Graphiti is an open-source framework designed to build knowledge graphs that handle real-time, temporal data, offering a structured approach for managing context crucial for LLM-based agents and advanced RAG pipelines. It addresses the limitations of traditional RAG by enabling agents to reason with state changes over time and collaborates with graph databases like FalkorDB to store and manage its temporal knowledge graphs. By combining semantic search with graph traversal, Graphiti efficiently retrieves context, allowing developers to build graphs that represent conversational history, user preferences, and other business logic. Its temporal-aware design and ability to preserve historical data by invalidating rather than deleting outdated information make it an essential tool for building sophisticated LLM agents and GraphRAG pipelines, offering capabilities that standard RAG systems lack.