Context Engineering vs Prompting

Context engineering is the broader discipline of designing everything a model sees, not just the user prompt. It encompasses system prompts, conversation history, retrieved documents, tool definitions, few-shot examples, and tool results that together shape model behavior — while prompt engineering focuses narrowly on crafting individual instructions. This distinction matters because in agentic systems, the prompt is a small fraction of the total context; the majority is dynamically assembled from tool results, retrieved data, and conversation state. Mastering context engineering is about architecting the entire information environment that an agent operates within, which determines the ceiling of what the agent can accomplish regardless of how good the underlying model is. This concept connects to system prompts and few-shot examples as individual components of context, context window budget for managing the finite space, and context assembly pipelines for automating context construction.