Validation with video

Version: always use the latest available version

This agent lets you verify a person’s identity without leaving the chat, combining real-time video selfie (liveness check), document capture, and automatic facial comparison (1:1).

Ideal for onboarding, access control, electronic signatures, or high-risk validations.

Prerequisites

✅ Checklist to start without blockers

You must have a Terms and Conditions PDF (public URL) to configure in the agent. This is mandatory.
The user must be able to record video from the chat (camera permissions must be enabled on the device).
Recommended: define the handling flow if you enable Human in the loop (who reviews, response times, and what happens if there is no response).

How does it work?

In less than 1 minute, the user completes the first two steps:

Record a video selfie

The user records a short real-time video, following simple instructions within the chat.

Capture photos of their identity document

Front and back photos of the identity document are requested to validate validity and authenticity.

Automatic facial comparison

The system applies the 1:1 technique, comparing a frame from the video with the image on the document (or with the official government source if enabled).

How is the Agent composed?

The Biometrics Agent is made up of three specialized agents:

Liveness check agent - Liveness validation via video selfie with OTP
Document verification agent - Verification of document authenticity and validity
Facial comparison agent - 1:1 facial comparison between the video and the document/government entity

If you only need to implement part of the process (for example, document only), see the configuration in Agents by stage.

Connection from Marketplace

Access the platform

Go to apps.jelou.ai. On the home screen you will see Brain Studio and Connect. In Brain Studio, select Marketplace.

Connect the integration

Search for KYC and you will see the integration details, including pricing, a trial feature, and availability by country. To use the free trial, click “Connect”. Once connected, the work canvas in Brain will open so you can configure and use it.

Initial configuration

To use video validation, select “Passive liveness biometrics”
The first thing to configure is the version found under “Advanced settings” (at the bottom).
None of the inputs (except the PDF) require changes if you will use the default values. If you need to change them, you can enter them in the corresponding field.
If you want to modify any input, go to the Settings section.

Configure Outputs

This agent has 8 error outputs and 1 success output. Each can be directed to:

A text input with a custom message.
Connect, only if you have this module.

Run tests

With the configuration you just completed, you can proceed to run tests using a skill or another agent as a precursor.

Important: The “Terms and Conditions PDF” field is mandatory. The document must explicitly include the use, storage, and handling of personal and biometric data. It must be presented and accepted before starting the verification process.

Configuration

Inputs
Outputs

Language

string

default:"Es"

Variable: language. Defines the language of the experience.Available values: Es (Spanish), En (English)

Maximum biometrics attempts

number

default:"2"

Variable: maxBiometricAttempts. Maximum number of biometrics attempts. Available values: 1-4

Notification email for exceeded retries

string

Variable: customerServiceEmail. Email to be notified when the maximum number of attempts is exceeded.

Terms and Conditions PDF (mandatory)

string

Variable: termsAndConditionsPdf. Public URL of the Terms and Conditions PDF.It must explicitly detail the handling of personal data, including the use, storage, and processing of biometric information. It must be shown to the user before starting the process.

Enable liveness introduction video

boolean

default:"false"

Variable: enableIntroVideo. Determines whether an introduction video should be shown to the user before starting the process.

URL to display liveness introduction media

string

Variable: introMediaUrl. URL shown to the user to load an introduction video or image.

Maximum OTP code duration

number

default:"1"

Variable: otpDuration. Maximum duration in minutes of the OTP code.Available values: 1-10

OTP code length in Chat

number

default:"4"

Variable: otpLength. Number of digits in the OTP code.Available values: 3-6

Verify document expiration date

string

default:"Disabled"

Variable: dateExpiracyCheck. Verifies the expiration date of the document.Available values: Disabled, Enabled

Enable government validation

boolean

default:"false"

Variable: flagGovValidation. Enables government validation; when false, the civil registry is not queried.

Accepted threshold for facial comparison

number

default:"60"

Variable: facematchThreshold. Minimum percentage required in facematch.Available values: 55-95

Enable Human in the loop

boolean

default:"false"

Variable: enableHumanInLoop. Indicates whether the review process with a human agent is activated.

Enable LipSync

boolean

default:"false"

Variable: enableLipSync. Enables LipSync to compare lip movement with audio.

Max. agent attempts

string

default:"Unlimited"

Variable: retriesAgent. Maximum number of agent attempts before blocking the user.Available values: Unlimited, 1, 2, 3

User block duration (in hours)

number

default:"2"

Variable: blockingInHours. Block duration (in hours) after exceeding attempts.Available values: integer (e.g.: 2, 24, 168)

Success

Biometric Validation Approved

Biometric validation has been completed successfully. Variable: biometricVerificationSuccess Response structure (JSON)

{
  "response_document_check_success": {
    "document_check": {
      "verified_fields": {
        "...": "Main document data"
      },
      "secondary_fields": {
        "...": "Secondary document data"
      },
      "details": {
        "...": "Document type details"
      },
      "image_quality_details": {
        "...": "Image quality details"
      },
      "images_extracted": {
        "...": "Images extracted from the document"
      },
      "gov_entity_fields": {
        "document_number": "Identity document number"
      }
    },
    "gov_entity_data": {
      "...": "Data obtained from the country's government entity"
    },
    "document_image_front_url": "URL of the front image of the document",
    "document_image_back_url": "URL of the back image of the document",
    "document_face_image_url": "URL of the face photo from the document"
  },
  "response_liveness_success": {
    "url_video_liveness": "URL of the video sent by the user",
    "url_selfie_image": "URL of the frame obtained from the video to validate liveness"
  },
  "result_facematch": {
    "facematch_result": "Facematch result (e.g. `approved` or `decline`)",
    "facematch_confidence": 99
  },
  "reporte_de_biometria": {
    "report": "URL of the generated biometrics report",
    "report_pdf": "URL of the generated biometrics PDF report"
  }
}

