Reality Testing Decisions
Use when tempted to predict, analyze, or debate whether something will work instead of testing it. When facing uncertainty about a decision, product, hire, or opportunity and relying on gut feeling or expert opinions.
Install
npx skills add apple-ouyang/book-to-skill --skill reality-testing-decisionsgit clone --depth 1 https://github.com/apple-ouyang/book-to-skill.git /tmp/book-to-skill
mkdir -p ~/.claude/skills
cp -r /tmp/book-to-skill/skills/reality-testing-decisions ~/.claude/skills/reality-testing-decisionsPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# 用实验替代预测 ## 任务目标 把"这个会不会成功?"的预测问题,转化为"怎么用最小成本验证?"的实验问题。 核心前提:我们对自己预测未来的能力严重高估。不要猜,不要感觉,去测试。 ## 为什么预测不可靠 Tetlock 收集了 82361 个专家预测,结论:**专家预测不如基本比率**(用历史平均值外推),而基本比率又不如小规模实验。 额外的教育、20年经验、博士学位——都不能提高预测精度。反而,媒体曝光度越高的专家,预测越不准。 **基本比率的力量**:共同基金 vs 指数基金——基本比率数据如此清晰,以至于选共同基金几乎必然让你退休时更穷。当基本比率足够明确时,直接用它,不需要再做实验。 结论:只要有可能,就应该完全避免预测,用实验代替。 ## 操作步骤 ### 第一步:识别你在预测什么 把当前的问题写出来。如果它的形式是: - "我觉得这个产品会有市场" - "我感觉这个人能胜任" - "我认为这个方向是对的" 这就是在预测。继续下一步。 ### 第二步:设计最小实验 问:**用什么最小行动,可以在现实中得到反馈?** 不同场景的实验设计: **产品/项目验证** - CarsDirect 案例:不争论"网上卖车有没有人买",直接建个假网站,第一天卖出 3 辆车 - 亦仁原则:号是消耗品,先发垃圾内容让市场反馈,不要等到完美再发 - 默认项目都是通的,默认数据都是假的——对项目乐观去了解细节,对收入数据悲观谨慎投入 **招聘/合作** - 面试表现和工作表现几乎不相关(医学院研究:面试排名第 700 和第 100 的学生,入学后表现无差异) - 工作样本 > 面试:让候选人做一个真实任务,用结果评判,不用印象评判 - 希望实验室:给潜在员工 3 周咨询合约,"面试表现最佳的人常常工作表现最差" - **Steve Cole / HopeLab**:$100k 设计项目,不赌一家公司,同时雇 5 家只做第一阶段($20k),用实际表现而非提案选人。从「OR 思维」转向「AND 思维」 **职业/方向选择** - 想读药学院?先去药房工作几周,哪怕无薪 - NI 无线传感器:副总裁在投入 200-300 万前,先接了一个大学教授的小项目,走通了再加大投入 - 绝对不投入做任何无法亲自体验和感受的项目 **大家都觉得不靠谱的想法** - 印度农民 app:库克和团队不看好,但让团队试了,农民收入提高 20%,最终 32.5 万农民使用 - 不试试怎么知道靠不靠谱? **创业验证** - **烘焙创业**(Heath Brothers 案例):想开烘焙店,不要直接辞职全职投入。先在当地农贸市场摆一个月摊位,观察:有没有回头客?有没有盈利?盲测中你的产品排名如何?用真实市场反馈替代"邻居都夸我的布朗尼"式的自我确认。 - **药学院实习**(Heath Brothers 案例):想读药学院的学生,先去 3 家不同药房各实习 1 个月,而不是直接报名。实习成本:时间。替代的错误成本:4 年学费 + 发现不喜欢这份工作。 ### 第三步:渐进式实验(当恐惧阻止你测试时) 强迫症律师佩吉的 7 步实验——当你觉得"万一出错怎么办"时,把实验拆得更小: 1. 把法律摘要带回家,做三次修改 2. 做两次修改 3. 做一次修改 4. 晚下班一小时,把摘要留在办公室 5. 按时回家,不做多余修改 6. 刻意留下一个标点错误 7. 刻意留下一个语法错误 结果:没有公司败诉,没有人被解雇,甚至没有人注意到错误。
Details
- Skill name
reality-testing-decisions- Repository
- apple-ouyang/book-to-skill
- Path
skills/reality-testing-decisions/SKILL.md- License
- MIT
- Repo updated
- 25 February 2026
FAQ
What does the Reality Testing Decisions skill do?
Use when tempted to predict, analyze, or debate whether something will work instead of testing it. When facing uncertainty about a decision, product, hire, or opportunity and relying on gut feeling or expert opinions.
How do I install the Reality Testing Decisions skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add apple-ouyang/book-to-skill --skill reality-testing-decisions. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Reality Testing Decisions popular?
It lives in apple-ouyang/book-to-skill, which has 190 GitHub stars and 25 forks.
Related skills
Book to Skill
190apple-ouyang/book-to-skill
把一本书拆解成 Claude Code 可执行的 Skill。当用户说"拆书"、"把这本书变成 Skill"、"book to skill"、"extract skill from book" 时触发。
Decision Tripwires
190apple-ouyang/book-to-skill
Use when committing resources to a project, investment, or relationship. When feeling the pull of sunk cost. When a plan has no defined exit criteria. When you catch yourself thinking "let's wait and see what happens.
Making Decisions
190apple-ouyang/book-to-skill
Use when facing a significant decision, evaluating options, or feeling stuck between choices. Especially when the decision is high-stakes, irreversible, or involves uncertainty about outcomes. Triggers on phrases like "should I", "I don't know whether to", "I'm torn between", or "help me decide".
Skill Forge Benchmark
189AgriciDaniel/skill-forge
Benchmark Claude Code skill performance with variance analysis, tracking pass rate, execution time, and token usage across iterations. Runs multiple trials per eval for statistical reliability, aggregates results into benchmark.json, and generates comparison reports between skill versions. Use when user says "benchmark skill", "measure skill performance", "skill metrics", "compare skill versions", "skill performance", "track skill improvement", "skill regression test", or "skill A/B test".
Test Suite Builder
398JetBrains/skills
Design and generate layered Kotlin + Spring tests that balance speed, realism, and regression value across unit, slice, and integration levels. Use when adding coverage for business logic, controllers, repositories, security, serialization, or end-to-end workflows, and when generic AI output would otherwise overuse `@SpringBootTest`, misuse mocks, or ignore MockK and coroutine testing idioms.
Actions CI Tuning
423mizchi/skills
Use when auditing or improving GitHub Actions workflows for a project. Covers cache setup (npm/pnpm/yarn), job parallelism, shard-based test splitting, artifact handling pitfalls, and Playwright-specific patterns. Trigger on: slow CI, cache miss, flaky shard jobs, merge-reports failures, or an explicit 'tune CI' request.
Browse the most popular Claude skills, all testing & qa skills, or write your own with the skills generator.