DejaView

This gives Claude a persistent knowledge graph it can read and write across sessions. Instead of starting from scratch every conversation, the agent calls agent_context() on startup to load your full context, then uses remember() to store new facts as typed relationships (works_at, decided, introduced_by) and recall() to query specific entities or connections. It's built on a REST API at dejaview.io where facts are stored as subject-predicate-object triples in your own isolated graph. Reach for this when you're tired of re-explaining your project setup, team members, or architectural decisions every single session, or when you want multiple specialized agents to share what they've learned without duplicating context.