What problem does it solve? When a project accumulates many AI skills, rules, and memories, the ecosystem can silently break: rules reference renamed skills, memories drift from the rules they summarize, and trigger chains dead-end. This Skill audits the whole asset structure rather than any single file, catching broken links, harmful duplication, and ambiguous handoffs before they mislead agent execution. ## Core Features & Use Cases - Five-dimension assessment: Statically checks three-layer consistency (skills/rules/memories/agents), redundancy health (intentional vs harmful), asset boundary overlap, and trigger-chain closure, then dynamically simulates an end-to-end cross-skill workflow to find handoff breaks. - Evidence-graded conclusions: Every finding is labeled as existing, referenced, hit, missing, or unobserved, so reports never overstate what was actually verified. - Actionable report: Produces a prioritized Markdown report with mermaid workflow diagrams, per-finding minimal fixes, and verification steps. - Use Case: After adding or renaming several skills in your agent configuration repository, ask for an ecosystem review to confirm no rule still points to a deleted skill and that the requirement-to-delivery chain still closes. ## Quick Start Ask the AI to run an ecosystem review on the current project to check consistency, redundancy, and trigger-chain closure across all skills, rules, and memories.
