Generates competing hypotheses and evaluates evidence to explain observed outcomes.
npx skills add https://github.com/xz1220/oh-my-kimi --skill trace-xz1220
Please help me install this Agent Skill. Skill: trace Source: https://github.com/xz1220/oh-my-kimi/tree/main/skills/trace Command: npx skills add https://github.com/xz1220/oh-my-kimi --skill trace-xz1220
It helps you explain why a specific observed result happened by structuring competing hypotheses and collecting supporting and contradicting evidence instead of looping on generic debugging or implementation.
Ask an AI to run the trace workflow by providing the observation you want to explain and requesting competing hypotheses with an evidence-backed ranking and the most discriminating next probe.
High-intent search queries and answers about installing and using this skill.
Root cause analysis for a performance regression involves generating competing hypotheses and collecting supporting and contradicting evidence for each. This approach grades evidence by strength to explain why the observed outcome occurred.
Structuring an investigation for ambiguous architectural postmortems requires preserving observations, hypotheses, and evidence gaps. This method applies falsification and re-ranking to competing explanations to find evidence-backed conclusions.
To stop generic debugging loops, generate multiple intentional hypotheses explaining the same observation. Prioritize stronger evidence like controlled reproduction and direct artifacts over weaker indirect claims.
Yes, hypothesis-driven investigation applies to experimental configuration trace-back. It processes evidence-dense scenarios by evaluating competing explanations and identifying a discriminating probe to collapse uncertainty.
When evidence gaps remain, identify a single highest-value discriminating probe. This probe quickly collapses uncertainty and resolves the critical unknown blocking your leading explanation.
Evidence-graded reasoning works for resource behavior analysis by downgrading weaker indirect claims. It prioritizes multi-source corroboration and controlled reproduction to validate hypotheses explaining the anomaly.
cheafiGenerate competing causal hypotheses and validate them against evidence.
sehoon787Generate competing hypotheses and rank evidence to triage ambiguous observations.
mattdurhamAnalyze logs, traces, and metrics to diagnose root causes across distributed systems.
mopeyjellyfishReproduce bugs from evidence and prove root causes with red signals.
