123 skills · sorted by relevance
Implement and troubleshoot large-model training with DeepSpeed across multiple GPUs. The skill covers memory optimization, parallel training, and numerical precision choices, alongside examples for storage transfers and models that divide work among specialized components.
Helps configure and troubleshoot DeepSpeed for training large models across multiple devices. Covers memory optimization, parallel training, and DeepNVMe storage transfers, with examples for tuning data movement and configuring models that divide work among specialized components.
Train Mixture of Experts models, which activate selected specialist networks for each input. The skill guides architecture, input routing, workload balancing, and GPU distribution using DeepSpeed or HuggingFace, with examples for training configuration and inference optimization.
Turn speedup requests into a measured optimization process with limits on time, cost, and variants. Compare each change against a baseline and correctness checks, then retain the fastest safe option with reproducible commands and measurements.
Integrate Flash Attention into transformer models to reduce processing time and GPU memory use for long sequences. Covers PyTorch integration, advanced attention options, and benchmarks that compare speed, memory use, and output accuracy.
Build and train Mixture of Experts models, which activate selected specialist networks for each input. Use DeepSpeed or HuggingFace workflows to configure routing, balance workloads, distribute experts across GPUs, and optimize inference.
Measure and optimize claude-flow v3 startup, search, attention processing, and memory use. Compare benchmarks against performance targets, track changes over time, and identify regressions to guide further optimization.
Measure and optimize claude-flow v3 startup, search, attention processing, and memory use. Compare benchmarks against performance targets, track changes over time, and identify regressions to guide further optimization.
Expert at diagnosing and fixing performance bottlenecks across the stack. Covers Core Web Vitals, database optimization, caching strategies, bundle optimization, and performance monitoring. Knows when to measure vs optimize. Use when "slow page load, performance optimization, core web vitals, bundle size, lighthouse score, database slow, memory leak, optimize performance, speed up, reduce load time, performance, optimization, core-web-vitals, caching, profiling, bundle-size, database" mentioned.
The English summary is being prepared. Showing the original description.
Analyze GTmetrix (and PageSpeed) reports for WordPress sites and produce a prioritized optimization plan, a custom WPCode-ready PHP optimization snippet, and a self-contained HTML summary report. Covers PDF report extraction with pypdf, live-page cross-referencing, WordPress-specific fix mapping (Elementor, WooCommerce, Breeze, reCAPTCHA, Google-tag duplication, WebP conversion), and staging-test discipline. Use when asked to review GTmetrix reports, optimize page speed or Core Web Vitals, create a custom performance optimization snippet, or produce an optimization report for a WordPress site.
The English summary is being prepared. Showing the original description.
Integrate Flash Attention into Transformer models to speed up attention, the calculation that relates input elements, while reducing GPU memory use. Covers PyTorch integration, performance measurement, output comparison, and a separate H100 optimization path.
Tune AgentDB memory use and search speed through quantization, which stores vectors more compactly, search indexes, caching, and batch operations. Includes configuration examples and benchmarks to help balance accuracy, speed, and storage as datasets grow.
Tune AgentDB memory use and search speed through quantization, which stores vectors more compactly, search indexes, caching, and batch operations. Includes configuration examples and benchmarks to help balance accuracy, speed, and storage as datasets grow.
Optimize transformer attention, the calculation that relates tokens to one another, using Flash Attention. The skill guides PyTorch integration, GPU memory reduction, performance measurement, and output comparisons for training or inference with long sequences.
Orchestrate Xcode build optimization by benchmarking first, running the specialist analysis skills, prioritizing findings, requesting explicit approval, delegating approved fixes to xcode-build-fixer, and re-benchmarking after changes. Use when a developer wants an end-to-end build optimization workflow, asks to speed up Xcode builds, wants a full build audit, or needs a recommend-first optimization pass covering compilation, project settings, and packages.
The English summary is being prepared. Showing the original description.
Diagnose and fix web performance issues including Core Web Vitals (LCP, INP, CLS), bundle size, asset optimization, render performance, and runtime efficiency. Use this skill whenever the user wants to improve page speed, fix Core Web Vitals, optimize assets, reduce bundle size, debug slow renders, or systematically improve a site's performance. Triggers on performance, page speed, Core Web Vitals, LCP, INP, CLS, FID, TTFB, bundle size, code splitting, image optimization, lazy loading, render blocking, slow page, performance audit, Lighthouse score. Also triggers when traffic or conversion is dropping due to perceived slowness.
The English summary is being prepared. Showing the original description.
Landing page audit and optimization for conversion. Covers above-the-fold design, value propositions, CTAs, social proof placement, form design, page speed, and mobile optimization. Use when the user asks about landing page optimization, conversion rate improvement, page audits, or CTA optimization.
The English summary is being prepared. Showing the original description.
Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention
The English summary is being prepared. Showing the original description.
Audit web pages for performance optimization following PageSpeed Insights guidelines. Use when analyzing page performance, optimizing web applications, reviewing performance metrics, implementing Core Web Vitals improvements, or when the user mentions page speed, performance optimization, Lighthouse scores, or Core Web Vitals.
The English summary is being prepared. Showing the original description.
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.
The English summary is being prepared. Showing the original description.
Audit a website for problems affecting search visibility, from search engine access and indexing to speed, page titles, content, and links. Receive evidence-based findings and a prioritized action plan, including checks for multilingual and local business sites.
Plan website improvements for organic search traffic through keyword research, content structure, page metadata, and technical checks. Get guidance on internal links, structured data that describes page content, loading speed, and ongoing performance monitoring.
Helps speed up LLM text generation by comparing methods that predict and verify multiple tokens at once. Covers draft model selection, Medusa prediction heads, Lookahead Decoding settings, and deployment examples for inference services.
Reduce LLM generation delays with speculative decoding, which proposes tokens ahead of verification. This skill compares small draft models, Medusa, and Lookahead Decoding, with implementation examples, tuning guidance, and deployment using vLLM.