This project investigates how avatar agency impacts practices of ascribing moral responsibility in emergent avatar communities (Metaverse, augmented Metaverse, and Human-Cybernetic Avatar societies). These communities challenge traditional frameworks of agency and moral responsibility. The unprecedented persistent and continuous use of multiple avatars by both humans and organisations, has the potential to permeate prior boundaries of the physical, the augmented and the virtual world. We propose a dynamic normative framework of ascribing moral responsibility that is grounded in a philosophical conceptualisation of avatar agency. This framework considers the way interactions between avatars, individuals, and organisations generate intertwined and mutual ascriptions of moral responsibility, potentially enhancing or, on the contrary, diminishing human agency and responsibility in emergent avatar communities. We construct our approach along five main objectives: 1) to clarify avatar agency and related responsibility pertaining to individual, organisational, and artificial members of emergent avatar communities; 2) to delineate the way avatar agency enhances or restrains individual agency and responsibility; 3) to correlate the way interactions between social entities in avatar communities generate mutually enhancing (or perhaps decreasing) ascriptions of moral responsibility to all participants; 4) to evaluate the extent in which the ethical characteristics of the community environment can generate normative attributions of responsibility; 5) to project normative evaluations of blaming and praising practices in avatar communities. The successful delivery of this ethical framework is an original contribution to the problem of moral responsibility and bears direct implications for future regulation, impacting societal moral practices of accountability and blaming.