Objective To summarize the major construction methods and phenotypic characteristics of Parkinson’s disease (PD) animal models reported in recent Chinese-language databases, with the aim of supporting the refinement of experimental model systems and subsequent mechanistic studies. Methods We systematically searched CNKI, SinoMed/CBM, Wanfang, and VIP databases for PD animal-model studies published from January 2022 to August 2025. Data were extracted on species, sex, age, modelling approach, modelling duration, and assessment methods. A database was built in Microsoft Excel 2024, and inductive analyses were used to organise the application patterns of model types and outcome measures. Additionally, we supplemented the dataset by extracting TCM-related syndrome information (syndrome label, syndrome judgment items, and syndrome-based modeling) for descriptive summarization. Results A total of 235 studies were included. Rodents predominated, with C57BL/6 mice and Sprague-Dawley rats most frequently used; males were used more often, and rodent age clustered around ~8 weeks. A small number of studies additionally employed other species such as Drosophila, Caenorhabditis elegans and non-human primates. Exogenous models comprised neurotoxin induction (e.g., MPTP, 6-OHDA, rotenone), multi-factor combination paradigms, and surgical lesions. Genetic-engineering models included overexpression or mutation of α-synuclein (α-syn), leucine-rich repeat kinase 2 (LRRK2), and PTEN-induced kinase 1 (PINK1), among others. Modelling durations were mainly acute (1–7 days) or chronic (4-8 weeks). Common readouts spanned behavioural performance, histopathology, and molecular assays, focusing on motor deficits, neuroinflammation, mitochondrial dysfunction, and synaptic plasticity. Among the studies in which TCM syndromes could be identified, 6.81% (16/235), the more frequent syndromes were Yin-deficiency with internal wind, Kidney-deficiency, and Liver-Yang hyperactivity; common judgment items included emotional irritability, dry/yellowed fur, and dry/constipated stools. Conclusions The PD animal-model toolbox is diverse. Neurotoxin models remain widely used owing to operational simplicity and reproducibility; multi-factor induction offers greater physiological relevance for disease progression; and genetic models facilitate dissection of specific pathogenic mechanisms. In future model construction, while integrating exogenous and genetic factors, it would be advisable to appropriately incorporate and standardize the reporting of TCM syndrome elements, and to introduce multi-omics and imaging technologies so as to enhance model stability and translational value, thereby laying a standardized foundation for studies of TCM and other integrated interventions.