This paper presents a novel edge-aware transformer-based framework for camouflaged object detection. By integrating a spatial attention module guided by structural cues extracted from edge information, the model is directed toward visually ambiguous regions that commonly hinder segmentation performance. This design enables more effective global context modeling and improves the delineation of camouflaged object boundaries. Extensive experiments on multiple benchmark datasets validate the effectiveness of the proposed approach, demonstrating consistent performance gains over state-of-the-art methods in all key evaluation metrics.