Skip to main content

​
Installation

You will receive an API key and a Client ID from SmartComply. Keep these secure and do not commit them to source control.
Add the SDK to your app’s pubspec.yaml, replacing YOUR_TOKEN with the token provided to you: 
dependencies:
  smartcomply_sdk:
    git:
      url: https://YOUR_TOKEN@github.com/Anitajallas/smartcomply-web-sdk.git
      path: flutter
Then run: 
flutter pub get
Store the git token in an environment variable or a secrets manager — never hard-code it in pubspec.yaml.

​
How It Works

1

Create Session

Your app calls createSession() using your apiKey and clientId. This returns a session token.
POST /v1/session/create
2

Verify Identity

Your app calls onboarding.verify() using the session token to confirm the user’s identity.
POST /v1/onboarding/verify
3

Liveness Check

Your app calls liveness.startCheck():
  1. On-device AI: Camera opens and ML Kit detects face actions locally (no network).
  2. Autoshot capture: An autoshot is captured and sent to receive an Entry ID. 
    POST /v1/liveness/create
  3. Video upload: The video is uploaded for final analysis. 
    POST /v1/liveness/submit
Face detection runs entirely on-device — no video is streamed live to the server.

​
Platform Setup

In android/app/src/main/AndroidManifest.xml, add inside <manifest>:
<uses-permission android:name="android.permission.CAMERA" />
<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.RECORD_AUDIO" />
In android/app/build.gradle (or build.gradle.kts):
android {
    defaultConfig {
        minSdk 21
    }
}

​
Initialisation

import 'package:smartcomply_sdk/smartcomply_sdk.dart';

final sdk = SmartComply(
  SDKConfig(
    apiKey: 'YOUR_API_KEY',        // Bearer token from SmartComply
    clientId: 'YOUR_CLIENT_UUID',  // UUID from your SmartComply dashboard
    environment: Environment.production,
  ),
);
ParameterRequiredDescription
apiKey✅API key issued by SmartComply. Used to authenticate session creation.
clientId✅UUID of your SDK config record. Ties the session to your account settings.
environmentOptionalControls which SmartComply backend the SDK connects to:
• Environment.production - Live system. Verifications use real databases and charge your account.
• Environment.sandbox - Testing system. Uses mock data and does not charge your account.
Defaults to sandbox.
timeoutOptionalHTTP timeout. Defaults to 15 seconds.

​
Step 1 — Create a Session

final session = await sdk.createSession();
print(session.token); // UUID session token
Must be called before any other SDK method. All subsequent calls automatically use the session token.

​
Step 2 — Verify Identity

// Verify BVN
final result = await sdk.onboarding.verify(
  onboardingType: OnboardingType.bvn,
  idNumber: '22476562817',
);

// Verify NIN
final result = await sdk.onboarding.verify(
  onboardingType: OnboardingType.nin,
  idNumber: '91014204738',
);

if (result.isVerified) {
  print('Verified at: ${result.verifiedAt}');
}
OnboardingTypeID
OnboardingType.bvnBank Verification Number
OnboardingType.ninNational Identification Number

​
Step 3 — Liveness Check

Opens the camera full-screen, guides the user through actions, captures a photo and video, and submits both to the backend: 
final result = await sdk.liveness.startCheck(
  context,
  identifier: '22476562817',     // user's ID number
  identifierType: 'bvn',         // 'bvn' or 'nin'
  country: 'nigeria',
  challengeActions: [
    ChallengeAction.blink,
    ChallengeAction.turnLeft,
  ],
);

print('Status: ${result.status}');           // "success"
print('Liveness: ${result.livenessStatus}'); // "processing" → "passed" | "failed"

​
Supported Challenge Actions

ChallengeActionAPI value sentDetection method
ChallengeAction.blinkBLINKML Kit eye open probability
ChallengeAction.turnLeftTURN_LEFTML Kit head euler Y angle
ChallengeAction.turnRightTURN_RIGHTML Kit head euler Y angle
ChallengeAction.turnHeadTURN_HEADML Kit head euler Y angle (either direction)
ChallengeAction.openMouthOPEN_MOUTHHeuristic (confirmed by backend video AI)
After liveness/submit, the status will be "processing" while the backend AI analyses the video. The final passed or failed status is delivered via webhook to your webhook_url.

​
Error Handling

try {
  await sdk.createSession();
} on SDKError catch (e) {
  print('Error ${e.statusCode}: ${e.message}');
}

​
Full Example

import 'package:flutter/material.dart';
import 'package:smartcomply_sdk/smartcomply_sdk.dart';

final sdk = SmartComply(
  SDKConfig(
    apiKey: 'YOUR_API_KEY',
    clientId: 'YOUR_CLIENT_UUID',
    environment: Environment.production,
  ),
);

Future<void> runVerification(BuildContext context) async {
  try {
    // 1. Session
    await sdk.createSession();

    // 2. Identity
    final identity = await sdk.onboarding.verify(
      onboardingType: OnboardingType.bvn,
      idNumber: '22476562817',
    );
    if (!identity.isVerified) return;

    // 3. Liveness
    final liveness = await sdk.liveness.startCheck(
      context,
      identifier: '22476562817',
      identifierType: 'bvn',
      country: 'nigeria',
      challengeActions: [ChallengeAction.blink, ChallengeAction.turnLeft],
    );

    print('Liveness submitted: ${liveness.status}');
    // Final result delivered via webhook
  } on SDKError catch (e) {
    debugPrint('SmartComply error: $e');
  }
}