Seafood mislabeling threatens customer rights and causes economic loss worldwide. The information on seafood misrepresentation in Thailand is still lacking, and the investigation and monitoring program must be well established. This study investigated the mislabeling status of imported seafood in Thailand using the DNA barcoding technique. A total of 45 imported seafood products from five distributors were included. Scientific, common, local, and market names of seafood samples were obtained from FAO and Fishbase databases. DNA was extracted, and PCR was performed using a universal primer targeting the COI gene. Species of each sample were identified with over 98% similarity based on COI sequence analysis. DNA sequence revealed 11 mislabeled samples. Among substituted species, Pangasianodon hypophthalmus and Thunnus maccoyii were found to be endangered species according to IUCN status. Products obtained from Brand-C showed the highest mislabeling rate (42.85%). The phylogenetic analysis adopted with the TIM2+F+I+G4 model showed the sequenced DNA similar to the NCBI database reference sequence. Overall, mislabeled products of imported seafood were found at the rate of 24.44%, suggesting that strict surveillance for seafood substitution should be implemented in Thailand.

New Zealand green-lipped mussel (Perna canaliculus) is a premium seafood product that may be substituted with morphologically similar mussels after shucking and cooking, particularly Asian green mussel (Perna viridis). This study developed a rapid, on-site duplex recombinase-aided amplification–lateral flow dipstick (dRAA–LFD) assay to authenticate P. canaliculus and differentiate it from P. viridis. Species-specific primers were designed from mitochondrial COI alignment and combined in a dRAA reaction. Reaction conditions were optimized at 37–42 °C and 15–60 min. Specificity was assessed against 11 non-target seafood species, and sensitivity was evaluated using 2-fold serial dilutions. The assay was further validated using DNA from boiled (85 °C, 5–15 min), steamed (105 °C, 10–30 min), and fried (185 °C, 30–90 s) mussels, and 15 restaurant products labeled as New Zealand mussel dishes. Optimal performance was achieved at 40 °C for 30 min, with no cross-reactivity. The LFD detection limits were 0.05 ng/reaction for P. viridis and 0.2–0.1 ng/reaction for P. canaliculus. All cooked samples remained identifiable, and commercial testing classified 13/15 products as P. canaliculus and 2/15 as P. viridis. Overall, the dRAA–LFD assay enables rapid, equipment-light authentication of cooked mussel products for routine screening.