Context engineering
lastmod 2026-11-10created 2026-04-143 edits
Context engineering is the practice of designing what information goes into a model's context window, and in what form, when an LLM-based agent performs a task. Going beyond prompt engineering, which polishes the wording of instructions, it covers the selection and compression of system prompts, tool definitions, retrieved documents, conversation history, and memory as a whole. Its core is managing the problem of context overflowing or getting polluted during long tasks.