Assessing the trade network connectivity is essential for understanding the trade network structure, optimizing trade development patterns, and improving uneven trade development along the “Belt and Road” (BRI). From the perspective of connectivity, this paper integrates the frontier algorithms in network science and constructs an analytical framework to identify the mesoscale structures, including the community structure, core-periphery structure, and backbone structure embedded in the network, and further explore the structural connectivity of the BRI trade network. The results show that: (1) The BRI trade network represents a trade pattern of “one superpower, many great powers”, with three major trade groups in Southeast Asia, the Middle East, and Northern Central and Eastern Europe in terms of geographical space. China is the super core of the BRI trade network, and the most considerable trade links are all centred in China. (2) Five distinctive trade blocs have formed in the BRI trade network. Nevertheless, the structure of the trade blocs shows significant geographical proximity, indicating that geographical distance still plays a vital role in the international trade system at the regional scale. (3) The BRI trade network demonstrates a significant core-periphery structure, with apparent trade clustering among the core countries within the trade network. Among them, nine countries led by China constitute the core structure, and the peripheral structure is large, reaching forty-four. (4) The trade links with China constitute the backbone structure of the whole trade network in the BRI region. In addition, the trade links related to energy trade and re-export trade are also crucial components of the BRI backbone structure. Methodologically, the analytical framework proposed for assessing the network structural connectivity has great potential to be widely applied to other disciplines and fields.