Context window
Everything a language model has in front of it at a given moment: your instruction, the files it has read and the results of its tools. What isn't in the context, the model can't use.
The context has three limits that matter in data work:
- It's limited. In long sessions the harness summarizes or drops older parts to make room, so an instruction from the start can get lost.
- It doesn't carry over by default. A new session starts blank: what you explained yesterday isn't included.
- It's all the agent knows about your data. If you don't tell it a table has several rows per customer, it doesn't know.
Common confusion: "the model already read my file, so it knows". It knows while the file is still in this session's context. Tomorrow, in another session, you give it again or leave it in an instructions file the harness loads at the start.
A quick test: at the start of a session, ask the agent what it knows about your table before asking for anything. Whatever it doesn't mention, it doesn't have.
Learn more
Related terms
- AI agent
A program that pairs a language model with tools, such as terminal commands and files, to carry out a task in several steps: it reads its context, acts, looks at the result and repeats.
- Agent harness
The program that wraps a language model and makes it an agent: it gives the model tools, asks for your permission, keeps the session and decides what goes into its context.
- Agent instructions file
A text file in your project that the agent's harness loads at the start of every session, so the agent knows from the start what it always needs to know: rules, facts about the data and how to check its work.