Equipment, materials and methods for the measurement of the biomechanical parameters governing knife stab attacks have been developed and data have been presented that are relevant to the improvement of standards for the testing of stab-resistant materials. A six-camera Vicon motion analysis system was used to measure velocity, and derive energy and momentum during the approach phase of the attack and a specially developed force-measuring knife was used to measure three-dimensional forces and torque during the impact phase. The body segments associated with the knife were modelled as a series of rigid segments:​ trunk, upper arm, forearm and hand. The velocities of these segments, together with knowledge of the mass distribution from biomechanical tables, allowed the calculation of the individual segment energy and momentum values. The instrumented knife measured four components of load:​ axial force (along the length of the blade), cutting force (parallel to the breadth of the blade), lateral force (across the blade) and torque (twisting action) using foil strain gauges. Twenty volunteers were asked to stab a target with near maximal effort. Three styles of stab were used:​ a short thrust forward, a horizontal style sweep around the body and an overhand stab. These styles were chosen based on reported incidents, providing more realistic data than had previously existed. The 95th percentile values for axial force and energy were 1885 N and 69 J, respectively. The ability of current test methods to reproduce the mechanical parameters measured in human stab attacks has been assessed. It was found that current test methods could reproduce the range of energy and force values measured in the human stab attacks, although the simulation was not accurate in some respects. Non-axial force and torque values were also found to be significant in the human tests, but these are not reproduced in the standard mechanical tests.