Foraging is essential for honey bee survival and colony reproduction, underpinning the pollination services that support agriculture, plant biodiversity, and ecosystem health. Using an agent-based model, we address how colonies allocate foraging effort across multiple resource sites under changing environmental conditions. Contrary to existing task allocation models of honey bee foraging, we introduce a holistic approach to dissect foraging by incorporating positive and negative drivers identified in the literature. To this end, we utilize a Finite-State Machine framework to simulate task allocation for varying food source qualities, predator stress on foragers, and unloading constraints. The model holistically integrates waggle dancing (recruitment), tremble dancing (unloading-limitation feedback), and stop signaling (predation-risk inhibition) within explicit behavioral states, using parameters grounded in empirical field experiments for biological realism. In simulations, the colony self-organizes into stable distributions of retrieval, recruitment, and inhibition behaviors, reallocating effort in response to dynamic changes in food quality, predator risk, and receiver availability. These results support the model as a compact, behaviorally grounded platform for ecological analysis of foraging dynamics and as a basis for future extensions toward richer environmental drivers and predictive colony-level monitoring.