applications, games, system software, or embedded systems.
Outline extension implementing protocol for plugin development domain
Create custom adapters for Rimitive view primitives. Use when building renderers for Canvas, WebGL, terminal, native platforms, or any tree-based target.
Generate production-quality Python code for sensor integration, sensor fusion, and perception systems that run in Pyodide (browser). Creates educational code for teaching IMU sensors, LiDAR simulation, depth perception, tactile sensors, and sensor fusion algorithms. Focuses on sensor data processing, calibration, noise modeling, and multi-sensor fusion for educational purposes.
Fixes 'multiple definition' linker errors by removing duplicate symbols.
Metasuperhypergraph orchestrator with fractal scale-invariant recursion. Activates maximum performance mode through renormalization-group optimization, power-law resource allocation, and homoiconic self-improvement. Triggers on "ultrawork", complex multi-domain tasks, or when parallelization/delegation yields superlinear returns.
Provide architecture-specific GPU optimization advice for NVIDIA GPUs (Ampere, Hopper, Ada). Use when discussing GPU architecture features, compute capability, or hardware-specific optimizations.
Advanced memory management techniques for the OpenCode memory system
Architecture and implementation details of FastEdgesGeometry. Use when modifying FastEdgesGeometry, understanding its optimization techniques, or debugging edge generation issues.
