kwp-productivity-memory-management

Tiered memory management provides persistent context by using a hot-cache for immediate recall of frequent contacts and a deep-storage directory for comprehensive knowledge retrieval. This architecture ensures AI agents maintain continuous awareness of organizational profiles and project metadata.

To setup context-aware memory, initialize the system by running the productivity start command to scan existing documents and chat history. This process builds structured Markdown files that maintain consistent profiles for people, projects, and organizational context.

Yes, you can use structured Markdown files to maintain organizational context. The system requires these files to consistently profile people, projects, and organizational context, enabling the AI to automatically translate internal jargon and project codenames into actionable information.

The best way to retrieve project-specific metadata is through contextual decoding. When referring to a colleague by nickname, the system instantly retrieves their full profile, role, and project status from the hot-cache or deep-storage layers to provide an informed response.

The limitation of this knowledge management approach is its strict dependence on structured Markdown files. Without maintaining consistent profiles for people and projects in these files, the hot-cache and deep-storage layers cannot accurately decode workplace shorthand or retrieve organizational context.