canonical: https://jentic.com/apis/fraudlabspro.com

# FraudLabs Pro APIs

FraudLabs Pro provides fraud screening and SMS verification for online merchants. Its APIs score orders and transactions for chargeback risk in real time, return recommended actions with signals such as IP proxy detection and address mismatch, accept post-transaction feedback that tunes the per-merchant model, and send and confirm one-time SMS codes for step-up authentication. The portfolio spans a version 2 API that combines screening, feedback, and SMS verification, a version 1 fraud-detection API for legacy integrations, and a standalone SMS verification module.

## For AI agents

An agent can score checkout orders for fraud risk, branch on the recommended action, send approved, rejected, or chargeback feedback to improve the model, and run SMS one-time-code step-up verification on borderline transactions. It can work through either the combined v2 API or the separate v1 fraud-detection and SMS verification endpoints.

## Scope

Use for: Real-time order and transaction fraud scoring, post-transaction feedback that tunes the per-merchant model, and SMS one-time-code step-up verification for online merchants

Not supported:
- payment processing
- refunds
- chargeback dispute filing
- voice OTP
- TOTP authenticator apps
- device fingerprinting

## APIs

| API | Category | Endpoints | Description |
| --- | --- | --- | --- |
| FraudLabs Pro API | security | 5 | Score an order for fraud risk in real time, send post-transaction feedback to improve the model, and run SMS step-up verification when needed. |
| FraudLabs Pro Fraud Detection | security | 2 | Score a transaction for fraud risk and submit post-transaction feedback through FraudLabs Pro's v1 fraud-detection endpoints. |
| FraudLabs Pro SMS Verification | identity-auth | 2 | Send a one-time SMS verification code to a customer's phone and confirm the code they enter. |

## Which API to use

| Need | API | Why |
| --- | --- | --- |
| Screen orders, send feedback, and run SMS verification through one current API | fraudlabs-pro | The v2 API combines real-time screening, order-result retrieval, post-transaction feedback, and SMS step-up in a single five-endpoint interface for new integrations. |
| Score transactions and send feedback on a legacy v1 integration | fraudlabs-pro-fraud-detection | The v1 fraud-detection API keeps the same scoring contract under /v1 paths for customers who already call it, without the SMS endpoints. |
| Send and confirm SMS one-time codes on their own | fraudlabs-pro-sms-verification | The standalone SMS verification module exposes just the send and result endpoints for step-up authentication without order screening. |

## Cross-API use cases

### Score an order then step up borderline cases with SMS

An agent scores a checkout order for fraud risk, and when the recommended action is review it sends a one-time SMS code to the customer and confirms the entered code before letting the order proceed. A successful verification softens the risk signal so the order clears without manual review.

Example prompt: Call the v1 screen endpoint for a fraud score; if the action is review, call the SMS verification send endpoint, prompt the customer for the code, then confirm it with the SMS verification result endpoint before approving

### Migrate a legacy fraud integration to the current API

An agent runs an existing v1 fraud-detection integration while trialling the v2 API in parallel, comparing scores and recommended actions on the same orders before switching the production path. Both share the same license key and scoring contract, so the cutover needs no new credential.

Example prompt: Send each order to both the v1 screen endpoint and the v2 screen endpoint, diff the returned scores and actions, and route production traffic to v2 once the results align

### Close the loop from screening to verified outcome

An agent screens an order, applies SMS verification on borderline cases, then records the realised outcome as feedback once the bank settles or disputes the charge. The feedback retrains the per-merchant model so later screens reflect what is actually fraud for that store.

Example prompt: Screen the order and run SMS step-up where needed via the v2 API, then post the settled or charged-back outcome to the v2 feedback endpoint so the model learns the result

## Why Jentic

- **Setup:** Wiring these APIs by hand means appending the license key to every request and building your own retry and error handling across the screening, feedback, and SMS verification endpoints. With Jentic One you install once, add the FraudLabs Pro operations from the Jentic directory, store the key one time, and your agent calls any of them.
- **Permission scoping:** FraudLabs Pro carries order details, phone numbers, and codes in the request body rather than the URL path, so scoping is operations-only. Your own rules limit the agent to the operations it needs, such as scoring an order or confirming a code, and leave out sending SMS or posting feedback unless you add them. Every operation the agent can run is one you chose.
- **Credential handling:** Your FraudLabs Pro license key is stored encrypted by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search the Jentic directory by intent, such as 'score an order for fraud risk' or 'send an SMS verification code', and Jentic returns the matching FraudLabs Pro operation with its input schema so the agent calls the right endpoint without reading the reference docs.

## Related vendors

- **Sift** — A higher-end fraud and trust platform with broader event types and configurable ML, an alternative for high-volume merchants with dedicated risk engineering.
- **MaxMind minFraud** — Focuses on IP intelligence and device-based fraud signals, an alternative when geolocation and IP signals dominate the scoring need.
- **Twilio Verify** — Multi-channel verification across SMS, voice, email, and TOTP, an alternative to the SMS one-time-code step-up flow at larger scale.
- **Fingerprint** — Provides a stable visitor identifier that complements order screening by catching returning fraudsters across cleared cookies.

## FAQ

### What can an AI agent do across the FraudLabs Pro APIs?

An agent can score orders and transactions for chargeback risk, retrieve a prior screening verdict, send approved, rejected, or chargeback feedback to tune the per-merchant model, and send and confirm one-time SMS codes for step-up verification. These span the v2 API, the v1 fraud-detection API, and the standalone SMS verification module.

### How do the three FraudLabs Pro APIs relate to each other?

The v2 API is the current interface and combines screening, result retrieval, feedback, and SMS verification in one place. The v1 fraud-detection API keeps the same scoring contract under /v1 paths for legacy integrations, and the SMS verification module exposes just the send and confirm endpoints for use on their own.

### Do I need separate credentials for each API?

No. All three APIs use the same FraudLabs Pro license key on every request, so a single credential works across order screening, feedback, and SMS verification.

### Which API should a new integration use?

New integrations should use the v2 API, which brings screening, feedback, and SMS step-up together under one interface. The v1 fraud-detection API is best kept for customers who already call its /v1 paths, and the standalone SMS module suits workflows that only need one-time-code verification.

### How does SMS verification fit into a fraud workflow?

When a screen returns a review recommendation, the merchant sends a one-time SMS code to the customer and confirms the entered code. A successful match shows the buyer controls the phone on file, which softens the risk signal and lets the order proceed without manual review.

### How is fraud scoring accuracy improved over time?

Every screen can be followed by a feedback call once the outcome is known, marking the order approved, rejected, or charged back. FraudLabs Pro uses that feedback to retrain the per-merchant model so future screens better reflect that store's real fraud patterns.