使用方法
- 复制安装命令，在自己的智能体环境安装前先检查源码包。
- 从原始文档核对所需工具、凭据和支持的宿主。
- 说明输入材料与期望交付物，并对照来源检查输出。
npx skills add https://github.com/yogsoth-ai/de-anthropocentric-research-engine --skill plan-effect-size开始前需了解
- 安装不包含模型订阅、API 额度或访问文件的权限。
- 仓库星数和安装量反映活跃情况，不等于独立质量或安全测试。
仓库原始内容
按源仓库内容呈现，保留标题、案例、代码、表格、链接以及原文引用的演示图片。
plan-effect-size
Purpose
Choose effect-size estimands and conversion rules appropriate to outcomes and study design.
Input contract
yaml
required: [outcome_schema, study_designs, comparison_target]
optional: [reported_statistics, variance_rules, transformation_options]
constraints: [estimand, direction, and conversion assumptions must be explicit]Procedure
- Match each outcome and design to a valid estimand.
- Define direction, scale, variance, and required conversions.
- Record unavailable statistics and assumptions for any conversion.
- Produce a harmonized effect-size plan with incompatibilities.
If the effect measure and transformation rules are fixed, consider design-data-extraction as the next tactic.
Output contract
yaml
produces: [estimand_plan, conversion_rules, variance_requirements, incompatibility_log]
delta_fields: [decisions, assumption_updates, uncertainties, open_questions]Quality gates
- Effects are comparable only under declared assumptions.
- Conversions remain traceable to reported statistics.
Failure and counterexamples
Do not convert incompatible constructs into a common effect size merely to increase sample size.
Provenance map
resolved: effect-size-planning
