Recognize Plan-Worthy Tasks
Plan when a wrong assumption would be expensive; skip it when the task is one file and obvious.
Plan mode pays off when changes touch multiple files, are hard to reverse, or depend on assumptions about existing code. It's overhead when the task is one file, well-scoped, or already obvious. Ask for a plan when you'd want to catch a wrong assumption cheaply; skip it when you'd just be re-confirming what you already agreed on. Match the discipline to the work.
Where you'll practice this
One Promptles scenario teaches this principle directly.
Plan-Worthy or Not
It's Monday morning and you have four small tasks queued for Claude Code. Two of them are well-scoped one-file changes; two are multi-file changes with hidden assumptions. The temptation is to either ask for a plan on every task (slow, annoying) or just send 'go' on every task…
Other principles in Multi-Step
Architect Before You Ask
Ask for a Plan First
Request a plan before any multi-file change — approve the approach, then let it code.
Break the Workflow into Steps
Number each step so Claude builds pieces, not monoliths.
Combine Context with Action
Coordinate Across Files
Delegate to a Subagent
Spawn a subagent for research-heavy subtasks; brief it on scope, limits, and report shape.