GitHub Copilot
Definition
GitHub Copilot is an AI assistant that suggests code completions and can generate code from comments or natural language. It is powered by LLMs trained on public code.
It focuses on inline completion and short generation (e.g. from a comment or docstring). For deeper project context, chat, and edit-in-place, see Cursor. Both are examples of LLM-powered coding tools; Copilot is lightweight and IDE-agnostic.
How it works
As you type, Copilot suggests completions (gray ghost text); Tab accepts. As you add comments or docstrings, it can generate a function or block. You can accept, edit, or reject. The model sees the current file and nearby context (and optionally other open files, depending on the IDE). Integrations exist for VS Code, JetBrains, Neovim, and others. Languages: most mainstream languages and frameworks are supported. Suggestions are based on public code and your current context; quality varies by language and task. No project-level rules; for project-specific behavior, use Cursor rules or similar.
Use cases
GitHub Copilot is for inline completions and generating code from comments across many languages and IDEs.
- Inline code completion and snippet generation while coding
- Generating functions or tests from comments or docstrings
- Speeding up development across many languages and frameworks
External documentation
- GitHub Copilot documentation — Setup, usage, and best practices