This study aims to present a comprehensive framework elucidating the trajectory from trust establishment to user adoption intention of chatbots. Through a quantitative analysis of the role of trust in user adoption intention of chatbots, we seek to reconcile and clarify inconsistencies found in previous research and evaluate the robustness of its antecedents. In total, 54 papers comprising 18,707 samples were summarized through the meta-analysis. We categorized trust antecedents based on the Heuristic Systematic Model, subsequently dissecting trust into cognitive and emotional dimensions to scrutinize their impact on user adoption intention. The findings indicate that among systematic factors, chatbot competence and risk exhibit strong correlations with emotional trust, whereas competence and personalization are positively correlated with cognitive trust. All heuristic factors (anthropomorphism, social presence, social influence) demonstrate relatively strong positive correlations with both cognitive and emotional trust. The interaction between emotional and cognitive trust is affirmed, with trust significantly fostering user adoption intention of chatbots. Moreover, this study tests the moderating effect of sample characteristics (culture, IT penetration), chatbot features (text-driven vs. voice-driven, task-oriented vs. conversation-oriented), and usage industry. Theoretical contributions and practical implications are also derived toward the end.