107 skills · sorted by relevance
Configure LLM inference and serving on NVIDIA GPUs with TensorRT-LLM. Covers quantization, which reduces numerical precision, batching requests during generation, and distributing models across GPUs to improve processing capacity and response times.
Configure LLM inference and serving on NVIDIA GPUs with TensorRT-LLM. Covers quantization, which reduces numerical precision, batching requests during generation, and distributing models across GPUs to improve processing capacity and response times.
Collect weekly snack requests from Slack, combine duplicate items with standing staples, and prepare a vendor cart within budget. Share the draft for the office manager's written approval before placing the order.
Patterns for managing MSBuild item groups: Include/Remove/Update semantics, item metadata, batching with %(Metadata), transforms, per-item filtering, and cross-product batching pitfalls. Only activate in MSBuild/.NET build context. USE FOR: diagnosing and fixing item group anti-patterns in .csproj files, reviewing item management for correctness, fixing CS2002 duplicate file warnings from SDK globbing, fixing targets that run more times than expected due to cross-product batching, fixing Include vs Update misuse on SDK-globbed items, fixing FileWrites registration for generated file clean support, moving generated files to IntermediateOutputPath. DO NOT USE FOR: target chain architecture (use target-authoring), property patterns (use property-patterns), incrementality (use incremental-build), general anti-patterns (use msbuild-antipatterns), non-MSBuild build systems.
The English summary is being prepared. Showing the original description.
Improve GSAP animation performance by reviewing movement properties, batching page measurements and updates, and stopping unused animations. It guides changes for smoother motion, including efficient handling of long lists, pointer tracking, and scroll-driven effects.
Set up Phoenix tracing for LLM applications in Python or TypeScript. The skill guides automatic and custom recording of individual operations, conversation grouping, metadata, feedback, and production settings such as batching and personal data masking.
Deploy language models with vLLM through an OpenAI-compatible API or offline batch processing. Tune response speed, throughput, and GPU memory use with quantization, which reduces numerical precision, and execution across multiple GPUs.
Deploy language model APIs and run batch generation with vLLM. Configure GPU memory use, quantization—lower precision model storage—and distribution across GPUs, then test response times and monitor how many requests the server handles.
Assess GraphQL APIs during penetration tests, from finding endpoints and mapping data structures to checking access controls, query limits, subscriptions, and uploads. Includes example queries, guidance for testing tools, and defensive controls for interpreting findings.
Compare batching, request scheduling, and memory management for a single model serving instance. Assess concurrent request capacity, response times, generation speed, and GPU time per result that meets the required quality and deadline.
Own concrete MSBuild ItemGroup and item-expression questions. USE FOR: Include, Remove, Update, item identity and metadata, transforms, filtering, batching, duplicate or overlapping items, and reviews that verify whether those operations are correct. Generated items stay in scope when the central defect is item identity, metadata, batching, duplicate declarations, or glob/Remove/Update semantics. For a generated artifact missing from compilation or output, wrong target timing/path, or FileWrites clean tracking without a broader item-semantics defect, use including-generated-files. The item operation may be broken, suspected, or already correct. Exclude property-only issues, general incrementality with no item question, broad reviews with no concrete item concern, and non-MSBuild systems.
The English summary is being prepared. Showing the original description.
Find and fix game performance problems methodically — measure with the engine profiler first, reason about the frame-time budget, locate the CPU-vs-GPU bottleneck, then apply the right fix: object pooling, draw-call batching, fewer allocations/GC spikes, and asset budgets. Engine- neutral method that pairs with each engine's profiler. Use when the user mentions performance, optimize, low/dropping FPS, frame drops, stutter, lag, profiler, frame budget, draw calls, batching, garbage collection/GC spikes, object pooling, or "the game runs slow".
The English summary is being prepared. Showing the original description.
