Capital breeders accumulate nutrients prior to egg development, then use these stores to support offspring development. In contrast, income breeders rely on local nutrients consumed contemporaneously with offspring development. Understanding such nutrient allocations is critical to assessing life‐history strategies and habitat use. Despite the contrast between these strategies, it remains challenging to trace nutrients from endogenous stores or exogenous food intake into offspring. Here, we tested a new solution to this problem. Using tissue samples collected opportunistically from wild emperor penguins Aptenodytes forsteri, which exemplify capital breeding, we hypothesized that the stable carbon (δ13C) and nitrogen (δ15N) isotope values of individual amino acids (AAs) in endogenous stores (e.g. muscle) and in egg yolk and albumen reflect the nutrient sourcing that distinguishes capital versus income breeding. Unlike other methods, this approach does not require untested assumptions or diet sampling. We found that over half of essential AAs had δ13C values that did not differ between muscle and yolk or albumen, suggesting that most of these AAs were directly routed from muscle into eggs. In contrast, almost all non‐essential AAs differed in δ13C values between muscle and yolk or between muscle and albumen, suggesting de novo synthesis. Over half of AAs that have labile nitrogen atoms (i.e. ‘trophic’ AA) had higher δ15N values in yolk and albumen than in muscle, suggesting that they were transaminated during their routing into egg tissue. This effect was smaller for AAs with less labile nitrogen atoms (i.e. ‘source’ AA). Our results indicate that the δ15N offset between trophic‐source AAs (Δ15Ntrophic‐source) may provide an index of the extent of capital breeding. The value of emperor penguin Δ15NPro‐Phe was higher in yolk and albumen than in muscle, reflecting the mobilization of endogenous stores; in comparison, the value of Δ15NPro‐Phe was similar across muscle and egg tissue in previously published data for income‐breeding herring gulls Larus argentatus smithsonianus. Our results provide a quantitative basis for using AA δ13C and δ15N, and isotopic offsets among AAs (e.g. Δ15NPro‐Phe), to explore the allocation of endogenous versus exogenous nutrients across the capital versus income spectrum of avian reproduction.