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 detect-feedback-loop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。
detect-feedback-loop
Purpose
Detect reinforcing or balancing feedback loops in a causal or process structure and state the evidence for each loop.
Input contract
yaml
required: [causal_graph, node_semantics, edge_polarity]
optional: [time_delays, observed_series, intervention_records]
constraints: [loop classification requires directed edges and polarity or transition evidence]Procedure
- Normalize directed relations, polarity, and delays.
- Enumerate simple cycles and identify reinforcing or balancing sign patterns.
- Compare loops with observations or intervention evidence where available.
- Report loop boundaries, uncertain edges, and testable implications.
Output contract
yaml
produces: [feedback_loops, loop_classification, supporting_evidence, uncertain_edges, testable_implications]
delta_fields: [findings, evidence_updates, uncertainties, recommended_jumps]Quality gates
- Every loop lists its ordered edges and polarity basis.
- Correlational cycles are not presented as causal loops without qualification.
Failure and counterexamples
Do not infer feedback from a static co-occurrence or omit time direction where it determines loop meaning.
Provenance map
resolved: detect-feedback-loop
