AI code completion
AI code completion predicts and suggests the code you are about to write, usually as inline grey text you accept with Tab. It works from the current file, nearby open files, and recent edits, and operates at the scale of a line, a block, or a function.
It is the oldest capability in this category and still the one people use most. Its strength is that it costs nothing to ignore: a wrong suggestion is dismissed by continuing to type.
Its limitation is scope. Completion sees local context and pattern-matches against it, so it is excellent at the obvious next line and unaware of the wider design. It will happily complete code that fits the file and contradicts the architecture.
What this means in practice
GitHub Copilot is still the strongest here, with Cursor's Tab and Windsurf's Supercomplete close. Claude Code has none at all, which is the main reason people run a terminal agent alongside an editor assistant rather than instead of one.
Prompts for the tools this applies to
Related terms
- AI code generation
AI code generation is producing working code from a natural-language description — a function, a component, a module, or an entire application.
- AI pair programming
AI pair programming is working alongside an AI assistant the way you would with a human pair: it suggests, you steer, and you review continuously rather than delegating a task and checking back later.
- Context window
The context window is the maximum amount of text a model can consider at once, measured in tokens and covering everything in the request: system prompt, your instructions, the files supplied, prior conversation, and the response being generated..