Previous research has demonstrated a close connection between the development of osteosarcoma (OS) and variations in the abundance of specific gut microbiota (GM). Generally speaking, GM play a role in human health mostly through metabolites. However, the causal relationship between GM, plasma metabolites, and OS remains unclear. Hence, in our study, we aim to clarify this relationship between GM, plasma metabolites, and OS, by employing a Mendelian randomization (MR) approach. In this study, pooled data were derived from the public genome-wide association study (GWAS) in GM (GCST90032172 to GCST90032644), plasma metabolites (GCST90199621 to GCST90204603) and OS (finngen_R10_C3_OSTEOSARCOMA_EXALLC). The two-sample and two-step MR methods were used for the current analysis: (1) genetic causality between GM and plasma metabolites on OS; (2) mediation effects of plasma metabolites. For evaluating the previously described causal relationship, the inverse variance weighted (IVW) method was primarily used, with complementary approaches including the weighted median, MR-Egger, weighted mode, and simple mode. Moreover, the MR-Egger intercept test and the mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) were employed to assess the horizontal multiplicity. The reliability of causality is verified by "leave-one-out" sensitivity analysis and the Cochran's Q test for heterogeneity. The STROBE-MR checklist for the reporting of MR studies was used in this study. First, according to the IVW results, 13 types of GM, specifically, were identified to have a potential causal relationship with OS. After FDR correction, Phocea was defined as a strain with a clear causal relationship with OS (FDR-adjusted p < 0.05). Second, a total of 48 plasma metabolites were identified to have a potential causal relationship with OS, including 30 currently known metabolites, 7 metabolites not yet studied, and 11 metabolite ratios. Finally, we further explored whether plasma metabolites mediate the causal relationship between Phocea and OS. And as a result, two plasma metabolites, Eugenol sulfate levels (mediated proportion: 7.74% (14.2%, 1.3%)) and N-acetylphenylalanine levels (mediated proportion: 3.52% (6.18%, 0.867%)), that may mediate the causal link between Phocea and OS were identified. All of the above results were subjected to sensitivity analysis. The causal relationship between GM, plasma metabolites, and OS was revealed in this MR study. Importantly, this study also demonstrated the mediating role of plasma metabolites levels of Eugenol sulfate levels and N-acetylphenylalanine levels in modulating the causal relationship between Phocea and OS. Of course, further research needs to be conducted to verify the above findings.