Developers · iOS Component SDK Developer
SDK

iOS Component SDK

The Assentify Component SDK for iOS gives developers direct, native control of individual identity components such as face capture, ID document capture, and data extraction, then lets them submit the collected data to the Digital Experience Platform (DXP).

Install · Initialize · Capture
Platform iOS native (Swift) SDK type Component SDK Also available Android, Flutter, React Native Delegate AssentifySdkDelegate
Which SDK is this? Assentify offers two SDK types. The Flow SDK embeds a whole published DXP flow, a complete guided journey, natively and is driven by a generated flow URL. The Component SDK, which this page documents, gives developers direct control of individual identity components (for example face capture, ID document capture, and extraction) and lets them submit the collected data to the Digital Experience Platform (DXP). For the Flow SDK, see the SDK overview.

Overview

The Assentify SDK has several capabilities and functionalities that empower developers to take complete control of the identity extraction process. The Assentify SDK complements the Assentify DXP platform (SaaS). The developer leverages the available data by using the response information from the supported components or data provided by the app user. The SDK allows the developer to submit all extracted data to Assentify to gain access to further features such as Compliance, AML, client profile, and other features that the DXP covers.

The key components that are supported and available in the SDK are:

Identification Document Capture

A standalone, event-driven component that extracts information from an ID card. The ID type is determined by the configuration set in Assentify DXP.

Face Capture / Matching

A standalone, event-driven component that performs face matching between a live source and a developer-supplied target image.

Context Aware Signing

Retrieves document tokens, creates a user document instance, and applies a base64 signature to produce a signed document URL.

Definitions

TermMeaning
StepIDThe StepID represents the ID of the Step in the Engagement block that you configure in the Flow Builder on the portal (for example, the StepID of Face Capture or ID Capture). You will need it to submit your data to the portal.

Requirements

To initialize the Component SDK you require the following values, all of which can be retrieved from the Assentify DXP portal. Once you have logged in, visit the Developers section and select the flow you would like to use for your application.

ValueWhere to retrieve it
API KeyIn the Developers section, create a new API Key or use an existing key. You can assign Claims to an API Key to limit the amount of access a specific API Key has.
Tenant IdentifierIn Developers › Flow Configuration, select the flow and Engagement block. The Tenant Identifier is shown in the flow information section for that block.
Interaction HashIn Flow Manager, select your flow. When the Digital Engagement Link loads on the right panel it has the format https://platform.assentify.com?interaction={HASH}. Copy the HASH part.
AvailabilityThe Assentify Component SDK is available for Android native, iOS native, Flutter, and React Native.

Initialization

To leverage the capabilities of the Assentify Component SDK, the developer must initialize the SDK on each application start-up. Initialization ensures the developer has the key information required to begin using the Component SDK.

Step 1 : environmentalConditions configuration

The environmentalConditions variable is instantiated as an object of type EnvironmentalConditions. It is configured with the following parameters.

ParameterDescription
enableDetectEnables the detection feature.
enableGuideEnables the guide feature to assist users during scanning.
CountDownNumbersColorCustom color (hex) for the countdown timer on the Face Match step, for example #610A3A.
HoldHandColorColor (hex) for the hand-holding guide or indicator, for example #FFC400.
MotionLimitNumber of frames taken before sending in case of a steady hand, for ID Cards on iOS. (Optional)
MotionLimitFaceNumber of frames taken before sending in case of a steady hand, for Face on iOS. (Optional)
BRIGHTNESS_HIGH_THRESHOLDAcceptable high brightness threshold, Android. (Optional)
BRIGHTNESS_LOW_THRESHOLDAcceptable low brightness threshold, Android. (Optional)
activeLiveTypeLocal active liveness type: actions (move your head left, right, up, down), wink (your eye, left or right), or blink. (Optional)
activeLivenessCheckCountThe number of times the local active liveness check will be requested. (Optional)
faceLivenessRetryCountThe number of times the user failed the face liveness before passing the step. (Optional)
minRam, minCPUCoresAny Android device with RAM or CPU smaller than these values will use manual scanning instead of automatic detection. (Optional)

Step 2 : assentifySdk configuration

The assentifySdk object is an instance of AssentifySdk and is initialized with several parameters.

