promote-memory

Automate memory promotion by running parallel five-critic reviews on generality, staleness, redundancy, evidence, and format, then promoting learnings via majority vote. This ensures only high-quality entries advance to your permanent memory.

The five-critic review process evaluates stored learnings across five dimensions: generality, staleness, redundancy, evidence, and format. Each critic runs in parallel, and learnings are promoted only when a majority of three or more critics approve.

Review and promote LEARN entries by triggering the memory promotion command, which initiates the five-critic council evaluation. Approved entries pass a majority vote and are promoted to maintain a high-quality memory file.

Yes, the memory promotion process applies directly to academic workflows. It reviews stored learnings for relevance and quality using the five-critic council, ensuring your academic knowledge base retains only validated, non-redundant entries.

The majority vote system requires three of five critics to approve promotion, meaning entries with split evaluations may not advance. This strict threshold ensures quality but may filter out niche learnings lacking broad generality or conventional formatting.

No external dependencies are required to run the memory promotion process. You only need a personal memory file containing LEARN entries to initiate the five-critic review and majority vote promotion.