Fraud Detection:Device Fraud Detection SDK for HarmonyOS

Prerequisites

Before you begin, make sure you have:

• HarmonyOS Next 0.0.71 or later, with a minimum API version of 12

• A physical device for testing — simulator debugging is not supported

• Bytecode packaging enabled — only bytecode packaging is supported

• The latest SDK version downloaded from the Alibaba Cloud official website, to meet privacy compliance requirements

Before using Device Fraud Detection, read the regulations on personal information processing and the Privacy Policy of Fraud Detection SDK.

Permissions

The SDK requires the following permissions. Grant all recommended permissions to maximize detection accuracy.

Install the SDK

1. Download the HarmonyOS SDK and extract the archive. The SDK is packaged as a standard HarmonyOS .har file.

2. Copy the .har file to your project's libs directory. In the oh-package.json5 file at your project root, add the version dependency for the authentication package:

   

3. In the oh-package.json5 file for your project module, add aliyundevice to the dependencies section:

   {
     "dependencies": {
       "aliyundevice": "file:../libs/HarmonyOS-AliyunDevice-xxx.har"
     }
   }

Obfuscation configuration

The .har package includes an obfuscation.txt file that prevents SDK APIs from being obfuscated. Do not remove this file — removing it causes functional failures.

API reference

The SDK provides two core methods: initWithOptions for initialization and getDeviceToken for token retrieval.

export class SecurityInitListener {
  // code: status code of the initialization result
  onInitFinish(code: number): void {}
}

public initWithOptions(ctx: Context,
                userAppKey: string,
                options: Map<string, string>,
                securityInitListener: SecurityInitListener): void;

export class SecurityToken {
  /**
   * Status code — see SecurityCode for meanings
   */
  public code: number = 0;

  /**
   * Device token returned by the SDK
   */
  public token: string = "";
}

public getDeviceToken(): SecurityToken

Status codes

Sample code

The following example shows a complete integration. ALIYUN_APPKEY is the app identifier — get it from Device App Management in the Alibaba Cloud console.

All calls use SecurityDevice.getInstance() as the single entry point.

import { SecurityCode, SecurityToken, SecurityDevice } from 'aliyundevice';

@Entry
@Component
struct Index {
  @State message: string = 'Aliyun Device';
  @State ALIYUN_APPKEY: string = "<your-appkey>";

  build() {
    Row() {
      Column() {
        Button(this.message)
          .fontSize(18)
          .fontWeight(FontWeight.Bold)
          .onClick((event: ClickEvent) => {
            // Initialize the SDK on app start, once per lifecycle
            SecurityDevice.getInstance().initWithOptions(getContext(), this.ALIYUN_APPKEY, null, null);

            // Wait at least 2 seconds before retrieving the token
            setTimeout(() => {
              let tokenObj: SecurityToken = SecurityDevice.getInstance().getDeviceToken();
              if (tokenObj.code == SecurityCode.SC_SUCCESS) {
                console.log("Aliyun Token: " + tokenObj.token);
              } else {
                console.log("Aliyun Code: " + tokenObj.code);
              }
            }, 2000);
          })
          .margin({ top: 10 })
      }
      .width('100%')
    }
    .height('100%')
  }
}

Replace <your-appkey> with your actual app key from the console.

In business scenarios that require fraud detection — such as account registration, login, and promotional activities — retrieve the device token and submit it to your server before calling the Fraud Detection API.

What's next

Use the device token to call the Fraud Detection API. The following references describe the request parameters and response fields for each scenario:

• Device risk detection events and returned parameters

• Registration fraud detection features and parameters

• Marketing fraud detection features and parameters

• Login fraud detection features and parameters

• Traffic promotion fraud prevention features and parameters