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 generate-competing-hypotheses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。
generate-competing-hypotheses
Purpose
Generate distinct, testable hypotheses that explain a target observation or research problem.
Input contract
yaml
required: [problem_statement, observations, scope_constraints]
optional: [existing_theories, analogies, anomaly_records]
constraints: [each hypothesis must imply a differentiating prediction or assumption]Procedure
- Extract the target phenomenon and unresolved explanatory gap.
- Generate mechanismally distinct explanations, including a null or status-quo hypothesis.
- Translate each explanation into predictions, assumptions, and disconfirming observations.
- Deduplicate equivalent explanations and retain the comparison set.
If competing hypotheses have explicit predictions and evidence links, consider score-object as the next tactic.
Output contract
yaml
produces: [hypothesis_set, mechanism_descriptions, prediction_set, assumption_register]
delta_fields: [hypothesis_updates, uncertainties, open_questions, recommended_jumps]Quality gates
- Hypotheses differ in mechanism or prediction, not wording only.
- A null or baseline explanation is considered where applicable.
Failure and counterexamples
Do not generate unfalsifiable narratives or discard inconvenient hypotheses before comparison.
Provenance map
resolved: generate-competing-hypotheses
