Inthis article, we propose a novel, model-independent, unified detection strategy based on disagreement Laplacian potential for effective identification of cyber anomalies in interconnected autonomous direct current (dc) microgrid (MG) clusters. This interconnection enables inter and intra-microgrid power exchange between them, thereby rendering efficient and maximum utilisation of the distributed energy resources. However, these clusters are vulnerable to cyber intrusion that disrupt the power sharing schemes. As such, we have investigated the effect of false data injections at various vulnerable points, including both voltage and current sensors in primary, and data transmission points in the distributed secondary, local, and global tertiary layers. Simultaneously, a logical evaluation in the form of data-integrity indices are formulated from the outcomes of the proposed detection scheme to provide for their timely mitigation. Ultimately a fault-ride through operation of the converters is achieved instead of alienating the misbehaving component, unlike previously reported works.