Context management
Context management is the practice of organizing, updating, and governing what an AI system knows.
Category
Industry concept
See all terms
What is context management?
Context management is the practice of deciding what information an AI agent has access to, keeping that information current, and controlling who and what can reach it. It covers the full lifecycle of a fact inside a company: where it lives, how it gets structured, when it changes, and which agent or user is allowed to read it. Without it, an agent's knowledge is based on its prompt, training data, or tool calls. Context management enables a company to centrally update and maintain its knowledge base, ensuring that all agents work from current information.
How does context management differ from context engineering?
Context engineering is what happens inside a single session: choosing which documents to retrieve, how to phrase a system prompt, how much of the context window to spend on background information versus the task. Context management operates a level above that, on the context itself. Good context engineering on top of poorly managed context still produces a confident answer built on incorrect knowledge.
How does context management work in practice?
Context management starts with pulling knowledge out of the tools and conversations where it lives and turning it into structured records: entities like customers, employees, and projects, and the relationships between them. It keeps those records current as the underlying sources change, resolves the same entity appearing differently across two systems into one record, and applies access controls so a piece of context only reaches the agents and people that are intended to see it. Qontext's context repository handles this as an ongoing, automatic process, so every agent reads from the same current, permission-aware source.