Execute multi-layer performance benchmarks for Scalex indexing, queries, and memory.
npx skills add https://github.com/nguyenyou/scalex --skill benchmark-nguyenyou
Please help me install this Agent Skill. Skill: benchmark Source: https://github.com/nguyenyou/scalex/tree/main/.claude/skills/benchmark Command: npx skills add https://github.com/nguyenyou/scalex --skill benchmark-nguyenyou
This Skill helps quantify Scalex performance, surface bottlenecks, and validate improvements by executing multi-layer benchmarks across indexing, querying, and memory usage.
Run the benchmark suite to establish a baseline of Scalex performance across the Scala repo.
High-intent search queries and answers about installing and using this skill.
To benchmark index build time and query latency, this Skill executes multi-layer performance tests using hyperfine and built-in timings across cold and warm indexing workloads, establishing a performance baseline for Scala repositories.
Generating a flame graph for profiling Scala performance is done using the async-profiler integration layer within the benchmark suite. It records stack traces and visualizes CPU hotspots to surface bottlenecks during indexing or querying operations.
Yes, you can use hyperfine to measure performance regression after code changes by running automated before and after comparison benchmarks. The Skill validates performance improvements by comparing query latency and index times across git refs.
You do not strictly need async-profiler and JFR for memory profiling and microbenchmarks, as the suite supports six independent profiling layers. However, JFR recording is required for Java flight recorder data and async-profiler is needed for flame graph generation.
Cold vs warm indexing affects benchmark results by measuring the initial build time without cache against subsequent cached query latency. This distinction surfaces memory usage patterns and reveals whether performance bottlenecks occur during initial population or repeated queries.
Memory profiling shows high usage during query operations because the benchmark captures heap allocations across typical query workloads. Analyzing these memory footprint metrics helps identify whether specific refs or complex queries cause unexpected memory spikes.
Drock91Automate mp CLI trading commands for scheduled DCA, limit orders, and stop-losses.
CarbeneAIRun baseline-aware browser benchmarks against URLs and report performance regressions.
kunallimayeProfile software performance with perf, valgrind, strace, and hyperfine.
jabrenaCompare Java profiling metrics before and after refactoring to validate optimizations.
