Cursor Master Quiz
Eight questions spanning everything Cursor — the inline edit loop, @-context, Rules, agents, indexing, and the context mistakes that produce confident edits in the wrong file. Take it as a diagnostic: whatever you miss tells you which topic to read next. No sign-up, instant score, and you can share how you did.
-
Q1 What does the Cmd-K inline edit loop do?Why
Select, describe, review the diff, apply. It's the fastest and most controlled way to use AI in Cursor — controlled precisely because the change stays scoped and visible.
-
Q2 What are @-mentions (@file, @folder, @codebase, @docs) for?Why
They replace hoping retrieval guesses right with telling the model precisely what to look at. If you already know the file, name it.
-
Q3 You add a whole folder as context 'to be safe'. What happens?Why
Extra context isn't neutral — it dilutes the signal. 'To be safe' is the instinct that quietly makes retrieval worse.
-
Q4 What's the most common way Cursor goes wrong?Why
Cursor acts on its best guess, and its best guess is well-written and wrong. Explicitly pointing at the relevant files is the whole fix.
-
Q5 You're handing Cursor a bug. What does it need?Why
A paraphrased error hides the actual location — the stack trace is the location. Without @-mentioned files, the model searches and guesses in a big repo.
-
Q6 What are Cursor Rules?Why
Rules encode conventions once, in the repo, so every edit follows them. Keep them concrete — 'write clean code' gives the model nothing to act on — and keep them current.
-
Q7 What's the reliable shape of an agent run?Why
The plan is where a wrong assumption is cheap to catch. Skip the checkpoint and your first look at the work is a huge diff built on a premise you never agreed to.
-
Q8 Why does leaving node_modules and build output indexed hurt?Why
The index is a map, and it's only useful if it maps the things that matter. Retrieval quality is context quality.