Financial fraud has a destructive effect on corporate reputation and market trust, and it is necessary to choose effective technical means and discriminative information to identify enterprises involved in financial fraud. This paper constructs a financial fraud identification model based on the XGBoost algorithm and introduces financial and non-financial variables to discriminate fraud. The model in this paper, using the XGBoost algorithm, achieves 94.11% accuracy in 50 seconds. Incorporating ESG information to analyze the financial fraud identification mechanism of a company, it is found that the company has financial fraud risk factors in environmental information disclosure, employee and supply chain management, governance structure, and monitoring system. This suggests that the model proposed in this paper has a better identification effect, and the integration of ESG information can provide a more in-depth understanding of the mechanisms of corporate financial fraud.