Route one call path to the replacement, reversibly
One call path moved to the replacement behind a flag, with both paths tested and the switch reversible.
The ticket
Route one call path to the new implementation behind a reversible switch.
Acceptance criteria
- Both the old and the new path are covered by tests
- The switch is a flag that can be reverted without a deploy
- Behaviour is identical on both paths, proven by a comparison test
- The plan states which path comes next
What lands as proof
A comparison test asserting both paths produce the same result, which is what makes the switch safe to flip.
Why teams defer it
- Each step is small and the sequence matters, so it needs somebody holding the whole plan in their head.
- Running two implementations at once doubles the surface until the migration finishes, and migrations stall.
Questions
- What does the agent actually change?
- The ticket is scoped to one outcome: route one call path to the new implementation behind a reversible switch. Work that serves that outcome is in scope, and anything outside it is left for a separate ticket, so the pull request stays reviewable.
- How do I know the work is done?
- The pull request carries the evidence, not only the diff. Here that means both paths under test, the switch documented and reversible, so a reviewer can confirm the result without reproducing the work locally.
- How much oversight does this need?
- The run stops for the plan, and the review bar is raised afterwards so that smaller suggestions block the merge too, not only serious findings.