ParameterDescription
apiKeyThe API Key retrieved from the DXP portal.
tenantIdentifierThe Tenant Identifier retrieved from the DXP portal.
interactionThe interaction (Interaction Hash) value.
environmentalConditionsPasses the previously defined environmentalConditions as part of the SDK configuration.
assentifySdkDelegateSets the current class (self) as the callback handler for SDK events.
performActiveLivenessFaceEnables the local active liveness check for face. (Optional)

Initialization delegates

DelegateDescription
onAssentifySdkInitSuccessAn initialization callback fired when the SDK has successfully been initialized.
onAssentifySdkInitErrorA callback executed upon failure to initialize the SDK.
Load templates after initUpon completion of the initialization, call the getTemplates function. This returns a List<TemplatesByCountry>. Store the result to be used later in the process when processing an ID card.

Retrieving the values from the DXP portal

  1. Visit the Assentify DXP and click on the Developers section. Then click on the Flow Configuration button.
  2. Select the flow you want to use with your application. Then from the drop-down menu, select the Engagement block you would like to target.
  3. Once you have selected an appropriate block you will see the information related to the flow. Retrieve the Tenant Identifier from this section.
  4. To retrieve the API Key, create a new API Key or use an existing key. You can assign Claims to an API Key to limit the amount of access it has.
  5. To fetch the Interaction Hash, click on the Flow Manager and select your flow. When the Digital Engagement Link loads on the right panel, its format is https://platform.assentify.com?interaction={HASH}. Copy the HASH part to retrieve the interaction hash.

BlockLoader

The Assentify BlockLoader is a core component provided to the developer to perform several functions and capabilities. The BlockLoader gives the developer core properties that must be set to give context to the Assentify DXP platform.

Built-in property
DeviceName
FlowName
TimeStarted
Application
UserAgent
InstanceHash

The properties above are built into the BlockLoader. However, the developer can access the DXP and add additional properties if needed, such as phoneNumber and clientIdentifier. These properties can transfer custom information to Assentify DXP, which can later be mapped to WebHooks or other platform capabilities. The developer has no limit on the number of properties that can be added to the BlockLoader.

How to add a property to the BlockLoader

  1. Go to the Flow Manager and select the flow you want to edit.
  2. Click the view more button on the Engagement Block.
  3. In the input field, type the name of the property. Then press the add button.
  4. Completed.

How to use the BlockLoader properties as a developer

Once the developer adds these properties using the DXP, the developer then defines the added properties inside their application, which they can use to set the BlockLoader step for submission. All custom properties added can be used to perform additional capabilities on the Assentify DXP, such as running decisions and mapping webhooks. To make use of the custom properties, declare them inside your app. Assentify uses OnBoardMe_Property_{propertyName} as the naming nomenclature for all custom properties.

Identification Document Capture

The Identification Document Capture is a standalone, event-driven component that provides a means by which the developer can extract information from an ID card. The ID type is determined by the configuration set in Assentify DXP. The developer or business user can log in to Assentify and manage these configurations anytime, even after the application has gone live. This provides flexibility to support as many ID cards as possible without recompiling and deploying the application.

This component returns many different types of information depending on the ID type being scanned, such as a Civil ID versus a Driving License. This is done using dynamic keys and mappings built into the SDK. A global model is available for developers to extract information from the base model and the available property dictionary.

The component has a built-in user interface that can be used by default, or the developer can use their own custom user interface by turning off the guides and default interface. The component leverages the native tools while providing complete control of what is displayed on the user interface using the events fired during the scanning process.

Methods

MethodDescription
startFlowInitializes and launches the entire flow, including the UI and events configured in the Admin Portal.
getTemplatesReceives all the configured supported Identification Document templates in the system (such as Passport, Lebanese Civil ID, Lebanese Driving License, French Resident Permit). You need the template when starting the ID scanning feature to inform the scanner which template you are using, and to create a list for the user to select which ID they would like to provide.
startScanIDCardBegins the ID scanning process.
startScanPassportBegins the passport scanning process.
startScanOtherBegins the scanning process for other ID types.
readPassportBegins NFC passport scanning.

startFlow parameters

The FlowEnvironmentalConditions parameter carries the following fields.

