Execute a metric command N times across 2-3 repo variants and emit markdown and TSV reports.
npx skills add https://github.com/arbazkhan971/godmode --skill bench-arbazkhan971
Please help me install this Agent Skill. Skill: bench Source: https://github.com/arbazkhan971/godmode/tree/main/skills/bench Command: npx skills add https://github.com/arbazkhan971/godmode --skill bench-arbazkhan971
This skill automates statistically valid performance comparisons by executing a user-provided metric command N times across 2-3 repo variants and reporting reproducible results.
Run /godmode:bench with a metric_cmd and 2-3 variants to compare their performance.
High-intent search queries and answers about installing and using this skill.
Automate performance benchmarking across git branches by executing a user-provided metric command N times across 2-3 repo variants. The process snapshots and restores your git workspace, runs the benchmarks, and calculates delta percentages to quantify performance changes.
Compare code performance variants by running a metric command N times across 2-3 states and calculating delta percentages using awk. The process enforces N>=3 runs and performs variance recovery with up to three retries to ensure reproducible results.
Yes, you can benchmark code changes safely because the process enforces a clean git workspace, automatically snapshots your starting state, and restores it after executing the performance comparisons across the specified variants.
Performance benchmarking results are reported and exported by emitting a tabular markdown report and a dedicated TSV file. These outputs include per-variant metrics like mean, median, and standard deviation, alongside a one-paragraph summary.
Statistically valid variance recovery in code benchmarking is a mechanism that enforces a minimum of three runs and performs up to three retry attempts. This ensures the collected performance metrics are reproducible and statistically reliable.
JoNaYeonValidate Git repository status and create a baseline branch before coding.
wshaddixCompare BenchmarkDotNet results against baselines to detect performance regressions in CI pipelines.
namanchopraAnalyze feature branch health and merge readiness from git data.
josephneumannAnalyze Git commit history to generate engineering retrospective reports.
