What problem does it solve? Over time an AI memory system accumulates bloated files, stale content, routing gaps, and learnings that never get captured — with no visibility into what is actually working. This Skill performs a full cross-source audit of the memory system so you can see what is overloaded, what is never retrieved, and what to restructure. ## Core Features & Use Cases - Cross-source correlation: Analyzes retrieval logs, session diary, learnings, and the memory corpus together to surface systemic patterns like overloaded files, wasteful retrievals, and learning gaps. - Utilisation and routing precision metrics: Runs deterministic scripts to estimate which memory sections were actually used (not just loaded) and how often suggested skills were invoked. - Longitudinal tracking: Saves dated assessment snapshots and compares against previous audits to show whether past changes improved the system. - Use Case: Run monthly to answer "how's memory doing" — get prioritized recommendations such as which files to split, which routing domains to tighten, and which stale sections need review, each backed by concrete metrics. ## Quick Start Ask the assistant to run a memory health audit and report what patterns it has noticed about the memory system.
