Now in closed beta.Book a demo to get started.
Use cases

Migrate off a deprecated library at each call site

Dependency and runtime upgrades|Everything this ticket has to prove is checked by machine, so it can be handed over as it is. A person still reads the pull request and decides whether to merge it.

A library the ecosystem moved on from, swapped out one call site at a time with a behaviour test at each.

The ticket

Replace a deprecated library with its maintained equivalent at every call site.

Acceptance criteria

  • No file imports the old library
  • Each migrated call site has a test covering the behaviour it relies on
  • The old package is removed from the manifest and the lockfile
  • Any behaviour the replacement does not cover is named in the description

What lands as proof

A search showing the old import is gone, and a test per call site proving the behaviour survived the swap.

Why teams defer it

  • The replacement is never a drop-in. The first call site takes an hour of reading and the rest take ten minutes each, but nobody knows that until they have done the first one.
  • Deprecation warnings are easy to filter out of a log, so the deadline never feels real.

Questions

What does the agent actually change?
The ticket is scoped to one outcome: replace a deprecated library with its maintained equivalent at every call site. 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 behaviour test at each site, the old import gone, so a reviewer can confirm the result without reproducing the work locally.
How much oversight does this need?
None is required before the run starts. Every claim this ticket makes can be checked by machine, so the agent works straight through and a person reviews the finished pull request like any other.

Ready to put the fleet to work?

Contact us for a demo with an expert.