What problem does it solve? Managing a self-improving harness learning subsystem requires a controlled bridge between CLI commands and user approval decisions. This Skill produces structured Tier 4 auto-update proposal payloads that the orchestrator surfaces via AskUserQuestion, ensuring no learning-driven change is applied without explicit human approval. ## Core Features & Use Cases - Proposal Payload Generation: Runs moai harness apply to fetch pending Tier 4 proposals and emits a structured JSON payload (proposal_id, target_path, field_key, new_value, confidence, recommended_action) for orchestrator consumption. - Apply/Rollback Orchestration: Coordinates approval writes to .moai/harness/proposals/, executes approved changes via the 5-Layer Safety Pipeline, and restores snapshots with moai harness rollback. - Lifecycle Management: Checks tier distribution and rate limits with moai harness status, and disables learning via moai harness disable while preserving YAML comments. - Use Case: When the harness observer accumulates enough pattern observations to propose updating a user-area skill's description or triggers, use this Skill to surface the proposal for approve/reject/inspect/defer and execute the chosen action safely. ## Quick Start Ask the assistant to check harness learning status and surface any pending auto-update proposals for approval.
