← All skills

empirical-prompt-tuning

85 / 100 · Low risk

Fine to try.

By nwiizo/workspace_2026 (GitHub) · other · 42 stars · scanned at commit 326f464

agent 向けテキスト指示（skill / slash command / task プロンプト / CLAUDE.md 節 / コード生成プロンプト）を、バイアスを排した実行者に動かしてもらい、両面（実行者の自己申告 + 指示側メトリクス）で評価して反復改善する手法。改善が頭打ちになるまで回す。プロンプトや skill を新規作成・大幅改訂した直後、またはエージェントの挙動が期待通りにならない原因を指示側の曖昧さに求めたいときに使う。

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'empirical-prompt-tuning' ] || [ -L ~/.agents/skills/'empirical-prompt-tuning' ]; then echo 'A skill folder named empirical-prompt-tuning already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/nwiizo/workspace_2026' "$d" && git -C "$d" checkout --quiet '326f464a3f34a0eff7a4cc5ab1a2e84f520ee8c0' -- 'tools/nwiizo-claude-plugins/plugins/prompt-engineering/skills/empirical-prompt-tuning' && mkdir -p ~/.agents/skills && cp -R "$d"/'tools/nwiizo-claude-plugins/plugins/prompt-engineering/skills/empirical-prompt-tuning' ~/.agents/skills/'empirical-prompt-tuning' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'empirical-prompt-tuning' ~/.claude/skills/'empirical-prompt-tuning' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for nwiizo/workspace_2026 shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/nwiizo/workspace_2026.svg)](https://zalavijaysinh.github.io/skilltrust/repos/nwiizo/workspace_2026/)

Score history