Loading...
Loading...
Found 155 Skills
Run an autonomous Humanize-governed vLLM SOTA performance loop for one LLM model: first perform the fixed fair vLLM/SGLang/TensorRT-LLM deployment search and benchmark, then start one RLCR loop that repeatedly decides the gap, profiles the current bottleneck, runs layer/kernel pipeline analysis, patches vLLM code, optionally uses ncu-report-skill for kernel evidence, and revalidates until vLLM matches or beats the best observed framework under the same workload and SLA.
Agent skill for benchmark-suite - invoke with $agent-benchmark-suite
Designs structured benchmarks for comparing algorithms, models, or implementations. Selects appropriate metrics (latency, throughput, memory, accuracy), designs representative test cases, captures hardware/software context, produces comparison tables with tradeoff analysis, and includes reproduction instructions. Triggers on: "benchmark", "compare performance", "which is faster", "latency comparison", "memory comparison", "run benchmark", "design benchmark", "compare implementations", "evaluate algorithms", "performance comparison", "throughput test", "speed test". Use this skill when comparing two or more implementations, algorithms, or models.
Calculate and benchmark social media engagement rates across platforms and variants. Use this skill when the user needs to compute engagement metrics, compare performance across accounts or posts, or set engagement benchmarks — even if they say 'what is my engagement rate', 'benchmark engagement', or 'social media KPIs'.
Research TikTok Creative Center or ad-library style datasets for winning ad patterns, regions, objectives, hook language, and creative signals without mixing paid ads with organic creator discovery.
Competitive landscape analysis — builds a competitive structure research framework covering market positioning (Porter five-forces), peer cross-comparison (PE/PB/ROE/revenue growth), market share estimation, competitive advantage assessment (moat), and potential disruptor identification. Triggers: "竞争格局", "竞争分析", "行业竞争", "市场份额", "竞争对手", "护城河", "波特五力", "竞争优势", "競爭格局", "競爭分析", "行業競爭", "市場份額", "競爭對手", "護城河", "波特五力", "competitive analysis", "competitive landscape", "market share", "competitive moat", "Porter five forces", "industry competition", "competitive advantage", "market positioning", "moat analysis", "NVDA vs AMD", "who are the competitors".
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup, per-iteration statistics, and anti-pattern avoidance. Also covers NVTX annotation for labeling profiler timelines. NOT for: running or analyzing profiler tools (nsys, ncu, Nsight Systems, Nsight Compute), writing kernels (Triton, CuTe, CUDA), applying optimizations (CUDA Graphs, gradient checkpointing, fusion), or interpreting roofline/SOL% metrics. Triggers: "measure throughput", "benchmark this function", "time my training loop", "samples per second", "NVTX annotate", "instrument my dataloader", "data load time", "kernel timing", "how do I time".
Guardrail framework for monitoring spend, efficiency, and CAC thresholds by channel.
Gatling Scenario Creator - Auto-activating skill for Performance Testing. Triggers on: gatling scenario creator, gatling scenario creator Part of the Performance Testing skill category.
Audit how a brand appears in AI-powered search (ChatGPT, Perplexity, Claude, Gemini). Use when user mentions "AI search," "how do I show up in ChatGPT," "AI discoverability," "AEO," "LLM visibility," or wants to understand their brand's AI presence.
Analyzes venture capital, investment trends, funding rounds, investor strategies, M&A activity, and funding patterns in specific markets or industries. Use when the user requests funding analysis, VC landscape research, investment trend analysis, or wants to understand investor activity and funding dynamics.
Use when designing or auditing computer science experiments, evaluation plans, baselines, metrics, ablations, datasets, statistical tests, benchmarks, validity threats, or reproducibility claims.