#agents

Spelling out do-this-then-that is now the common failure mode. Describe the goal, the constraints and how Claude knows it is done, then leave the method alone.

Spin up a subagent for codebase search or verification so only the conclusion returns to your main context.

Write the plan in one session, then open a fresh Claude with no context and ask it to review the plan like a staff engineer.

Use `claude --worktree` (or git worktrees) so multiple agents edit isolated checkouts without colliding.

Define custom subagents with `isolation: worktree` so delegated edits happen on a temporary branch checkout.

Write the tests first, then set the goal to making them pass. Claude keeps working across turns until the condition is met instead of stopping after each reply.