Conductive hydrogels hold great promise for next-generation wearable electronics and intelligent systems due to their ability to combine tissue-like compliance with electronic functionality. While liquid metal particles (LMPs) offer substantial potential for improving hydrogel performance, their tendency to coalesce, stemming from high surface energy and low viscosity, poses a significant challenge to dispersion stability. To overcome this limitation, this work presents a biopolymer-mediated stabilization strategy employing gelatin, whose abundant amino and carboxyl groups form coordination bonds with LMPs, effectively inhibiting their aggregation. The resulting uniformly dispersed LMPs enable rapid, initiator-free polymerization of acrylic acid, yielding hydrogels with outstanding mechanical properties, including high stretchability (> 700%), robust elasticity (46.5 kPa), and excellent fatigue resistance. Furthermore, these gelatin-metal hydrogels (GMHs) exhibit high electrical conductivity (0.15 S·m⁻¹) and pronounced strain sensitivity, with gauge factors of 1.85 and 3.66 for strains below 260% and between 260% and 500%, respectively. This combination of electrical characteristics allows GMHs to function as high-performance biomimetic electronic skins capable of precise and stable human motion monitoring. When integrated with convolutional neural networks (CNNs), the system further enables real-time handwriting recognition. Beyond sensing applications, the GMHs also demonstrate photo-thermal conversion capability, which can be leveraged for electricity generation. Overall, this study establishes a versatile interface-engineering strategy for the design of multifunctional hydrogels, contributing to the development of sustainable and scalable functional hydrogel systems.