TrustRAG is a configurable and modular Retrieval-Augmented Generation framework designed to provide reliable input and trusted output. It features a DeepResearch component that achieves deep information search through layered queries, recursive iteration, and intelligent decision-making. The system parses user queries into sub-queries for intent understanding, includes processing condition judgment for token budget and action depth limits, and allows users to flexibly adjust and optimize each component according to specific application needs.