Execute Go benchmarks and flag regressions exceeding 10% against a saved baseline.
npx skills add https://github.com/PeterBooker/veloria --skill benchmark-peterbooker
Please help me install this Agent Skill. Skill: benchmark Source: https://github.com/PeterBooker/veloria/tree/main/.claude/skills/benchmark Command: npx skills add https://github.com/PeterBooker/veloria --skill benchmark-peterbooker
This Skill helps identify performance regressions in Go code by running benchmarks and comparing results against a baseline, ensuring code changes do not negatively impact performance.
Run benchmarks on all packages in the project.
High-intent search queries and answers about installing and using this skill.
Detect performance regressions in Go by executing benchmarks and comparing current `ns/op`, `B/op`, and `allocs/op` metrics against a saved baseline. The analysis flags any degradations exceeding a 10% threshold.
Run Go benchmarks with memory and allocation tracking by executing the benchmark tool across performance-critical packages. This captures `B/op` and `allocs/op` metrics to identify potential bottlenecks introduced by recent code changes.
The best way to compare Go benchmark results against a baseline is to save initial metrics and diff them against current results. This trend analysis isolates regressions in operations, memory, and allocations exceeding the 10% threshold.
Benchmark regression detection works by targeting performance-critical packages like `internal/index/`, `internal/repo/`, and `internal/api/search/`, running tests, and flagging metric degradations exceeding 10% compared to the saved baseline.
A 10% threshold is used to flag performance regressions in Go benchmarks. If current `ns/op`, `B/op`, or `allocs/op` metrics exceed the saved baseline by more than 10%, the analysis flags the degradation.
You can use benchmark regression detection after refactoring a search algorithm to ensure changes haven't introduced a performance bottleneck. It executes targeted Go benchmarks and compares the results against your previously saved baseline metrics.
ahravCompare Criterion.rs benchmark results against a saved baseline.
kis9aExecute Go benchmarks and analyze ns/op metrics to identify bottlenecks.
PeterBookerExecute Go unit tests for specified or all packages.
rlagycksMeasure web, API, and build performance metrics and compare results.
