Shellfish and salmonid aquaculture operations in Eastern Canada attract several mobile epibenthic species as a result of added structural complexity and increased food availability (bivalve fall-off and waste salmonid feed). It is not clear whether the aggregation of predators and scavengers below coastal farms contributes positively or negatively to their population dynamics, due to concerns about the quality of food items found under farms. We conducted an 18-month laboratory study to investigate the effect of diets composed of 1) mixed items, 2) mussels (Mytilus edulis), and 3) salmonid feed on the performance and condition of the rock crab, Cancer irroratus. Diet had no impact on crab survival but several negative consequences were observed in crabs fed the salmonid feed diet when compared to the mixed diet: reduced 1) moulting rates during the second growing season, 2) inter-molt growth, 3) gonad and hepatopancreas indices, 4) hemolymph dissolved compounds, 5) hepatopancreatic glycogen, and 6) shell hardness. Crabs fed the mussel diet had similar performance and condition when compared to the mixed diet. Fatty acid composition of muscle, gonad, and hepatopancreas tissues revealed that a salmonid feed diet decreased n3/n6 ratio when compared to a mixed or a mussel diet; those differences were mostly due to increases in the proportions of terrestrial (18:1n9 and 18:2n6) and decreases in proportions of marine essential (20:5n3 and 22:6n3) fatty acids. Together, these results point to a minimal impact of a mussel-only diet on crabs, whereas the salmonid feed diet resulted in negative impacts on condition. Our experimental results explored the consequences of a ‘worst-case scenario’ in which crabs were forced to feed on a single item for a long period of time; the realized impact in field settings will depend on other factors such as consumption of alternate food items underneath a farm, proportion of time spent in farms, and level of overlap between crab habitat and aquaculture facilities.