What problem does it solve? After fixing a bug, teams often move on without understanding why it happened, leading to the same class of bugs recurring. This Skill provides a structured retrospective framework that turns each fixed bug into documented prevention mechanisms and captured knowledge. ## Core Features & Use Cases - Five-Dimension Analysis: Categorizes root causes (missing spec, cross-layer contract, change propagation, test coverage gap, implicit assumption), reviews why earlier fixes failed, and defines prevention mechanisms. - Bayesian Reasoning Framework: Guides hypothesis formation with priors, evidence evaluation, belief updating, and confidence thresholds when multiple root causes are plausible. - Knowledge Capture: Directs findings into .trellis/spec/ guides and hands off spec decisions to the trellis-update-spec workflow with no-op as a valid outcome. - Use Case: After fixing a production bug where an API returned timestamps in seconds instead of milliseconds, run this analysis to classify it as an implicit assumption bug, identify similar risks across the codebase, and record a prevention checklist in the spec docs. ## Quick Start Ask the AI to run a break-the-loop analysis on the bug you just fixed and produce the structured root cause and prevention report.
