Climate risk assessments for critical infrastructure are central to identifying and predicting vulnerabilities early in the asset life cycle, enabling proactive mitigation before impacts occur through the implementation of appropriate technical and nature-based solutions. However, such assessments often rely on quantitative indices that can be difficult for non-technical stakeholders to interpret. To address this challenge, this paper presents an open-source decision support platform that combines OpenStreetMap site characterization, qualitative pre-screening, a quantitative IPCC AR6-aligned risk chain, and a downstream nature-based solution (NbS) recommendation layer. The approach uses Large Language Models (LLM) to translate analytical outputs into accessible narrative explanations. End-to-end site-characterization processing across three tested European demonstration sites took between 29 and 70 seconds. An exploratory ablation study investigated the faithfulness of AI-generated explanations using three complementary metrics. It showed that the generated hazard assessments remained factually grounded and free from fabricated numerical values. Including example reports in the prompt further improved the reliability of explanations for more complex risk indicators. While the results demonstrate the potential of AI-assisted climate risk communication, expert evaluation of stakeholder utility is identified as the most important next step.