FieldDescription
logoUrlURL of the logo displayed in the flow header or main screen. (Optional)
svgBackgroundImageUrlURL of the SVG background image used for the screen background. (Optional)
textColorPrimary text color in HEX format, for example #000000. (Optional)
secondaryTextColorSecondary text color in HEX format, used for subtitles or less prominent text. (Optional)
backgroundCardColorBackground color for cards and containers in HEX format. (Optional)
accentColorAccent color used for highlights and interactive elements. (Optional)
backgroundColorDefines the main background style of the screen. Can be Solid (single color) or Gradient (multiple colors with an angle). (Optional)
clickColorDefines the background style applied on click or pressed states. (Optional)
backgroundTypeDefines how the background should be rendered, for example color or image. This is the only required field.
LanguageOptionally set the language that the OCR will process the extracted content from the cards. (Optional)
enableNfc, enableQrTriggers NFC scanning after scanning a passport, or QR scanning after scanning a card.
blockLoaderCustomPropertiesMap containing the custom properties configured in the Admin Portal.
flowDelegateSpecifies the callback function that will be called during flow completion.
fromCurrent view controller.

startScanIDCard parameters

ParameterDescription
scanIDCardDelegateSpecifies the delegate function that will be called during the scanning process.
templatesByCountryThe object of the templates that are supported to scan.
LanguageOptionally set the language that the OCR will process the extracted content from the ID card. (Optional)

startScanPassport and startScanOther parameters

ParameterDescription
scanPassportDelegate / scanOtherDelegateSpecifies the delegate function that will be called during the scanning process.
LanguageOptionally set the language that the OCR will process the extracted content from the ID card. (Optional)

How to initialize the flow

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.startFlow().
  3. Configure the startFlow parameters.

How to initialize ID card scanning

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.startScanIDCard().
  3. Configure the startScanIDCard parameters.
  4. The startScanIDCard method accepts a templatesByCountry parameter that includes all the available retrieved templates.

How to initialize passport scanning

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.startScanPassport().
  3. Configure the startScanPassport parameters.
  4. Completed.

How to initialize NFC passport scanning

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.readPassport().
  3. Check NFC support.
  4. Enable NFC Tag Reading from Xcode.

How to initialize Other ID scanning

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.startScanOther().
  3. Configure the startScanOther parameters.
  4. Completed.

Face Capture / Matching

The Face Capture and Matching step is a standalone, event-driven component that allows the developer to perform face matching between a live source and a target image. The developer offers the target image, while the selfie camera captures the source image. All the features and capabilities of the component are driven by the configuration that the developer or business user can set on Assentify DXP. The developer may choose to override this configuration.

How to use the Face Acquisition component

  1. Initialize the AssentifySdk object.
  2. Call AssentifySdk.startFaceMatch().
  3. Configure the startFaceMatch parameters.
  4. Supply the target base-64 encoded face image.
  5. Add the component to the ViewController as a subview.

Context Aware Signing

How to initialize it

  1. Call startContextAwareSigning.
  2. Wait for onHasTokens: the tokens from your selected document.
  3. Fill the tokens and call createUserDocumentInstance.
  4. Wait for onCreateUserDocumentInstance and call the signature function with the base64 signature.
  5. Wait for onSignature and get the document URL.

Wrap Up

The Wrap Up component is a built-in step that allows the developer to set the TimeEnded value. The developer can supply an end time for the flow.

Submission

Submission is a built-in function of the SDK that allows the developer to initiate the data transmission from the integrating app to Assentify DXP. Submission can be used to add additional properties to the BlockLoader and perform data aggregation and manipulation before sending it to Assentify. For example, the developer can send additional eKYC information, user-specific data, or data to bridge between the Assentify DXP and the back-end services of the integrating application. The developer may also perform additional functions such as custom date formatting, translation, transliteration, and calculations.

One of the key features of the Assentify Component SDK is being able to submit all the collected data to Assentify DXP, allowing business users to review all the collected data. The simplest way to submit data to Assentify is to leverage the existing code provided in the Demo Application (Android, iOS).

