canonical: https://jentic.com/apis/faceidentity-beta.azurewebsites.net/faceidentity-api-documentation # Faceidentity Beta Face Identity API Documentation Face Identity is a beta authentication service hosted on Azure App Service that exposes a customer registration and token issuance flow. The two endpoints register a customer record and issue an authentication token that can be used to call protected resources. The surface is minimal and intended as the auth layer in front of a face identity verification stack. ## For AI agents Register a customer and obtain an authentication token for the Face Identity beta service. Used as the auth layer in front of face verification operations. ## Scope Does not handle face verification, document checks, or biometric matching - use for the customer registration and token issuance bootstrap only. ## Capabilities - Register a new customer record on the Face Identity service - Issue an authentication token for an existing customer - Bootstrap a session against the Face Identity beta endpoint - Verify the customer registration request payload was accepted - Refresh access by re-issuing a token from the customer credentials ## Use cases ### Customer Onboarding for Face Identity An application that consumes the Face Identity beta service must first register the end-customer and then exchange those credentials for a token. The two-call flow registers the customer at `/authentication/customer/registration` and immediately calls `/authentication/customer/token` to obtain a session token used by downstream face verification operations. Designed for low-volume integration testing rather than production identity workloads. Example prompt: POST customer registration payload to `/authentication/customer/registration`, then POST the same credentials to `/authentication/customer/token` and store the returned token for the session. ### Token Refresh in a Verification Pipeline A scheduled face verification job needs a current token before invoking downstream operations on the Face Identity service. The agent calls the token endpoint with stored customer credentials to obtain a fresh token, with no separate refresh-token flow exposed in the spec. Suited to short-lived workers that pick up credentials from a secret store and run a single verification batch. Example prompt: Call POST `/authentication/customer/token` with the stored customer credentials, capture the returned token, and attach it as the Authorization header for subsequent Face Identity calls. ### AI Agent Bootstrap Step An AI agent that orchestrates a face verification workflow uses Jentic to discover and execute the auth bootstrap before running the verification operations. Jentic exposes both registration and token issuance as searchable operations, so the agent can register and authenticate without reading the Azure-hosted documentation page. Useful when face identity is one step in a longer KYC-style flow. Example prompt: Use jentic.search with query 'register face identity customer', execute the registration operation, then load and execute the token operation to obtain a session token for the agent. ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/authentication/customer/registration` | Register a new customer | | POST | `/authentication/customer/token` | Issue an authentication token for a customer | ## Key resources - **Authentication** — Customer registration and token issuance for the Face Identity service ## Why Jentic - **Setup:** Wiring the Face Identity API by hand means targeting the faceidentity-beta.azurewebsites.net host and chaining the customer registration and token routes yourself. Through Jentic you install once, import the Face Identity API from the API Directory, and your agent calls it: the spec declares no authentication scheme, so there is no standing secret to store. - **Permission scoping:** Face Identity carries customer details in the request body rather than a resource id in the path, so you limit the agent to the operations it needs: customer registration and token issuance. You choose those operations, so nothing beyond that bootstrap is included. - **Credential handling:** The Face Identity API declares no authentication scheme, so your own Jentic One instance holds no standing secret for it, and the session token returned by the token endpoint never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'register a Face Identity customer' or 'issue a Face Identity token', and Jentic returns the two operations with their request schemas so the agent chains registration and token issuance without reading the Azure docs page. ## Related APIs - **Onfido** — Production-grade identity verification with document and biometric checks - **Veriff** — Identity verification platform with document, liveness, and biometric flows - **FaceCheck.ID** — Reverse face search across the open web for the same probe image ## FAQ ### What authentication does the Face Identity API Documentation service use? The OpenAPI spec does not declare a global security scheme. The two endpoints accept a customer credentials payload and return a token in the response - that token is then used as a bearer credential for downstream Face Identity operations not covered by this spec. ### Can I register a new customer with the Face Identity API? Yes. POST a customer credentials payload to `/authentication/customer/registration.` The response confirms registration and the same credentials can then be POSTed to `/authentication/customer/token` to obtain a session token. ### What are the rate limits for the Face Identity API? Rate limits are not declared in the spec. The service runs on Azure App Service in beta, so production-grade quotas are not published - treat this as a beta endpoint and avoid high-volume traffic. ### How do I obtain a session token through Jentic? Run pip install jentic, call jentic.search with 'register face identity customer', execute the registration operation, then load and execute `/authentication/customer/token` to receive the bearer token. ### Is this Face Identity service production-ready? No. The host is faceidentity-beta.azurewebsites.net and the service is in beta. The two-endpoint surface covers registration and token issuance only and does not include the full verification flow. ### Can I limit what my agent is allowed to do with the Face Identity API? Yes. Because you run Jentic One yourself, your own rules decide which of the two Face Identity operations the agent may call: customer registration at `/authentication/customer/registration` and token issuance at `/authentication/customer/token.` You can grant the agent just the token issuance step, or both, so it never reaches beyond the registration and token bootstrap you approve. The spec declares no authentication scheme, so there is no standing secret to hand out, and the session token the endpoint returns stays out of the agent's prompt and logs.