1 min lesson
Self-test product depth, no notes
For each case in "Self-test product depth, no notes", name the signal and the response you would use.
Step 1 of 2
Self-test product depth, no notesDrill 2
Set a timer and explain each surface out loud as if to a non-Cursor engineer. If you stall, that surface is a study gap, not a nerves problem.
What it predicts vs ⌘K vs Agent.
When it feels magic and when it gets in your way.
One failure mode you've actually hit.
Multi-file edits, runs commands, iterates.
Where it loops or over-edits a diff.
How you supervise it on a risky change.
Pulling files, docs, symbols into the prompt.
Why grounding beats a blind ask.
What happens when you over-stuff context.
Embeddings over your repo power retrieval.
Symptoms of a stale or failed index.
Why a fresh clone sometimes answers worse.
The tell that ends a screen early: describing Agent modeCursor's full-capability mode: the AI can read the codebase, write and edit files, move them and run terminal commands. Contrast with Ask mode, which is read-only. Press Enter for the full definition. and Tab as if they're the same thing or calling indexing “the search.” A real power user has felt the seams between these and can name a moment each one failed them. Vague praise reads as someone who installed Cursor for the interview.