What problem does it solve? Development sessions produce valuable insights that are typically lost when the session ends, forcing teams to relearn the same lessons repeatedly. This Skill captures lessons, solutions, and patterns from each work cycle and stores them in a searchable knowledge base so future sessions start better equipped. ## Core Features & Use Cases - Multi-Agent Analysis Pipeline: Spawns a coordinated team of analyzer agents (context-analyzer, lesson-extractor, pattern-matcher, solution-writer) to review cycle artifacts like git diffs, test results, and spec drift. - Quality-Filtered Knowledge Capture: Applies novelty and specificity checks, deduplicates against existing lessons via search, and classifies items by type (lesson, solution, pattern, preference) and severity before storing them via the ca learn CLI. - Spec and Verification Drift Detection: Compares final implementation against original EARS requirements and the epic's Verification Contract to capture workflow-quality lessons. - Use Case: After completing a feature epic, run this Skill to analyze what changed versus the plan, confirm high-severity findings with the user, store concrete lessons, and deprecate outdated ADRs before closing the epic. ## Quick Start Ask the AI to run the compound phase to reflect on this cycle, extract lessons learned, and store them for future sessions.
