Cursor

Cursor is a VS Code fork from the company Anysphere that turns AI from a completion plugin into a way of working. The key difference from completion tools is that it reads the whole project: you can say “make this function’s error handling match the rest of the codebase” and it will go look at the rest of the codebase.

Multi-file changes are where it earns its keep — renaming a concept, adjusting an interface and updating every call site. That work is far faster here than pasting back and forth into a chat window. Changes are presented as diffs, which you can accept piece by piece or all at once.

The cost is switching editors. If your settings, keybindings and extensions live elsewhere, the migration is real, and the VS Code version it forks from sometimes lags upstream.

Key features

• Tab completion: goes beyond the current line and predicts where you will edit next, so repeated Tabs carry a change through.
• Inline edit: select code, describe the change in plain language and review the result as a diff.
• Agent chat: describe a task in the sidebar and it searches the codebase, edits multiple files and runs terminal commands.
• Codebase indexing: the project is indexed so relevant files are found automatically; you can also @-mention files or docs.
• Project rules: write your team’s conventions into rule files so the AI follows them on every edit.
• Model choice: pick between models from several vendors; a subscription raises limits on the stronger ones.

How to use

1. Download the installer for your OS from the official site, install it and sign in.
2. On first launch, import your VS Code settings, keybindings and extensions in one step.
3. Open a project folder and let it finish indexing.
4. Use inline edit for small changes and the agent for multi-file tasks; read each diff before accepting it.

Best for

• Cross-file refactors: interface changes, renames and style unification, with every call site updated.
• Getting to grips with an unfamiliar codebase: asking “where is this implemented?” beats global search.
• Existing VS Code users: the interface and extension ecosystem are nearly identical.
• Solo developers and small teams who want completion and agent work in one tool.

Strengths and limitations

Strengths

• Strong whole-project understanding; multi-file edits are much faster than copy-pasting into chat.
• Next-edit prediction in Tab completion speeds up ordinary editing too.
• High compatibility with the VS Code ecosystem, so most extensions just work.

Limitations

• You have to switch editors, which is costly for JetBrains and other ecosystem users.
• The underlying VS Code version sometimes lags upstream, and some extensions misbehave.
• Heavy agent use burns through quota quickly, and the pricing model has changed more than once.
• Reachable from mainland China, though signing in occasionally takes patience.