cos-consolidate-memory

Weekly memory consolidation for an AI Chief of Staff promotes learnings, supersedes stale facts, and decays or prunes memory. It produces a reviewable diff and changelog by consolidating the week's captures into a canonical Markdown file.

Automate memory management by running the weekly consolidation process, which identifies stale facts and decays or prunes outdated data automatically. It then captures outcomes and closes loops, outputting a reviewable diff for verification.

The best way to review changes during memory consolidation is by examining the reviewable diff and changelog automatically generated by the process. This diff highlights all promoted learnings and superseded facts before finalizing the canonical Markdown file.

Yes, you need Python installed to execute the weekly memory consolidation process. Python runs the necessary scripts to parse system files, close loops, catch calendar problems, and generate the final reviewable diff.

Yes, memory consolidation catches calendar problems early and closes open loops automatically during the weekly process. It also captures outcomes and coaches the user by reviewing the system's weekly captures and corrections.

You should avoid running memory consolidation if your system files lack the weekly captures and corrections required for processing. Running it without proper inputs prevents the system from accurately promoting learnings or producing a meaningful reviewable diff.