[Objective] To accurately reflect the carbon emission differences on the load side， fully account for the total carbon emissions generated during system operation， and explore the potential of low-carbon demand response， a low-carbon operation optimization model for multi-energy microgrids considering indirect carbon emissions from electricity transactions is proposed. [Methods] First， an operation optimization model of multi-energy microgrids was established based on the energy flow constraints of electricity， heat and gas， as well as the complementary demand response characteristics of multi-energy carriers. To precisely differentiate carbon accounting responsibilities among load-side entities， indirect carbon emissions from purchased electricity were quantified according to power transaction paths， and the source and path allocation of purchased power were incorporated into optimization decisions. Furthermore， a low-carbon operation optimization model for multi-energy microgrids integrating demand response was developed. The model was formulated in the MATLAB environment using the YALMIP toolbox and solved by the CPLEX solver. [Results] Compared with the conventional demand response scheme without indirect carbon emission costs， the daily system operation cost decreases by 3.90%， and the total system carbon emissions drop by 552.63 kg， representing a reduction of 50.84%. In contrast to the regional average carbon emission factor method， the indirect carbon emissions of the system under the power transaction contract path are reduced by 547.46 kg， achieving a decline of 44.23%. [Conclusions] The proposed method provides theoretical and model references for the economical and low-carbon operation of multi-energy microgrids， and effectively stimulates the enthusiasm of end-users for low-carbon response.