Modes: Ask, Agent, and custom postures
So far you’ve let Cursor run wherever it landed by default - you opened the chat, typed a request, and it edited files and ran commands without you thinking about how much freedom it had. On a one-line fix that’s fine. The moment a change reaches into code a dozen call sites depend on, it stops being fine, and the thing that controls it is the mode.
A mode is not a different model and it’s not a different chat. It’s the same Cursor reading the same project at a different posture - how far it’s allowed to go before you look. Cursor’s current mode surface is Agent, Ask, and Custom; Plan and Debug are common custom-mode patterns, and the exact built-ins can change between releases. The skill this chapter teaches is not memorising a shortcut. It’s the judgment: given what’s in front of you, which posture does this task actually want?
The mode picker
Section titled “The mode picker”Use the mode picker next to the chat input, or the current keyboard shortcut shown in Cursor’s mode documentation and shortcut settings. Cursor documents Ctrl+. for opening or switching modes; shortcuts can vary by platform and release. (mode documentation)
The current picker centres on these:
- Ask - read-only Q&A. It answers and explains; it does not change your files.
- Agent - full execution: it plans, edits across whatever files it judges relevant, runs commands, and self-corrects until the task is done. In practice this is the mode you’ll sit in most.
- Custom - user- or product-defined postures. A Plan posture can clarify, research, and hand you an editable markdown plan; a Debug posture can reproduce and instrument a failure before fixing it. Treat those names and their exact controls as version-sensitive.
The reason there’s a chord for this and not just a setting buried in preferences is that you’re meant to switch often - sometimes several times inside one piece of work. The natural rhythm is to Ask to understand unfamiliar code, Plan the approach, then drop into Agent to build it. That’s one task climbing the ladder of latitude as your confidence grows, not three separate conversations.
The moves of this chapter
Section titled “The moves of this chapter”- Ask & Agent - the read-only detent for when you can’t yet name the change, then full execution on a contained change where reviewing a finished diff beats narrating every step - Ask and Agent: read-only Q&A, then full execution
- Plan - a custom posture for clarifying questions, read-only research, and an editable markdown artifact before a line of code is written - Plan: clarify, research, then a plan you can edit
- Debug - a custom posture for bugs a straight Agent pass can’t crack, plus the corresponding CLI workflow where available - Debug: a custom debugging posture
The arc climbs from understanding to execution, then adds custom postures where the task benefits from a different procedure. The first lesson changes what you think the job is before you commit to it. The second hands over the latitude to do the work. The Plan and Debug lessons show how to encode two useful postures without treating their names as permanent product guarantees. By the end you shouldn’t be picking modes off a list. You should feel which posture a task is asking for before you type a word.