How to submit data to Assentify DXP

  1. Retrieve the step definition model so that there is a reference to the List<StepDefinitions>.
  2. Retrieve the output properties model.
  3. Retrieve the output properties for the face from OutPropertiesModelFace.
  4. Define variables for the wrap up, blockLoader, ContextAwareSigning step, face step, and ID capture step.
  5. Loop through the steps and populate the output properties into a map.
  6. Ensure that each SubmitRequestModel is then set into a MutableList<SubmitRequestModel>.
  7. Call the assentifySdk.startSubmitData() method.
Preparing the submit listYou must prepare your list of steps with the outputProperties obtained for each step, for example:
sharedStepDefinitionsStepFace = SubmitRequestModel(item.stepId, StepsName.FaceMatch, values);
sharedStepDefinitionsStepDoc  = SubmitRequestModel(item.stepId, StepsName.DocumentCapture, values);

Data Models

BaseResponseDataModel

FieldDescription
outputPropertiesA map containing the extracted data with hash keys for submission.
transformedPropertiesA map with extracted data that may be transformed if translation is needed.
additionalDetailsExtra extracted data, particularly for "Other Scanning."
imageUrlThe URL of the ID image.
facesURLs of faces extracted from the ID.
identificationDocumentCaptureThis model contains all extracted data.
baseImageFaceThe selfie image used for "Face Match."
secondImageFaceThe image provided for comparison in "Face Match."
percentageMatchMatch percentage.
isLiveLiveliness detection result.
error messageThe error message, if any.
successA boolean indicating success.

TemplatesByCountry

FieldTypeDescription
nameStringThe name of the country.
sourceCountryCodeStringThe ISO-2 country code.
flagStringThe country flag image.
templatesList<Templates>The list of templates by country.

KycDocumentDetails

FieldTypeDescription
nameStringThe name of the document.
orderIntThe order of the page (front, back, etc.).
templateSpecimenStringThe default specimen that represents the ID type being used (Civil ID, Driving License, etc.).
templateProcessingKeyInformationStringThe template Id.
hasQrCodeBooleanWhether the ID has a QR code.

Templates

FieldTypeDescription
idIntThe template Id.
sourceCountryFlagStringThe flag of the country as a URL.
sourceCountryCodeStringThe ISO-2 country code.
kycDocumentTypeStringThe name of the template.
sourceCountryStringThe country name.
kycDocumentDetailsList<KycDocumentDetails>A list of KycDocumentDetails.

ConfigModel

FieldTypeDescription
stepIdIntIdentifier for the step.
flowNameStringName of the flow.
blockNameStringName of the block.
instanceHashStringHash value of the instance.
stepDefinitionAny?Step definition, can be any type.
flowInstanceIdStringIdentifier for the flow instance.
tenantIdentifierStringIdentifier for the tenant.
blockIdentifierStringIdentifier for the block.
flowIdentifierStringIdentifier for the flow.
customPropertiesMap<String, Any>Custom properties map with string keys and any values.
defaultLanguageIdIntIdentifier for the default language.
instanceIdStringIdentifier for the instance.
userStateStepMapMap<String, List<UserState>>Mapping of user states.
stepDefinitionsList<StepDefinitions>List of step definitions.

StepDefinitions

FieldTypeDescription
stepIdIntIdentifier for the step.
stepDefinitionStringDefinition of the step.
customizationCustomizationCustomization options.
outputPropertiesList<OutputProperties>List of output properties.

OutputProperties

FieldTypeDescription
idIntIdentifier for the output property.
keyStringKey for the output property.
displayNameStringDisplay name for the output property.
isRequiredBooleanWhether the property is required.
isExcludedBooleanWhether the property is excluded.
typeIntType identifier for the property.

Customization

FieldTypeDescription
ProcessMrzBoolean?Whether to process MRZ (Machine Readable Zone).
StoreCapturedDocumentBoolean?Whether to store captured documents.
PerformLivenessDetectionBoolean?Whether to perform liveness detection.
StoreImageStreamBoolean?Whether to store image streams.
SaveCapturedVideoBoolean?Whether to save captured video.
identificationDocumentsList<IdentificationDocuments>?List of identification documents.

PassportResponseModel

FieldTypeDescription
destinationEndpointString?Destination endpoint.
passportExtractedModelPassportExtractedModel?Extracted passport model.
errorString?Error message.
successBoolean?Success status.

PassportExtractedModel

FieldTypeDescription
outputPropertiesMap<String, Any>?Output properties map.
transformedPropertiesMap<String, String>?Transformed properties map.
extractedDataMap<String, Any>?Extracted data map.
imageUrlString?URL for the image.
facesList<String>?List of detected faces.
identificationDocumentCaptureIdentificationDocumentCapture?Identification document capture details.

OtherResponseModel

FieldTypeDescription
destinationEndpointString?Destination endpoint.
otherExtractedModelOtherExtractedModel?Extracted model for other data.
errorString?Error message.
successBoolean?Success status.

