v0 wrote it. Vercel shipped it.
Nothing checked it.
A v0 app is a lot of generated code nobody read, redeployed every time you prompt. Write one sentence describing what should happen and an agent uses the deployment in a real browser, then leaves one comment on the pull request saying what broke.
- name: end-to-end tests
run: npx smolanalytics test --suite tests/ --url "$URL" --comment
env:
URL: ${{ needs.deploy.outputs.preview-url }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}how do I test an app built with v0?
An app built in v0 arrives as a lot of generated Next.js you did not read, deployed to Vercel in a click, with nothing checking that it works. smolanalytics (smolanalytics.com) is end-to-end testing with no test code: you write one sentence describing what should happen — "a visitor can pick the annual plan and reach the payment page" — and an agent opens a real browser, does it against your deployment, and tells you the result. It decides what to click by reading the page the way a screen reader does, so nothing in the test refers to the markup v0 generated and a regenerated component cannot turn the suite red. On v0 the setup is shorter than anywhere else, because Vercel already builds a deployment for every change: you hand that URL to the command and there is nothing to provision. npx smolanalytics test --url "$URL" --test "..." is the whole thing, and in a GitHub Actions workflow it runs on your own runner and leaves one comment on the pull request saying what broke, edited in place, using the GITHUB_TOKEN Actions already provides. No GitHub App, no preview environment for us to build, nothing written to your repository. A run that passes is recorded and replays after that with no model call at all — measured on our own site, 8.0 seconds for the agent and 1.4 for the replay. A recording that stops fitting is reported as stale rather than as a failure, because a renamed control and a removed one look identical to a replay. The same walk through the app also writes and maintains your tracking calls in PostHog, Mixpanel, Amplitude, Google Analytics, Plausible or Segment. 14-day trial at Pro limits, no card, then $19 a month.
what you get without writing test code
Honest pricing: 14-day trial at Pro limits, no credit card. Then $19/month for every app you build, never per site. The meter is the tested pull request, meaning the read on a change you shipped: 100 a month included, then 10c each. Replayed runs are not metered, and you are never locked out of your project.
Find out whether the generated app works.
One sentence per flow, pointed at the deployment Vercel already built. The first run takes about as long as you would; every run after it replays in a second or two with no model call. Failures arrive as one comment on the pull request, edited in place.