Summary Background First tarsometatarsal joint (TMT-1) hypermobility might cause hallux valgus deformity (HV), and recurrence following surgical correction. Anatomic findings, indicating tibialis anterior tendon (TAT) involvement in TMT‑1 stabilization, led to the development of cross-glide test allowing clinical TMT‑1 stability testing. Cross-glide test function was evaluated in anatomical specimens and in the clinical setting, compared to simulated weight-bearing computer tomography (CT) analysis. Methods Cross-glide test was evaluated in 6 healthy lower leg specimens before and after TAT transection. Clinical testing was performed prospectively in 36 feet (6 controls, 21 HV, 9 recurrent HV); consecutive weight-bearing CT analysis was performed. Results from clinical testing were compared to CT analysis. Results TMT‑1 instability significantly increased in anatomic specimens following TAT transection (p = 0.009). In the clinical setting, all healthy feet were cross-glide test negative, 62% of HV cases and all recurrent HV feet were positive. In the CT analysis- Compared to controls the HV cases revealed significantly increased MT‑1 internal rotation (p = 0.003) and decreased dorsal angle (p = 0.002), considered as collapsing forefoot signs; HV recurrent cases revealed similar results. Positive cross-glide tested cases revealed increased MT‑1 internal rotation values (p < 0.001) and dorsal angle values (p < 0.001) in CT analysis. Strikingly, cross-glide test positive HV cases revealed significantly increased internal TMT‑1 rotation (p = 0.043) in CT analysis, and HV and IMT (intermetatarsal) angle were significantly higher (p = 0.005, p = 0.006). 15 HV recurrence cases, treated with TMT‑1 arthrodesis, revealed no recurrence during follow-up. Conclusion Cross-glide test allows reliable clinical TMT‑1 instability testing, via TAT tension, and is less laborious than CT analysis. We recommend TMT‑1 arthrodesis in cases with instability in clinical testing, to avoid HV recurrence.