mcpField
H
hyperfine-benchmarking

hyperfine-benchmarking

github.com/akillnessTesting & quality53248 installsUpdated 4d ago
Repository

About hyperfine-benchmarking

hyperfine-benchmarking is a skill for benchmarking CLI commands with repeatable conditions. It helps developers compare command variants, assess script changes, and prepare benchmark results for review.

Key features

  1. 01Compare multiple CLI commands
  2. 02Run warmups and repeated measurements
  3. 03Sweep command parameters
  4. 04Export JSON and Markdown results
  5. 05Record benchmark conditions

Use cases

  1. 01Compare alternative command implementations
  2. 02Measure performance before and after script changes
  3. 03Attach benchmark evidence to a pull request
  4. 04Investigate noisy or inconsistent measurements

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/akillness/jeo-skills --skill hyperfine-benchmarking

After installing, type/hyperfine-benchmarkingin Claude Code to confirm it loaded.

Details

From the project README

hyperfine-benchmarking

Benchmark CLI commands with reproducible methodology instead of one-off time output.

When to use this skill

  • Compare two or more command variants (tool-a vs tool-b)
  • Validate performance impact before/after script changes
  • Attach benchmark evidence to PRs

Instructions

  1. Confirm tool availability.
  2. Keep input/workdir/environment stable across compared commands.
  3. Run warmups and enough runs for stable variance.
  4. Export JSON/markdown outputs for review.
  5. Summarize relative speedup + risk notes.

Examples

Availability check
hyperfine --version
Two-command comparison
hyperfine \
  --warmup 3 \
  --min-runs 10 \
  'cmd_a --with flags' \
  'cmd_b --with flags'
Parameter sweep
hyperfine \
  --warmup 3 \
  --parameter-list mode fast,balanced,thorough \
  'mytool --mode {mode} input.txt'
Export artifacts
hyperfine \
  --warmup 3 \
  --min-runs 10 \
  --export-json benchmark.json \
  --export-markdown benchmark.md \
  'cmd_a' 'cmd_b'

Best practices

  • Prefer relative speedup and confidence ranges over single-run claims.
  • Do not compare commands with different semantics unless outputs are normalized.
  • If variance is high, increase runs or reduce background noise before concluding.
  • Record dataset/path and exact command strings in PR text.

References

Comments

Markdown supported