Climate adaptation decisions involve heterogeneous stakeholders, partial information, and institutional constraints. This paper presents a role-based Multi-Agent Computer Model (MACM) to simulate climate adaptation deliberation across Living Labs. The main contribution is a configurable architecture that keeps behavioural mechanisms fixed while externalising context-specific inputs. Each stakeholder is represented as one agent that may play one or more roles: Expert Evaluator, Disseminator, Positioning Agent, and Decision-Maker. The simulation proceeds through four phases—initialization, information exchange, positioning and influence, and final decision-making. We argue that this role-based design improves both interpretability and cross-case reusability in social simulation of climate adaptation governance.