Android

Drop-in identity verification for Android: Kotlin, coroutines and CameraX capture screens. Your backend creates the session; the SDK runs the document and liveness steps.

package com.blink.kyc:blink-kyc-android:1.4.0 minSdk 24 host ComponentActivity

Prepare the environment

Add the Blink Maven repository and the dependency:

settings.gradle.ktskotlin

dependencyResolutionManagement {
    repositories {
        google()
        mavenCentral()
        maven("https://blink-pay.net/maven/releases")   // Blink SDK repository
    }
}

app/build.gradle.ktskotlin

android {
    defaultConfig { minSdk = 24 }
}

dependencies {
    implementation("com.blink.kyc:blink-kyc-android:1.4.0")
}

The SDK declares the camera permission and its capture screens; they are merged into your manifest and the drop-in screens request the runtime permission themselves.

Verification flow

Get a session token from your backend (Session API), then run the capture from a coroutine:

VerifyActivity.ktkotlin

import com.blink.kyc.BlinkError
import com.blink.kyc.BlinkKyc
import com.blink.kyc.BlinkStepError
import com.blink.kyc.DocumentType

lifecycleScope.launch {
    try {
        val outcome = BlinkKyc(baseUrl = "https://kyc-api.blink-pay.net", sessionToken = sessionToken)
            .document(DocumentType.PASSPORT)   // PASSPORT | NATIONAL_ID | ID_CARD | DRIVING_LICENCE
            .face()                            // liveness + face match
            .present(this@VerifyActivity)      // the SDK owns the camera UI
            .onProgress { p -> Log.d("kyc", p.step) }
            .run()

        // outcome.result: VERIFIED | REJECTED | REVIEW — a device copy.
        // Ask your backend for the authoritative result before acting.
    } catch (e: BlinkStepError) {
        Log.w("kyc", "${e.step}: ${e.code}")        // e.g. DOCUMENT_UNREADABLE
    } catch (e: BlinkError) {
        Log.e("kyc", "${e.code} (${e.httpStatus})") // switch on e.code
    }
}

Call	Effect
`BlinkKyc(baseUrl, sessionToken, timeoutMs, theme, strings)`	Create the client. Only the first two are required.
`.document(type, side)`	Enable the document step. Both are optional.
`.face()`	Enable liveness and face match.
`.present(activity)`	Use the built-in capture screens.
`.capture(hooks)`	Supply the media yourself (headless).
`.onProgress { }`	Progress events.
`.status()`	Suspend: read-only session progress from Blink.
`.run()`	Suspend: run the journey, returns `SessionResult(result, detail)`.

If you enable neither step, run() performs the full capture. The session's purpose decides which steps Blink accepts.

Headless capture

Bring your own camera: return the document image and the liveness frames, and the SDK handles challenges, uploads and the verdict.

HeadlessCapture.ktkotlin

BlinkKyc(baseUrl, sessionToken)
    .document(DocumentType.NATIONAL_ID)
    .face()
    .capture(object : CaptureHooks {
        override suspend fun document(): ByteArray = grabDocumentJpeg()
        override suspend fun liveness(actions: List<String>): List<ByteArray> = recordFrames(actions)
    })
    .run()

UI customisation