Context engineering
Deciding what goes into the model’s context, in what order and when, across a whole task.
Prompt engineering is about the wording of one request. Context engineering is about the whole flow: which instructions are always present, which files load on demand, what tool results are trimmed, what gets summarized. For agents it matters more than any single prompt.
Related terms
- Context window: The maximum number of tokens the model can read and write in one request.
- Agent Skills: Folders with instructions, scripts and references that an agent loads only when the task needs them.
- RAG (retrieval-augmented generation): Finding relevant documents first and putting them into the prompt, so the model answers from them.
