# Perf Benchmarker

> Use when running performance benchmarks, establishing baselines, or validating regressions with sequential runs. Enforces 60s minimum runs (30s only for binary search) and no parallel benchmarks.

- Skill: `majiayu000/perf-benchmarker` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/perf-benchmarker`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/perf-benchmarker/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/perf-benchmarker

---


# perf-benchmarker

Run sequential benchmarks with strict duration rules.

Follow `docs/perf-requirements.md` as the canonical contract.

## Required Rules

- Benchmarks MUST run sequentially (never parallel).
- Minimum duration: 60s per run (30s only for binary search).
- Warmup: 10s minimum before measurement.
- Re-run anomalies.

## Output Format

```
command: <benchmark command>
duration: <seconds>
warmup: <seconds>
results: <metrics summary>
notes: <anomalies or reruns>
```

## Output Contract

Benchmarks MUST emit a JSON metrics block between markers:

```
PERF_METRICS_START
{"scenarios":{"low":{"latency_ms":120},"high":{"latency_ms":450}}}
PERF_METRICS_END
```

## Constraints

- No short runs unless binary-search phase.
- Do not change code while benchmarking.