Show HN: I made an open-source memory layer for agents

This open-source memory layer addresses a fundamental requirement for sophisticated AI agents: persistent, structured knowledge management. The ability to auto-extract entities and relationships, coupled with semantic search, is crucial for agents to maintain context and perform complex reasoning over time. Offering both self-hostable and cloud options, alongside an MIT license, maximizes adoption flexibility for developers and enterprises. This directly tackles the pain point of building robust agent architectures without proprietary lock-in. The market implications are significant, as it provides a foundational component for developing more intelligent, stateful AI applications, potentially accelerating innovation in agentic workflows across various industries.