The features
that ship for you.
App testing, beta testing, visual regression, accessibility, load testing — every test surface a modern product team needs, packaged as one desktop app.
The lineup
Eight test surfaces.One desktop app.
Run loop
Plan. Drive. Observe. Learn. The four phases of every Viziera run.
From the PR webhook to the status check, every run follows the same deterministic loop — and produces the same artifacts.
Plan
Viziera scans your repo for changed files, maps them to scenarios, and prioritizes the ones likely to surface a regression first.
Drive
The desktop driver opens a real browser, handles auth (including 2FA), and executes scenarios with self-healing selectors and retries.
Observe
Screens, network, console, and AI signals are captured. Visual regression diffs against baseline; the AI tester flags anomalies.
Learn
After every run, Viziera updates its baselines, flakiness scores, and per-finding history. The signal-to-noise ratio improves every cycle.
Selectors that heal. Not selectors that flake.
When a CSS selector breaks, Viziera falls back to text, aria-label, semantic role, and visual position. Most refactors that used to nuke a suite now just keep running.
An AI user that finds the bugs you missed.
Logs into your app, crawls every reachable page, and prioritizes exploration by likely bug density. Each finding comes back with a screenshot, repro script, and severity.
Wired to your repo. Not bolted on after the fact.
A bearer-token webhook ties Viziera to GitHub, GitLab, or Bitbucket. PRs get a status check, a comment with screenshots, and a delta of new findings vs. the base branch.
Workflows around tests. Not just tests on a schedule.
If/else, foreach, parallel branches, wait-for-event, and per-step error handlers. The control flow you want in production code — applied to your test runs.
How we're different
Not a framework. Not a cloud silo.
How Viziera compares to roll-your-own Playwright suites and the SaaS test platforms in your spam folder.
Playwright / Cypress
- Selectors break on every refactor
- No autonomous exploration
- Headless — misses native dialogs
- CI integration is your problem
Cloud test platforms
- Your credentials live in their cloud
- Screenshots stored off-machine
- Per-minute pricing on every run
- Limited native / desktop support
Viziera
- Self-healing selectors out of the box
- Autonomous AI tester included
- Real desktop control — not just browser
- Local-first runs, keytar-backed secrets
- Token-auth webhook CI in under an hour
Scenarios as code. Reviewed like any other PR.
Author scenarios in TypeScript and the runner auto-discovers them — or record one with the Chrome extension and let Viziera generate the file for you. Either way, they sit in your repo and review like any other code.
// Drop into ./viziera/scenarios — the runner discovers it
export default scenario({
name: 'Login — happy path',
url: 'https://app.example.com',
viewports: ['desktop', 'mobile'],
run: async (page) => {
await page.type('#email', 'qa@viziera.dev')
await page.click('Sign in')
await page.expect('Welcome back')
}
})Ready to test like a user?
Start with one scenario. Scale to a full release pipeline.