design-of-experiments — Claude Code skill: how to install
design-of-experiments creates structured factorial and Taguchi test designs to evaluate multi-parameter systems. It minimizes test iterations while mapping how variable interactions impact performance.
no license 139 installs ★ 203 updated 2026-09-01
npx skills add lyndonkl/claude --skill design-of-experiments -a claude-codeWhat design-of-experiments does
Instead of tweaking parameters one by one, this skill applies systematic design of experiments methods. It walks through objective setting, factor categorization into control and noise variables, and test matrix generation. The output is a clear design-of-experiments.md document containing execution protocols and statistical verification strategies.
The skill embeds strict guardrails: orthogonality checks to keep factor correlation below 0.3, run order randomization to prevent time trends, and power analysis keeping Type II error risk at or below 0.20. It also assesses the draft against an internal rubric, blocking delivery if the score is below 3.5. Skipping run randomization will immediately distort outcomes due to lurking environmental drift.
What it helps with
- Screen down 10 to 30 candidate variables to identify the critical few
- Map interactions across parameters without running full factorials
- Configure systems to remain resilient against uncontrolled noise
- Generate orthogonal test matrices in
design-of-experiments.md - Perform statistical power calculations prior to test execution
Example prompts for the agent
- Draft a screening experiment for 14 backend configuration settings with a strict run budget.
- Build a Taguchi inner-outer array to test application throughput under variable network latency.
- Generate a central composite design matrix and write out
design-of-experiments.md.
How to install design-of-experiments
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill design-of-experiments -a claude-code2Manually: put the files in place
~/.claude/skills/design-of-experiments/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/design-of-experiments/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill design-of-experiments -a codex2Manually: put the files in place
.agents/skills/design-of-experiments/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill design-of-experiments -a cursor2Manually: put the files in place
.cursor/skills/design-of-experiments/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lyndonkl/claude skills/design-of-experiments2Manually: put the files in place
.github/skills/design-of-experiments/SKILL.md1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill design-of-experiments -a gemini-cli2Platform CLI
gemini skills install https://github.com/lyndonkl/claude --path skills/design-of-experimentsManually: put the files in place
.devin/skills/design-of-experiments/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill design-of-experiments -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about design-of-experiments
How do I install design-of-experiments in Claude Code?
Run npx skills add lyndonkl/claude --skill design-of-experiments -a claude-code. The file will be placed in ~/.claude/skills/design-of-experiments/SKILL.md.
What does the design-of-experiments skill do?
It structures multi-factor testing using fractional factorials or Taguchi arrays. This approach exposes parameter interactions with significantly fewer execution runs than brute-force testing.
What quality checks does the skill enforce before delivery?
It scores the test plan against a built-in rubric requiring an average rating of at least 3.5. Mandatory guardrails include randomized run sequences, orthogonality checks, and predefined statistical analysis plans.