Loading...
Loading...
Found 193 Skills
Analyze Huawei Ascend NPU profiling data to discover hidden performance anomalies and produce a detailed model architecture report reverse-engineered from profiling. Trigger on Ascend profiling traces, NPU bottlenecks, device idle gaps, host-device issues, kernel_details.csv / trace_view.json / op_summary / communication.json. Also trigger on "profiling", "step time", "device bubble", "underfeed", "host bound", "device bound", "AICPU", "wait anchor", "kernel gap", "Ascend performance", "model architecture", "layer structure", "forward pass", "model structure". Runs anomaly discovery (bubble detection, wait-anchor, AICPU exposure) alongside model architecture analysis (layer classification, per-layer sub-structure, communication pipeline). Outputs a separate Markdown architecture report alongside anomaly analysis.
End-to-end SGLang SOTA performance workflow. Use when a user names an LLM model and wants SGLang to match or beat the best observed vLLM and TensorRT-LLM serving performance by searching each framework's best deployment command, benchmarking them fairly, profiling SGLang if it is slower, identifying kernel/overlap/fusion bottlenecks, patching SGLang code, and revalidating with real model runs.
Aspire platform help — word-of-mouth commerce for influencer marketing, product seeding, affiliate tracking, UGC sourcing, and paid social. Covers Discovery (170M+ profiles, Quickmatch AI, image recognition), Campaign Management (lifecycle tracking, content approval, term sheets), Product Seeding (Shopify gifting, shipping), Affiliate Tracking (promo codes, attribution), UGC & Content (library, repurposing for ads), Paid Social (TikTok Spark Ads, Meta whitelisting), Creator Payments (free processing). Integrates with Shopify, WooCommerce, Meta, TikTok, Pinterest, Klaviyo, CJ, Impact, ShareASale/Awin. Use when Aspire discovery isn't surfacing the right creators, product seeding orders aren't syncing with Shopify, affiliate tracking isn't attributing sales, content approvals are bottlenecked, not sure which Aspire plan fits, or integrations aren't connecting properly. Do NOT use for influencer strategy across platforms (use /sales-influencer-marketing) or affiliate program design (use /sales-affiliate-program).
End-to-end prospect research pipeline: Apollo enrichment → personalized email + call scripts → draft review → Apollo sequence load. Eliminates manual research bottleneck. Use when: 'research prospect', 'prospect [company]', 'build cadence for', 'outreach for [company]', 'research-to-cadence', 'enrich and sequence', 'new prospect batch'.
Builds generative AI applications on Amazon Bedrock. Covers model invocation (Converse API, InvokeModel), RAG with Knowledge Bases, Bedrock Agents, Guardrails, and AgentCore. Use when invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, troubleshooting Bedrock errors (ThrottlingException, AccessDeniedException), or choosing models (Claude, Llama, Nova, Titan). ALSO USE for prompt caching setup and debugging, quota health checks and throttling diagnosis, cost attribution and tracking, migrating between Claude model generations (4.5 to 4.6 to 4.7), chunking strategies, API selection (Converse vs InvokeModel), guardrail capabilities, and model selection. NOT for custom model training, Rekognition, or Comprehend.
Diagnoses and optimises slow SQL queries using EXPLAIN ANALYZE. Covers identifying bottlenecks (sequential scans, bad estimates, heap fetches), index strategy, query rewrites, and verification. Invoked when the user asks to optimize a query, fix a slow database query, or improve database performance.
Guide performance profiling with Instruments, diagnose hangs, memory issues, slow launches, and energy drain. Use when reviewing app performance or investigating specific bottlenecks.
Optimize bulk API requests with batching, throttling, and parallel execution. Use when processing bulk API operations efficiently. Trigger with phrases like "process bulk requests", "batch API calls", or "handle batch operations".
A framework for classifying product decisions based on impact and reversibility. Use this when you feel like a bottleneck for your team, when you have a massive backlog of choices to make, or when you need to justify spending weeks of research on a single high-stakes problem.
When the user wants to plan growth using the AARRR framework, diagnose growth bottlenecks, or map actions across the customer lifecycle. Also use when the user mentions "growth funnel," "AARRR," "pirate metrics," "acquisition activation retention," "customer lifecycle metrics," or "growth framework."
Use this skill whenever planning, designing, reviewing, or improving search and recommendation systems for a two-sided trust marketplace built on OpenSearch — covers user-intent framing, product-surface architecture, index design, query understanding, retrieval strategy, ranking, search-plus-recs blending, measurement, and a dashboard-and-alerting layer for ongoing decision making. Triggers on tasks involving marketplace search, homefeeds, ranking, relevance tuning, OpenSearch query DSL, analyzers, synonyms, golden sets, NDCG, A/B testing, or diagnosing an existing retrieval system. Use this skill BEFORE marketplace-personalisation when planning new work; hand off when the diagnosed bottleneck is personalisation-specific.
Intel VTune and AMD uProf profiling skill for microarchitecture analysis. Use when analyzing hotspots, microarchitecture bottlenecks, memory access patterns, pipeline stalls, or using the roofline model. Covers VTune Community Edition (free) and AMD uProf as a free alternative. Activates on queries about VTune, uProf, microarchitecture analysis, pipeline stalls, memory bandwidth, roofline model, or hardware performance analysis.