OtherExtractedModel

FieldTypeDescription
outputPropertiesMap<String, Any>?Output properties map.
transformedPropertiesMap<String, String>?Transformed properties map.
extractedDataMap<String, Any>?Extracted data map.
additionalDetailsMap<String, Any>?Additional details map.
transformedDetailsMap<String, String>?Transformed details map.
imageUrlString?URL for the image.
facesList<String>?List of detected faces.
identificationDocumentCaptureIdentificationDocumentCapture?Identification document capture details.

IDResponseModel

FieldTypeDescription
destinationEndpointString?Destination endpoint.
idExtractedModelIDExtractedModel?Extracted model for ID data.
errorString?Error message.
successBoolean?Success status.

IDExtractedModel

FieldTypeDescription
outputPropertiesMap<String, Any>?Output properties map.
transformedPropertiesMap<String, String>?Transformed properties map.
extractedDataMap<String, Any>?Extracted data map.
imageUrlString?URL for the image.
facesList<String>?List of detected faces.
identificationDocumentCaptureIdentificationDocumentCapture?Identification document capture details.

FaceResponseModel

FieldTypeDescription
destinationEndpointString?Destination endpoint.
faceExtractedModelFaceExtractedModel?Extracted model for face data.
errorString?Error message.
successBoolean?Success status.

FaceExtractedModel

FieldTypeDescription
outputPropertiesMap<String, Any>?Output properties map.
extractedDataMap<String, Any>?Extracted data map.
baseImageFaceString?Base image face.
secondImageFaceString?Second image face.
percentageMatchInt?Percentage match between faces.
isLiveBoolean?Whether the face is live.
identificationDocumentCaptureIdentificationDocumentCapture?Identification document capture details.

IdentificationDocumentCapture

FieldTypeDescription
nameAny?Name of the document holder.
surnameAny?Surname of the document holder.
documentTypeAny?Type of the document.
birthDateAny?Birthdate of the document holder.
documentNumberAny?Document number.
sexAny?Sex of the document holder.
expiryDateAny?Expiry date of the document.
countryAny?Country of the document.
nationalityAny?Nationality of the document holder.
idTypeAny?ID type of the document.
faceCaptureAny?Face capture details.
imageAny?Image of the document.
isSkippedAfterMRZFailsAny?Whether MRZ failed after skipping.
isFailedFrontAny?Whether the front check failed.
isFailedBackAny?Whether the back check failed.
skippedStatusAny?Skipped status.
capturedVideoFrontAny?Captured video of the front.
capturedVideoBackAny?Captured video of the back.
livenessStatusAny?Liveness status of the document holder.
isFrontAuthAny?Whether the front is authorized.
isBackAuthAny?Whether the back is authorized.
isExpiredAny?The document isExpired (true, false).
isTamperingAny?The document is tampered (true, false).
tamperHeatMapAny?The URL to the tampering heatmap.
isBackTamperingAny?The back of the document has been tampered.
backTamperHeatmapAny?The URL to the heatmap of the back-side of the document.
originalBackImageAny?The URL to the back-side of the image.
ghostImageAny?The URL to the ghost image.
idMaritalStatusAny?The marital status.
idDateOfIssuanceAny?The date of issuance.
idCivilRegisterNumberAny?The civil register number.
idPlaceOfResidenceAny?The place of residence.
idProvinceAny?The province.
idGovernorateAny?The governorate.
idMothersNameAny?The mother's name.
idFathersNameAny?The father's name.
idPlaceOfBirthAny?The place of birth.
idBackImageAny?The URL to the back image.
idBloodTypeAny?The blood type.
idDrivingCategoryAny?The driving category.
idIssuanceAuthorityAny?The issuance authority that issued the ID.
idArmyStatusAny?The army status.
idProfessionAny?The profession of the ID holder.
idUniqueNumberAny?The unique number on the ID card.
idDocumentTypeNumberAny?The ID document number.
idFeesAny?The fees paid.
idReferenceAny?The reference data associated with the ID card.
idRegionAny?The region.
idRegistrationLocationAny?The registration location.
idFaceColorAny?The face color.
idEyeColorAny?The eye color.
idSpecialMarksAny?Any special marks (tattoos, etc.).
idCountryOfStayAny?The current country of stay.
idIdentityNumberAny?The Identity Number of the ID.
idPresentAddresssAny?The card holder present address.
idPermanentAddresssAny?The card holder permanent address.
idFamilyNumberAny?The family number.
identityNumberBackAny?The identity number found on the back of the ID card.

