Browser automation for Kotlin Multiplatform. Write a workflow once and run it inside Android WebView, iOS WKWebView or JVM Chromium. Use a Playwright-inspired Kotlin API, or give an AI agent the same browser through compact commands.

Website · Documentation · Examples

[!NOTE] 0.0.1 alpha. Available on Maven Central. APIs and platform coverage may change.

Add to commonMain.dependencies in your module’s build.gradle.kts:

implementation("io.github.bernaferrari:kmpwright-core:0.0.1")

One dependency covers Android, iOS and JVM. Gradle selects the platform variant. Requires mavenCentral() in your project's repositories.

repositories { mavenCentral() }

kotlin {
    sourceSets {
        commonMain.dependencies {
            implementation("io.github.bernaferrari:kmpwright-core:0.0.1")
        }
    }
}

implementation("io.github.bernaferrari:kmpwright-compose:0.0.1")

Fill a form, click its button and verify the rendered result. This demo uses a self-contained page, so it runs without a server or an account.

import com.bernaferrari.kmpwright.core.*

suspend fun demo(page: Page) = with(page) {
    setContent(demoHtml)
    getByLabel("Name").fill("Ada")
    getByRole("button", name = "Say hello").click()
    expect(getByRole("status")).toHaveText("Hello, Ada!")
}

Result: the page displays Hello, Ada! and the assertion passes. Locators wait for actionable elements; assertions wait for the expected result. The same workflow runs on Android, iOS and JVM.

val demoHtml = """
    <!doctype html>
    <html lang="en">
    <meta charset="utf-8">
    <title>KMPwright demo</title>
    <label for="name">Name</label>
    <input id="name">
    <button onclick="document.getElementById('greeting').textContent =
        'Hello, ' + document.getElementById('name').value + '!'">Say hello</button>
    <p id="greeting" role="status"></p>
    </html>
""".trimIndent()

Connect your platform's browser, then pass its Page to the shared workflow. Run suspending APIs in a coroutine, on the main thread for mobile WebViews.

Browser(AndroidPlatformContext(activity, webView)).use { browser ->
    demo(browser.page())
}

Browser(IOSPlatformContext(webView = webView)).use { browser ->
    demo(browser.page())
}

launchBrowser(executablePath).use { launched ->
    demo(launched.browser.newPage())
}

page.runCommand("snapshot -i")
// @e1 [input type="email"]
// @e2 [button] "Sign in"

page.runCommand("fill @e1 ada@example.com")
page.runCommand("click @e2")

Expose runCommand() as a tool to your LLM SDK. The model reads snapshots and acts on element references, using the documented agent-browser command subset.

Features follow each backend’s abilities. Inspect page.capabilities() before using an optional feature; the capability matrix explains partial and unsupported behavior.

KMPwright adapts browser behavior from agent-browser into Kotlin APIs. It is independent of Playwright. The upstream comparison records the pinned revision and coverage; upstream source is not included in this repository.

For contributions, see development and tests. For artifact verification, see the release guide.

MIT · Third-party notices