Local-first execution
Run your full suite on your machine in seconds. No cloud queues, no per-seat runners.
Home
Open source visual test automation
Build browser tests visually. Ship them as code.
Design test flows in a visual workspace, then export clean Playwright and Gherkin your team can run locally, review in git, and trust in CI.
How it works
From visual design to reviewable code in four steps.
A visual IDE built for the modern testing loop. No black boxes, no lock-in.
Step 01
Build flows by defining steps
Create test cases visually by chaining steps together. Define user actions, assertions, and navigation in an intuitive flow builder.
Step 02
Generate locators from the builder
If locators are required, generate them directly from the visual builder on your live application—no hand-written selectors required.
Step 03
Save and generate artifacts
Save your flow to automatically generate standard, reviewable artifacts like Gherkin feature files, Playwright step definitions, and locator JSON.
Step 04
Run tests and view reports
Execute tests locally with zero cloud latency and view rich, interactive reports. Everything is stored as plain text, making it easy to version control.
Review in PRs
The output looks like code your team already writes.
Every visual flow becomes Gherkin features, Playwright steps, and locator JSON in your repo—ready for normal code review, not a 5 MB proprietary diff.
feat: add checkout visual smoke test #42
Open
tests/appraise/features/checkout-smoke.feature Gherkin Diff
@@ -1,3 +1,11 @@ Feature: Checkout smoke - # TODO: Write manual test steps+ Scenario: Returning buyer completes a card checkout+ When I add "Everyday Hoodie" to the cart+ And I use the saved shipping address+ And I pay with "visa_success"+ Then the order confirmation should be visible+ And the confirmation total should equal "$64.00" tests/appraise/steps/checkout.steps.ts TypeScript Step Definition Diff
@@ -0,0 +1,15 @@+import { expect } from "@playwright/test";+import locators from "../locators/checkout.locators.json";++When("I add {string} to the cart", async ({ page }, itemName) => {+ await page.getByRole("link", { name: itemName }).click();+ await page.locator(locators.product.addToCart.selector).click();+ await expect(page.locator(locators.cart.summary.selector)).toBeVisible();+});++When("I pay with {string}", async ({ page }, cardName) => {+ await page.locator(locators.payment.savedCard.selector).selectOption(cardName);+ await page.locator(locators.payment.submit.selector).click();+}); tests/appraise/locators/checkout.locators.json JSON Locator Diff
@@ -4,4 +4,7 @@ "cart": { "summary": "[data-testid='cart-summary']" },+ "product": {+ "addToCart": "[data-testid='add-to-cart']"+ },+ "payment": {+ "savedCard": "[data-testid='saved-card']",+ "submit": "[data-testid='submit-payment']"+ }Built for engineering teams
Everything you need to adopt visual testing without giving up code control.
Repo-owned artifacts
Features, steps, and locators live in git. Review them in PRs like any other code.
Extensible step library
Add custom TypeScript steps in seconds and wire them into visual flows.
Try it in 15 minutes
Install locally and prove the workflow yourself.
npx create-appraisejs@latest AppraiseJS is fully open source. Your tests, locators, and reports are plain files in your repository—inspect them locally, review them in PRs, and run them in any CI pipeline.
Quick start npx create-appraisejs@latest
Readable scenario structure—QA can follow this without opening the visual tool. The steps map cleanly to a smoke test I'd sign off on.
Approved.
Standard Playwright step definitions importing locators from JSON. I can debug these locally and extend them when we need custom logic.
Approved.
Selectors live in version-controlled JSON, not a hosted database. When a `data-testid` changes, it's one file to grep and patch.
Approved.