DocumentTokensModel

FieldTypeDescription
displayNameStringToken name.
idintToken id.
templateIdintDocument id.
tokenTypeEnumintToken type.

CreateUserDocumentResponseModel

FieldTypeDescription
documentIdintDocument id.
isPdfboolIs HTML or PDF.
templateInstanceStringDocument as base64 after filling the tokens.
templateInstanceIdintThe new document id.

SignatureResponseModel

FieldTypeDescription
fileNameStringDocument name.
signedDocumentboolDocument as base64 after signed.
signedDocumentUriStringDocument as URL after signed.

Callback Events

The following callback events are fired against BaseResponseDataModel during processing.

EventDescription
onSend()Starts processing.
onRetry()Invoked when processing fails.
onComplete()Called when processing is complete.
onEnvironmentalConditionsChange()Provides real-time updates on environmental conditions like brightness and motion changes.
onWrongTemplate() (ID Card)Called when trying to scan a wrong ID based on the submitted template.
onCurrentLiveMoveChange() (Face)Provides real-time updates about what the current action required from the user is.
onStartNfcScan() (NFC)Called when NFC starts.
onCompleteNfcScan() (NFC)Called when NFC processing is complete.
onErrorNfcScan() (NFC)Invoked when processing fails.
onCollectingManualImages() (Face)In manual scan mode, after the user clicks "Take Photo", this function notifies the user that multiple images will be captured in order to verify liveness detection. After that, the onSend function is called.

Language Transformation

A list of LanguageTransformationModel objects is created and populated with different language transformation configurations. Each model specifies the type of language transformation (either translation or transliteration), a key-value pair, the target language, and the data type for that transformation.

ParameterDescription
languageTransformationEnumThe type of transformation (transliteration or translation).
Key and value stringsPlaceholders ("Your Key", "Your Value") for specific content you want to work with.
LanguageThe target language (for example English, Arabic, Azerbaijani).
DataTypeThe type of data being transformed (text, date, number).

FAQ

Can you set what data the ResponseModels have from the admin page?

No, the response models are static and are based on responses from our core ID orchestrator.

When the onComplete callback is triggered (for example for the passport), can that data be retrieved or a hook triggered on our backend?

When the callback is triggered, the application receives all the information. It is the responsibility of the application to transfer that data to any specific backend at this point. If the data can be transferred at a later stage or when it has been submitted to our platform, then yes, you can define a webhook in the connector block to be triggered after submission to inform your backend or send you specific data.

What are the KycDocumentDetails on the ScanIDCard? How would I know how many to add to the list?

The template chosen from the getTemplates() callback.

Are there any actions allowed to be triggered programmatically on the scan screens (ScanIDCard, FaceMatch, etc.)?

For features like ScanIDCard and FaceMatch, these are returned as fragments, allowing you to draw on top of them. For ScanIDCard, you receive each page in order via the onComplete callback. The camera will automatically start scanning the next page after each callback.

On the FaceMatchActivity there is a static document image being loaded to be matched against the face. How can I get the document scanned on the previous step to pass it to the face match?

You can obtain the imageUrl from the onComplete callback for document scanning, convert it to base64, and use it in face match.

What happens if the document had multiple sides? How can I find which document had the photo of the user to pass to the FaceMatch?

You must choose an order (page) that contains a face to set it for face match. When the document URL is retrieved from the document scan, we flag that it has a face extracted based on the template.

What data should you submit on startSubmitData? Are these specified from the admin page?

You must prepare your list of steps with the outputProperties obtained for each step. For example:

sharedStepDefinitionsStepFace = SubmitRequestModel(item.stepId, StepsName.FaceMatch, values);
sharedStepDefinitionsStepDoc  = SubmitRequestModel(item.stepId, StepsName.DocumentCapture, values);

On the StepNames file there are various keys used for uploading the data on startSubmitData. How do these correlate with the admin page? Where can I find this mapping?

In Flow Builder, after selecting a flow from Flow Manager, you are presented with the various blocks that make up the flow. To view the additional keys per flow, select the magnifier icon on the BlockLoader; you can press the magnifier icon and see all the data there. You can also define and add new data points.