Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by kael-bit • Uncategorized
A memory engine for AI agents that organizes, promotes, and decays memories over time using a three-layer lifecycle and semantic clustering.
Persistent, organized memory with forgetting and promotion mechanisms.
Semantic search combined with keyword search and topic browsing.
Automatic memory consolidation and lifecycle management without external dependencies.
engram-rs is a Rust-based memory engine designed for AI agents, providing a structured memory lifecycle with automatic decay, promotion via LLM evaluation, and semantic deduplication. It organizes memories into a self-organizing topic tree, enabling agents to browse and recall relevant knowledge efficiently. The system runs as a single binary with SQLite storage, requiring no external dependencies, and supports integration with various LLM providers for enhanced intelligent features.