← Back to Terminology Glossaryconcepts

Context Window

The maximum amount of text (tokens) that an LLM can read and process in a single generation query.

The context window dictates how much of your codebase the AI model can understand at once. Managing your context window—by removing heavy build folders, pruning large files, and using fine-grained context commands (like Cursor's @files or Claude Code's narrow search)—is critical to preventing hallucinated bugs.