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 build-current-reality-tree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。
build-current-reality-tree
Purpose
Build a sufficient-cause Current Reality Tree linking undesirable effects to a small root set.
Input contract
yaml
required: [undesirable_effects, causal_evidence]
optional: [candidate_causes, system_boundary]
constraints: [each AND/OR connection requires a stated causal rationale]Procedure
- Normalize undesirable effects into observable statements.
- Connect effects with sufficient-cause relationships, marking AND requirements.
- Trace converging chains to a small root set and test each link.
Output contract
yaml
produces: [current_reality_tree, root_cause_candidates, causal_rationale]
delta_fields: [findings, assumption_updates, uncertainties, decisions]Quality gates
- Every effect has an incoming explanation or is marked unexplained; root candidates explain multiple effects where evidence supports it.
Failure and counterexamples
Do not infer causality from co-occurrence, and do not force a single root when independent roots remain.
Provenance map
deep-insight/current-reality-tree: resolved.
