Distributed Storybook testing.

Putting renders inside the coding agent's loop can multiply render counts by 100x to 1000x.

The extra snapshots are wildly effective, but wildly unaffordable with a pay-per-snapshot service like Chromatic.

Selfie lets your agent take its own snapshots.

$0 per snapshot

Get started on GitHub
  • ~500 stories per minute on M4 MacBook Air
Join the list

The agent checks its own work

# JSON naming the diff images
selfie review

# after looking at them
selfie approve <batch> --all
git commit

# nothing left to ask
selfie review
  • One JSON document. selfie review names every changed image, and its exit code says whether anything blocks.
  • It knows what kind of change it is. What moved or gained spacing is kept apart from what really changed, so the harmless ones get approved together.
  • Point at the problem. Write a note on an image, and your agent gets your words, the exact box, and a picture of what you saw.
  • Approvals follow Git. An approval counts wherever history reaches it, so a commit, a rebase, or a merge never asks again.

It's early. Follow along.

Selfie is piloting with StorybookJS. The local loop, CI, and commit statuses are built, and in use on Selfie's own repository. It isn't on npm yet, so setup starts from a checkout: you'll need Node 26 or later, pnpm, and Docker. GitHub sign-in and private repositories aren't built yet.

See Selfie reviewing itself, no sign-in needed.

Or read the quickstart