Important fields:

response_document_check_success: Identity document information
response_liveness_success: Video submitted to validate liveness
result_facematch: Facial comparison result
reporte_de_biometria: Report URLs in web and PDF format

JSON response variable definitions: response_document_check_success:

document_check: Object with all extracted and validated document information
verified_fields: Main verified fields of the document (names, surnames, dates, etc.)
secondary_fields: Secondary fields extracted from the document
details: Details of the recognized document type
image_quality_details: Results of the image quality validations
images_extracted: URLs of images extracted from the document (portrait, signature, etc.)
gov_entity_fields: Fields used to validate against the government entity
gov_entity_data: Data obtained from the country’s government entity (varies by country)
document_image_front_url: URL of the front image of the document
document_image_back_url: URL of the back image of the document
document_face_image_url: URL of the face photo extracted from the document

response_liveness_success:

url_video_liveness: URL of the video selfie submitted by the user during the liveness check
url_selfie_image: URL of the frame extracted from the video used to validate liveness

result_facematch:

facematch_result: Facial comparison result (approved if matched, decline if not matched)
facematch_confidence: Comparison confidence percentage (0-100)

reporte_de_biometria:

report: URL of the biometrics report generated in web format
report_pdf: URL of the biometrics report generated in PDF format

Errors

Service Error

A problem occurred with the service. Variable: serviceError

Document Validation Unsuccessful

The user’s document could not be validated. Variable: documentCheckError

Terms and Conditions Not Accepted

The user has not accepted the Terms and Conditions required to continue. Variable: termsNotAcceptedError

Process Abandoned

The user abandoned the process. Variable: userExitProcess

Error saving to Datum

An error occurred while creating or updating the record in Datum. Variable: datumError

Liveness check unsuccessful

The user’s liveness could not be detected. Variable: livenessError

Facial Recognition Verification Error

Facial recognition verification has not been completed successfully. Variable: facematchError

Time Expired

The time to complete the process has expired. Variable: timeExpiredError

User Blocked

User Blocked. Variable: userBlocked

Report

By default a database called “Biometrics Agent Report” is created with the following fields:

Field	Description
Biometrics Code	Unique process identifier
Date and Time	Verification timestamp
Biometrics Result	Final process status
Identification Number	User’s document
Identification Type	Document type
First Names	User’s first names
Last Names	User’s last names
Date of Birth	Date of birth
Phone	User’s phone number
Email	Email address
Address	User’s address
Liveness Check Result	Liveness check status
Video Selfie	Video URL
Selfie Photo	URL of extracted photo
Document Check Result	Document validation status
Front Document Photo	URL of front image
Back Document Photo	URL of back image
Face Photo on Document	URL of document face photo
Government Entity	Entity queried
Facematch Result	Facial match percentage
Biometrics Report	Web report URL
Failure Description	Error details (if applicable)
Observations	Additional notes

The biometric report is available for download in PDF format and contains all the information detailed in the table.

Frequently asked questions

Does the agent respond if I don't enter data in any field?

Yes, the agent can respond because all fields have default values. However, selecting the agent version is mandatory; we recommend always selecting the latest available public version.

What documents are needed during the biometric process?

Photos of the identity document from both sides are required: front and back.

Is it possible to skip any step in this agent?

No. This agent is designed as a complete flow: video selfie + document + facematch. If you only need part of it (for example, document only), you must install the agents by stage.

Can I customize the texts in this agent?

Yes, you can modify the interaction texts. If you need to add questions, conditionals, or other advanced customizations, we suggest consulting with the Jelou team.

Can I change or remove the terms and conditions?

No. The Terms and Conditions PDF is mandatory to use this agent.

What happens if the OTP expires or the user abandons the flow?

If the OTP expires, the entire process expires and the user must restart. If the user abandons the flow, the process remains unfinished and expires according to the agent/experience logic.

How many attempts does the user have before being blocked?

It depends on your configuration:

Maximum biometrics attempts (maxBiometricAttempts): controls biometrics attempts.
Max. agent attempts (retriesAgent): controls agent retries before blocking.
User block duration (blockingInHours): defines how many hours the block lasts.

Getting Started

AI Agents

Brain

Nodes

Databases

Variables

Tools

Functions

Configuration

Marketplace

Enterprise

Billing

Prerequisites

✅ Checklist to start without blockers

How does it work?

How is the Agent composed?

Connection from Marketplace

Configuration

Success

Errors

Report

Frequently asked questions

Getting Started

AI Agents

Brain

Nodes

Databases

Variables

Tools

Functions

Configuration

Marketplace

Enterprise

Billing

​Prerequisites

​✅ Checklist to start without blockers

​How does it work?

​How is the Agent composed?

​Connection from Marketplace

​Configuration

​Success

​Errors

​Report

​Frequently asked questions

Prerequisites

✅ Checklist to start without blockers

How does it work?

How is the Agent composed?

Connection from Marketplace

Configuration

Success

Errors

Report

Frequently asked questions