LLM-governed social robots are increasingly positioned to decide who receives scarce assistance first in real-world settings. Since prioritization norms vary across cultures by age, status, and group size, failure to calibrate pluralistically can scale into unequal access. Yet LLM moral audits remain largely English-centered, rarely test embodied contexts, with pluralistic calibration remaining a diagnostic gap, one with intensifying risks in LLM-robot deployment. We introduce a gradient-based audit framework for multilingual evaluation of LLM moral trade-off behavior against cultural preference gradients. Grounded in nine cross-domain social robotics reviews (covering >8,000 studies), we derive symmetry-controlled scenarios across care, education, and services, translating the Moral Machine Experiment's “whom to spare” into “whom to assist first” dilemmas with preserved identity trade-offs (many vs. few; young vs. old; higher vs. lower status). We audit four LLMs across four country-language pairs in four prompting regimes (57,600 decisions), benchmarked against country-specific MME preference gradients. Ordinal concordance tests whether models differentiate between cultural contexts; a governance typology surfaces vulnerabilities in gradient differentiation, directional tendency, and deliberation behaviour. We find persistent, culturally asymmetric gradient tracking failures that prompting alone cannot reliably correct: quality calibration is nearly twice as strong for Western-language decisions as for Chinese and Japanese; high determinism in majority-first trade-offs often erases cross-cultural gradients; partial sensitivity to age- and status-based norms risks sidelining minority groups. Prompting effects are uneven; only contrastive cultural exemplars produce the most consistent gains, while reasoning-only prompts can worsen gradient tracking. Our results motivate multilingual, pluralistic audits as an LLM-robot pre-deployment gate and suggest that model-level factors are a more robust lever than prompting alone.