Using Complexity and Priority Effectively

Complexity is your estimate of how much effort a task needs:

• Small: under 1 hour, usually one file
• Medium: 1 to 2 hours, a few files
• Large: more than 2 hours, several parts of the code

These times are a rough guide, and nothing enforces them. A task bigger than Large is better split into a goal with smaller tasks.

The field is hidden on a new board. The board owner turns it on with Complexity & Scope under Field visibility in the board's Settings.

Priority is always on the task form, and new tasks start at Medium:

• Critical: drop everything, such as a security issue or a production bug
• High: important work that should happen soon
• Medium: normal work
• Low: nice to have, done when there's time

When an agent asks for work, Stride looks only at the Ready column. It offers the highest priority task first and, among tasks with the same priority, the one nearest the top of the column.

It skips goals, blocked tasks, tasks marked Human Task, tasks assigned to someone else, tasks whose key files overlap work already in Doing or Review. When the agent's API token lists the agent's skills, it also skips tasks that need a skill the agent doesn't have.

Complexity doesn't change the order.

Consider both together:

• High priority and small: a quick win, do it now
• High priority and large: important, but plan it and think about splitting it
• Low priority and small: fills the gaps between bigger tasks
• Low priority and large: may never get done, so ask whether it's needed

A large task can take longer than a claim's 60 minutes, which is another reason to split it.

Revisit complexity and priority as work progresses:

• Raise the priority when something becomes urgent
• Drag tasks within Ready to change the order of tasks with the same priority
• Correct the complexity when an estimate was wrong
• Lower the priority of tasks that matter less now

Keep the board matching reality.