[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-c202a21e-ddfd-4085-b8fd-8e9d2ce83832":3,"$flcFIkJ0G5p2c26zXEaQ45rWSnjzrhl53RWuHf7-Z50s":43},{"id":4,"title":5,"description":6,"categoryId":7,"moduleId":8,"tags":9,"prompt":10,"icon":11,"source":12,"sourceUrl":13,"authorId":14,"authorName":15,"isPublic":16,"stars":17,"runs":18,"createdAt":19,"updatedAt":19,"module":20,"category":27,"packages":34},"c202a21e-ddfd-4085-b8fd-8e9d2ce83832","experiment-designer","在规划产品实验、撰写可测试的假设、估计样本量、优先排序测试或以实际统计严谨性解释A\u002FB测试结果时使用。","cat_life_career","mod_other","alirezarezvani,other","---\nname: experiment-designer\ndescription: Use when planning product experiments, writing testable hypotheses, estimating sample size, prioritizing tests, or interpreting A\u002FB outcomes with practical statistical rigor.\n---\n\n# Experiment Designer\n\nDesign, prioritize, and evaluate product experiments with clear hypotheses and defensible decisions.\n\n## When To Use\n\nUse this skill for:\n- A\u002FB and multivariate experiment planning\n- Hypothesis writing and success criteria definition\n- Sample size and minimum detectable effect planning\n- Experiment prioritization with ICE scoring\n- Reading statistical output for product decisions\n\n## Core Workflow\n\n1. Write hypothesis in If\u002FThen\u002FBecause format\n- If we change `[intervention]`\n- Then `[metric]` will change by `[expected direction\u002Fmagnitude]`\n- Because `[behavioral mechanism]`\n\n2. Define metrics before running test\n- Primary metric: single decision metric\n- Guardrail metrics: quality\u002Frisk protection\n- Secondary metrics: diagnostics only\n\n3. Estimate sample size\n- Baseline conversion or baseline mean\n- Minimum detectable effect (MDE)\n- Significance level (alpha) and power\n\nUse:\n```bash\npython3 scripts\u002Fsample_size_calculator.py --baseline-rate 0.12 --mde 0.02 --mde-type absolute\n```\n\n4. Prioritize experiments with ICE\n- Impact: potential upside\n- Confidence: evidence quality\n- Ease: cost\u002Fspeed\u002Fcomplexity\n\nICE Score = (Impact * Confidence * Ease) \u002F 10\n\n5. Launch with stopping rules\n- Decide fixed sample size or fixed duration in advance\n- Avoid repeated peeking without proper method\n- Monitor guardrails continuously\n\n6. Interpret results\n- Statistical significance is not business significance\n- Compare point estimate + confidence interval to decision threshold\n- Investigate novelty effects and segment heterogeneity\n\n## Hypothesis Quality Checklist\n\n- [ ] Contains explicit intervention and audience\n- [ ] Specifies measurable metric change\n- [ ] States plausible causal reason\n- [ ] Includes expected minimum effect\n- [ ] Defines failure condition\n\n## Common Experiment Pitfalls\n\n- Underpowered tests leading to false negatives\n- Running too many simultaneous changes without isolation\n- Changing targeting or implementation mid-test\n- Stopping early on random spikes\n- Ignoring sample ratio mismatch and instrumentation drift\n- Declaring success from p-value without effect-size context\n\n## Statistical Interpretation Guardrails\n\n- p-value \u003C alpha indicates evidence against null, not guaranteed truth.\n- Confidence interval crossing zero\u002Fno-effect means uncertain directional claim.\n- Wide intervals imply low precision even when significant.\n- Use practical significance thresholds tied to business impact.\n\nSee:\n- `references\u002Fexperiment-playbook.md`\n- `references\u002Fstatistics-reference.md`\n\n## Tooling\n\n### `scripts\u002Fsample_size_calculator.py`\n\nComputes required sample size (per variant and total) from:\n- baseline rate\n- MDE (absolute or relative)\n- significance level (alpha)\n- statistical power\n\nExample:\n```bash\npython3 scripts\u002Fsample_size_calculator.py \\\n  --baseline-rate 0.10 \\\n  --mde 0.015 \\\n  --mde-type absolute \\\n  --alpha 0.05 \\\n  --power 0.8\n```\n","","imported","https:\u002F\u002Fgithub.com\u002Falirezarezvani\u002Fclaude-skills","user_system_seed","SkillOPIC",true,239,1439,"2026-05-16 14:03:23",{"id":8,"name":21,"slug":22,"icon":23,"description":24,"sort":25,"createdAt":26},"其他","other","mdi-page-next-outline","其他类型Skill",5,"2026-05-16 12:53:40",{"id":7,"name":28,"slug":29,"icon":30,"description":31,"moduleId":8,"sort":32,"skillCount":33,"createdAt":26},"职场发展","career","mdi-briefcase-outline","面试准备、简历优化、职业规划",4,575,[35],{"id":36,"skillId":4,"version":37,"fileName":38,"fileSize":39,"filePath":40,"fileHash":41,"manifest":42,"createdAt":19},"b78ae8a1-bd6b-4007-a30b-b2b74eae38a1","1.0.0","experiment-designer.zip",4911,"uploads\u002Fskills\u002Fc202a21e-ddfd-4085-b8fd-8e9d2ce83832\u002Fexperiment-designer.zip","d3928e338a6b6b3e2ac3657728445cda481dd1cc1a9728f9ad12eda7007e7dbf","[{\"path\":\"SKILL.md\",\"isDirectory\":false,\"size\":3129},{\"path\":\"references\u002Fexperiment-playbook.md\",\"isDirectory\":false,\"size\":1934},{\"path\":\"references\u002Fstatistics-reference.md\",\"isDirectory\":false,\"size\":1546},{\"path\":\"scripts\u002Fsample_size_calculator.py\",\"isDirectory\":false,\"size\":3024}]",{"code":44,"message":45,"data":46},200,"success",{"items":47,"stats":48,"page":51},[],{"averageRating":49,"totalRatings":49,"ratingCounts":50},0,[49,49,49,49,49],{"limit":52,"offset":49,"hasMore":53,"nextOffset":52,"ratedOnly":16},15,false]