knowledge-ops

Cross-system knowledge syncing works by checking for existing entries before writing, applying frontmatter-based metadata, and routing updates to the correct storage layer to keep memory files, knowledge graphs, and external data stores aligned without duplication.

Organizing project context involves selecting the appropriate storage layer—quick-access memory, knowledge graphs, durable repo notes, or external data stores—based on the specific task requirements and retrieval needs.

Frontmatter-based metadata is required during knowledge ingestion to enable proper indexing, duplicate checks, and accurate storage-layer selection, ensuring that reference notes and conversation exports remain searchable and durable across systems.

Deduplicating session notes requires checking for existing entries before writing, ensuring synthesized research updates related context in the knowledge graph rather than creating redundant copies across memory and durable repositories.

Yes, conversation exports can be kept searchable by routing them to durable repository notes with frontmatter-based metadata and index updates, ensuring they remain durable and retrievable alongside other project context.

Storage-layer selection limitations include the strict requirement for frontmatter-based metadata on all entries and the need for duplicate checks before writing, which may slow down ingestion when syncing across multiple external data stores simultaneously.