性能
性能分析与优化
videocut:高清化
Ceeon
视频高清导出。2-pass编码+锐化，匹配或超越原片画质。触发词：高清化、高清导出、导出高清、渲染高清
hyperflow-status
jeremylongshore
Hyperflow project status. Use to see current hyperflow state — "what is hyperflow doing", "show task progress", "where are we". Read-only — reports in-flight tasks, memory count, and progress. Never modifies state or runs work.
hyperflow-cache
jeremylongshore
Hyperflow memory manager. Use to view, search, add, edit, prune, or clear hyperflow project memory — "show memory", "search memory for X", "clear memory", "what does hyperflow remember about Y". CRUD over .hyperflow/memory/ only — never touches source code.
chrome-extension-development
Mindrally
Expert guidelines for Chrome extension development with Manifest V3, covering security, performance, and best practices
business-central-development
Mindrally
Expert guidance for Microsoft Dynamics 365 Business Central development using AL language and extensions
angular-development
Mindrally
Expert guidance for Angular and TypeScript development focused on scalable, high-performance web applications
angular
Mindrally
Expert in Angular TypeScript development with scalable, high-performance patterns
data-analyst
Mindrally
Data analysis best practices with pandas, numpy, matplotlib, seaborn, and Jupyter notebooks.
data-analysis-jupyter
Mindrally
Expert guidance for data analysis, visualization, and Jupyter Notebook development with pandas, matplotlib, seaborn, and numpy.
c-sharp
Mindrally
Guidelines for C# development including Blazor, Unity game development, and .NET backend best practices
android-development
Mindrally
Android development guidelines for Kotlin with clean architecture, MVI pattern, Material Design, and best practices for building robust mobile applications
deep-learning
Mindrally
Comprehensive deep learning guidelines for neural network development, training, and optimization.
astro
Mindrally
Expert in Astro framework with static generation and partial hydration patterns
huggingface-tokenizers
NousResearch
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
peft-fine-tuning
NousResearch
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
huggingface-tokenizers
NousResearch
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
optimizing-attention-flash
NousResearch
Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.
llama-cpp
NousResearch
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
peft-fine-tuning
NousResearch
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
vercel-react-native-skills
vercel-labs
React Native and Expo best practices for building performant mobile apps. Use
Executive Resume Writer
Paramchoudhary
Create C-suite and VP level resumes emphasizing strategic leadership
Resume Bullet Writer
Paramchoudhary
Transform weak resume bullets into achievement-focused statements with metrics and impact
prospect-research
Salesably
Creates comprehensive prospect profiles and knowledge capsules for sales preparation. Use this skill when researching new prospects, preparing for calls, onboarding to new accounts, or updating prospect intelligence.
brewcode:text-optimize
kochetkov-ma
Optimizes text and docs for LLM token efficiency. Use when - optimizing prompts, reducing tokens, compressing text, condensing verbose content. Trigger keywords - optimize, reduce tokens, compress, condense, slim, tighten, too verbose, shrink.