RAGFlow is an open-source Retrieval-Augmented Generation (RAG) engine that fuses retrieval pipelines with agent capabilities to provide an enhanced context layer for large language models. It emphasizes deep document understanding to create high-quality semantic chunks for accurate retrieval, and it can orchestrate agent workflows that use retrieved context for multi-step reasoning and tool use. RAGFlow is distributed under an Apache-2.0 license, maintained in public repositories (ragflow and ragflow-docs), and provides Docker-based deployment artifacts and documentation to support local and production deployments.