Skip to content
Orbit Docs

Deep Review

← All terms · Orbit-internal concepts

A more thorough AI review of a queued prompt, typically run on Opus high-effort or another premium model. Used when stakes warrant extra scrutiny.

What it is

Deep Review uses a more capable (and expensive) model to evaluate a draft prompt — typically Opus high effort. Configured per tier; high-complexity tier defaults to manual reminder, where the user can choose between Auto Review (Flash) and Deep Review (Opus) depending on the situation.

When you would use it

You run Deep Review when a prompt's blast radius is large enough that a thorough review pays for itself — typically refactors, architectural changes, or anything affecting production.

Common operations

  • Reviewing high-stakes prompts with deeper analysis
  • BYOK premium-model review for important work

Related terms

Where this is taught

No learning path uses this term yet. Browse the Learning Atlas for guided sequences through related ideas.