Anomalous Behavior Chain Analysis

Problem

News, business moves, and market signals are often easier to narrate than to verify. A practical issue is that primary facts, secondary interpretations, and sentiment-driven speculation get mixed together, so surface explanations can be mistaken for underlying motives. This skill separates the event into a verifiable fact layer and an interpretation layer, then checks whether the behavior deviates from what would normally be expected before building a traceable analysis memo.

How it works

The skill produces a staged analysis rather than a loose opinion. Key stages include source verification to distinguish primary and secondary sources, competing-hypothesis testing to avoid relying on a single official narrative, abstract proposition extraction to turn concrete actions into reusable logic, chain reasoning to connect intermediate conclusions, multi-scenario branching to model different possible paths, and signal dashboards plus dynamic calibration to record follow-up evidence. The output should include falsifiable conditions, such as “if X happens, this conclusion should be revised,” instead of open-ended predictions.

Boundaries

It is suited to business analysis, investment reasoning, policy interpretation, and industry trend assessment. It is not a substitute for fact-checking, legal compliance review, or real-time trading decisions. The input should preserve enough original event context; if there is no clear anomaly, the analysis may stop with a brief interpretation of conventional behavior.