Treat context switches as the budget line they are — the switch census (how fragmented the week really is), the batching moves that consolidate scattered same-kind work, the calendar defrag that turns Swiss cheese into slabs, and the switch-cost line for saying no. Use when asked my day is fragmented to death, count my context switches, batch my meetings and reviews, or defend against calendar Swiss cheese. Produces the fragmentation census, the batching plan, the defrag moves, and the protective phrases.
The English summary is being prepared. Showing the original description.
Own concrete MSBuild ItemGroup and item-expression questions. USE FOR: Include, Remove, Update, item identity and metadata, transforms, filtering, batching, duplicate or overlapping items, and reviews that verify whether those operations are correct. Generated items stay in scope when the central defect is item identity, metadata, batching, duplicate declarations, or glob/Remove/Update semantics. For a generated artifact missing from compilation or output, wrong target timing/path, or FileWrites clean tracking without a broader item-semantics defect, use including-generated-files. The item operation may be broken, suspected, or already correct. Exclude property-only issues, general incrementality with no item question, broad reviews with no concrete item concern, and non-MSBuild systems.
The English summary is being prepared. Showing the original description.
Expert guide for parallel tool execution, MCP batching (vibes_batch), asynchronous Promise.allSettled pipelines, subagent concurrent fan-out, and latency-optimized tool calling loops / Panduan ahli eksekusi tool paralel dan batching MCP.
The English summary is being prepared. Showing the original description.
Turn a brand's positioning into a 30-day, multi-platform content batch — 3-7 pillars, subtopic matrix mapped to funnel stage and persona, pillar pieces plus derivative fan-out, proof elements in at least half the slots, all registered in the editorial calendar store and quality-gated. Use when "content batch", "batch my content", "30 days of content", "month of content", "content batching machine", "content pillars", "fill the content calendar", "plan a month of posts", or any request to produce a full month of multi-platform content in one run.
The English summary is being prepared. Showing the original description.
Identify and exploit GraphQL API vulnerabilities by leveraging Introspection queries to dump the entire database schema, performing query batching to bypass rate limits (brute forcing), and extracting deeply nested unauthorized data via graph relationship abuse.
The English summary is being prepared. Showing the original description.
Exploit GraphQL API architectural features to execute highly efficient brute-force, Credential Stuffing, and Denial of Service (DoS) attacks. Utilize Query Batching and Alias injection to bypass rate limits by packing thousands of requests into a single HTTP POST request.
The English summary is being prepared. Showing the original description.
Diagnose and eliminate the GraphQL N+1 resolver explosion using Facebook DataLoader's per-request batching, plus the second-order defenses (persisted queries, depth limits, cost analysis). Use when a GraphQL query that returns N items in a list triggers N+1 database round trips, when designing nested resolvers that load associations, when a list query times out under load, or when you need to budget query cost before resolvers run. NOT for REST-style endpoint design (use api-versioning-strategy), generic database query optimization (use database-query-optimizer), or schema design (use graphql-server-architect).
The English summary is being prepared. Showing the original description.
Otimiza inferência de LLM com NVIDIA TensorRT para máxima vazão e latência mínima. Use para implantação em produção em GPUs NVIDIA (A100/H100), quando você precisa de inferência 10-100x mais rápida que PyTorch, ou para servir modelos com quantização (FP8/INT4), batching em voo e escalabilidade multi-GPU.
The English summary is being prepared. Showing the original description.
Serve LLMs com alta throughput usando PagedAttention do vLLM e continuous batching. Use ao fazer deploy de APIs LLM em produção, otimizar latência/throughput de inferência, ou servir modelos com memória GPU limitada. Suporta endpoints compatíveis com OpenAI, quantização (GPTQ/AWQ/FP8) e tensor parallelism.
The English summary is being prepared. Showing the original description.
GraphQL saldırıları — introspection, aliased query batching, rate limit bypass
The English summary is being prepared. Showing the original description.
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
The English summary is being prepared. Showing the original description.
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.
The English summary is being prepared. Showing the original description.