# hyperfine-benchmarking

hyperfine-benchmarking — a skill that runs commands with warmup cycles and collects timing statistics. Compare tool variants or validate performance changes before committing, with exportable evidence.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 251 installs · ★ 61 repo stars
- Author: akillness
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/akillness/jeo-skills/tree/HEAD/.agent-skills/hyperfine-benchmarking/SKILL.md
- Updated: 2026-10-24
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/hyperfine-benchmarking

## What it does

Available for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. The skill wraps hyperfine to go beyond one-off `time` runs: it executes multiple passes with warmup, computes mean and variance, and exports results to JSON and Markdown suitable for code review.

Built-in safeguards ensure stable environment, identical inputs across compared commands, and enough repetitions for confident conclusions. Examples cover two-command comparison, parameter sweeps, and guidance on interpreting results—when variance is high, add more runs rather than drawing conclusions from noise.

Requires hyperfine in `PATH`. Absolute speed depends on command complexity and system load, but relative speedup of one variant over another (with all else equal) comes out reliably.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add akillness/jeo-skills --skill hyperfine-benchmarking -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/hyperfine-benchmarking/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/hyperfine-benchmarking/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add akillness/jeo-skills --skill hyperfine-benchmarking -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/hyperfine-benchmarking/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add akillness/jeo-skills --skill hyperfine-benchmarking -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/hyperfine-benchmarking/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install akillness/jeo-skills .agent-skills/hyperfine-benchmarking/SKILL.md
```

2. **Manually: put the files in place**
- File: `.github/skills/hyperfine-benchmarking/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add akillness/jeo-skills --skill hyperfine-benchmarking -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/akillness/jeo-skills --path .agent-skills/hyperfine-benchmarking
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/hyperfine-benchmarking/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add akillness/jeo-skills --skill hyperfine-benchmarking -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/akillness/jeo-skills/tree/HEAD/.agent-skills/hyperfine-benchmarking/SKILL.md

## When to use it
- measure performance difference between two tool versions
- prove a script optimization actually speeds things up
- gather timing statistics for a performance report
- test which command parameters run fastest
- attach benchmark evidence to a pull request

## Example prompts
- Compare curl and wget download speed, run each 10 times with 3 warmups and show the results.
- Run a parameter sweep testing fast, balanced, and thorough modes, then export all three to JSON and Markdown.
- Benchmark my script before and after my changes, export both results so I can see the speedup.

## Similar
- [cost-trend](https://affpapa.org/en/nevibecode/skills/cost-trend.md) — skill · 683 installs
- [performance-benchmark-specialist](https://affpapa.org/en/nevibecode/skills/performance-benchmark-specialist.md) — skill · 457 installs
- [shell-integration](https://affpapa.org/en/nevibecode/skills/shell-integration.md) — skill · 123 installs
- [nushell-pro](https://affpapa.org/en/nevibecode/skills/nushell-pro.md) — skill · 2800 installs
- [bash-script-generator](https://affpapa.org/en/nevibecode/skills/bash-script-generator.md) — skill · 742 installs
- [shell-expert](https://affpapa.org/en/nevibecode/skills/shell-expert.md) — skill · 141 installs

From the same set: [akillness/jeo-skills](https://affpapa.org/en/nevibecode/by/akillness/jeo-skills.md) — 99 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
