In this work it was demonstrated that sample endogenous polyphenols are selectively driving the gold-nanoparticle (AuNPs)-formation process when representative food samples were used as natural sources of reducing compounds. The process of AuNPs formation was characterized by UV-visible spectroscopy and was described by a sigmoidal curve (R (2) ≥ 0.990) which gave information about the polyphenol concentration at which the localized surface plasmon resonance (LSPR) absorption reached its half-value, X (c) (50) , and about AuNPs production per polyphenol concentration unit, K (AuNPs). The behavior of phenolic acids was different, with lower X (c) (50) and higher K (AuNPs) values than flavonoids. For the food samples tea, apple, pear, wine, and honey X (c) (50) values were 0.22, 7.3, 11.5, 20.4, 30.3, and 53.5 (mg mL(-1)) and K (AuNPs) values were 28.7, 0.70, 0.60, 0.20, 0.14, and 0.10 (mg(-1) mL), respectively. Excellent correlation between K (AuNPs) and total phenolics (TP) was obtained (r = 0.98, p-value < 0.05), implying K (AuNPs) is a novel marker for evaluation of food sample antioxidant capacity in vitro. The K (AuNPs) values of samples indicated their antioxidant capacity was in the order: tea > apple > pear > wine > honey. The reproducibility of the AuNPs formation approach was excellent, not only for polyphenol standards (RSD < 6 % for X (c) (50) and RSD < 11 % for K (AuNPs)) but also for food samples (RSD < 9 % for X (c) (50) and RSD < 15 % for K (AuNPs)). Transmission electronic microscopy (TEM) enabled confirmation of the formation of stabilized Au-nanospheres from endogenous polyphenols with very well-defined sizes under 20 nm diameter for all the food samples investigated.