BE

benchmark-guardian

Validates performance claims through rigorous, data-driven benchmarking.

Install

mkdir -p .claude/skills/benchmark-guardian && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/11044" && unzip -o skill.zip -d .claude/skills/benchmark-guardian && rm skill.zip

Installs to .claude/skills/benchmark-guardian

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Use for JS/TS performance-sensitive changes, benchmark additions, benchmark regressions, benchmark result JSON, or any claim that code is faster, smaller, cheaper, or not slower.
178 chars✓ has a “when” trigger
Intermediate

Key capabilities

  • State a benchmark hypothesis
  • Add focused benchmarks for iteration
  • Write structured JSON to `benchmarks/results/*latest.json`
  • Run correctness tests or fuzzers for the same path
  • Update evidence documentation for performance
  • Include command, output path, key rows, and caveats in final response

How it works

This skill guides the benchmarking workflow by requiring a hypothesis, focused benchmarks, structured JSON results, and correctness tests. It also updates performance documentation.

Inputs & outputs

You give it
performance-sensitive code changes or benchmark additions
You get back
structured JSON benchmark results, updated performance documentation, and a final response with caveats

When to use benchmark-guardian

  • Benchmark code optimization
  • Verify performance regressions
  • Generate benchmarking docs

About this skill

Benchmark Guardian

Do not make performance claims from intuition.

Workflow

  1. State the benchmark hypothesis:

    • target behavior,
    • metric,
    • fixture,
    • command,
    • acceptance threshold or baseline.
  2. Prefer focused benchmarks while iterating:

npx evidence-kit add-benchmark --name core --language js
npm run bench:evidence
  1. Write structured JSON to benchmarks/results/*latest.json.

  2. Run correctness tests or fuzzers for the same path before saying a performance change is safe.

  3. Update evidence docs:

npm run docs:perf
npm run docs:perf:search -- <benchmark topic>

Result Rules

  • Keep timeout, error, unsupported, and bad-result rows visible.
  • Report exact command and output file.
  • Report fixture-level rows, not only aggregate scores.
  • If benchmark scope changed, run npm run bench:scope or npx evidence-kit scope.
  • If the benchmark came from external research, cite the fetch manifest and source note path.

Final Response

Include the command, output path, key rows, and caveats such as dirty worktree, missing baseline, or local machine noise.

When not to use it

  • When making performance claims from intuition without benchmarks
  • When not needing structured JSON benchmark results
  • When not requiring correctness tests or fuzzers for performance changes

Limitations

  • Do not make performance claims from intuition
  • Benchmark results must be written to `benchmarks/results/*latest.json`
  • Final response must include command, output path, key rows, and caveats

How it compares

This skill enforces a structured benchmarking workflow with explicit steps for hypothesis, iteration, and documentation, unlike ad-hoc performance testing.

Compared to similar skills

benchmark-guardian side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
benchmark-guardian (this skill)02moReviewIntermediate
react-best-practices223moNo flagsIntermediate
react-component-performance25moNo flagsAdvanced
dependency-upgrade265moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry