GET STARTED

Introduction

Learn how you can prevent fraud and improve user experiences with Trustfull.

What is Trustfull?

Trustfull is a fraud prevention and identity risk intelligence platform that helps businesses make fast, explainable decisions about who to trust online. It analyzes email addresses, phone numbers, IPs, devices, and behavioral signals in real time, combining proprietary models with open-source intelligence (OSINT) and trusted data sources, to detect fake users, bots, and high-risk activity before fraud happens. Designed for mid-market and enterprise companies, Trustfull delivers low-latency risk scores with clear reason codes, enabling teams to block abuse, apply the right level of friction, and stay compliant without hurting user experience or conversion.

How it works?

Fraud AgentsActivate data collectionPhone NumbersCarriers, Breaches, Online Presence...Email AddressesSMTP Providers, Registrars, Blacklists...IP AddressesGeoIP Providers, Proxy Lists, VPNs...WebsitesSSL Authorities, WHOIS, Web Archives...BrowsersSession Data, Plugins, Canvas Data...DeviceDevice Graphs, Fraud Networks, Apps...INGESTFraud ModelsSelect or create risk modelYour ModelCustom ModelTrain agents to detect your fraud typeOnboardingStandard ModelSignup fraud preventionPhone RiskStandard ModelOTP abuse, spam detectionEmail RiskStandard ModelDisposable, deliverabilityMerchant RiskStandard ModelDomain reputation, fake companiesLead QualificationStandard ModelIntent scoring, qualityFree trial abuseStandard ModelMulti-account detection, abuse patternsEXECUTEDashboardDecide and take actionModel ConfigRules, scorecards & explainationsData AnalyticsMonitor agents performanceModel InsightsMonitor models performanceTeam AccessRole-based permissionsAPI KeysManage access credentialsIntegrationsConnect external services

There are five ways to send data to Trustfull and get a score back. This page explains what each one is for, and it sets out a path to follow if you are starting from scratch.

If you are new here, read The Free Trial and Before You Start, then work through Your First Steps in order. Between them they take you from an empty account to your first score.

If you learn better from a worked example than from a description, skip ahead to Recipes and follow one of those instead.

The Free Trial

You can try Trustfull before you pay anything.

Every new account comes with a free trial, and the trial runs for 15 days. When the account opens, Trustfull adds 50 € of credit to it. That credit is yours to spend as you like.

Credit is what pays for lookups. Each lookup takes a small amount out of your balance, so 50 € will cover a good number of them while you find your feet. There is nothing special about trial credit. It behaves exactly like credit you have paid for, and it works with every integration method on this page.

📘

The sandbox does not touch your credit

Requests you send to the Sandbox are free of charge, so they never come out of your 50 €.

The sensible approach is to build and debug your code against the sandbox first. Then spend your free credit on real lookups, once you want to see real answers.

⚠️

If the credit does not arrive

Open the dashboard and look at your balance. If the 50 € is not there, you do not need to send an email and wait for a reply.

Click Ask AI in the dashboard header. That opens our AI assistant. It can answer the question on the spot, and if somebody needs to look into it properly, it will open a support ticket for you.

Before You Start

You need three things. No integration works without them.

What you needHow to get it
A Trustfull accountYou can open an account using the signup link or book a demo with our sales team.
An API keyOpen the Dashboard and look under API Keys. It is also called an App Key.
Credit on your balanceA new account starts with 50 € of free credit. Every lookup spends a little of it. See Billing to add more.
⚠️

Keep your API key on your server

The API key identifies your account. Anyone who holds it can spend your credit.

Never put the key in browser code. Never put it in a mobile app. Never put it in a public repository. Send every request from your own backend.

The Five Integration Methods

All five send Trustfull much the same kind of data. What changes between them is who does the sending, and how many records travel at a time.

MethodUse it whenYou needPage
REST APIYou want a score inside your own product, in real time.A backend, and an API key.Your First API Call
Run CheckYou want to check one person by hand.The dashboard only.Run Check
UploadsYou want to check many people at once.A CSV file.Uploads
MCP ServerYou want an AI assistant to run the checks.An MCP client.MCP Server
JavaScript SDKYou want browser and device signals.A page you can edit.Session Integration

REST API

This is the main integration method, and it is the one most customers build on. Your backend sends a request, and Trustfull sends a score back.

Every request goes to https://api.fido.id, with your key in the x-api-key header.

Start with Your First API Call. It shows a working request in cURL, Python and Node.js. Then read the reference page for the product you want, such as Trustfull Phone or Trustfull Email.

Run Check

Run Check lives in the dashboard. You type in a phone number or an email address by hand, and Trustfull runs the check and shows you what comes back.

This is a real lookup rather than a preview, so it returns exactly what the API would return, and it costs the same credit.

It is the quickest way to see what Trustfull actually gives you, before you commit to writing any code. See Run Check.

Uploads

Uploads takes a CSV file and runs the same check on every row in it.

Reach for it when you have a whole list of people to score and calling the API once per person would be tedious. A single file can hold up to 200,000 rows.

See Uploads for the file format and the rules.

MCP Server

The MCP Server connects an AI assistant to Trustfull, which lets the assistant score a phone number, an email address or an IP address by itself, in the middle of a conversation.

It suits fraud workflows where an agent is doing the driving rather than a person. See MCP Server.

JavaScript SDK

The other four integration methods send Trustfull data you already have. This one is different, because it gathers data you do not have yet.

You add a small script to your page, and it quietly collects browser and device signals while the visitor is there. Your backend then asks Trustfull what those signals mean.

This is what you want if your problem is bots and automated signups. See Session Integration, and the recipe Approve or reject a signup.

Your First Steps

Follow these five steps in order. They work for every integration method.

1. Get your API key

Open the Dashboard and look under API Keys.

If there is no key waiting for you there, click Ask AI in the dashboard header and say so. The assistant will get one issued for you.

Your 15 day trial begins the moment the account opens, and the 50 € of free credit is already sitting on the balance. There is nothing to claim and nothing to activate, so you can start straight away.

2. See a result before you write code

Open Run Check in the dashboard and run a single check by hand.

Once you have seen a real result on screen, the response format in the next step will make far more sense.

3. Test without spending credit

Send your first requests to the Sandbox.

The sandbox costs nothing and returns made-up data. You choose which answer comes back by setting the customer_id field to high, good, moderate, bad or poor.

This means you can write and debug the whole integration before any real money is involved.

If you would rather follow a worked example, there is a recipe for exactly this: Make a sandbox call.

4. Make one real call

Follow Your First API Call and send a single real request to https://api.fido.id.

Check that a score comes back, and that your own code reads it correctly. This one is paid for out of your free credit.

The recipe Assess the risk of a mobile phone number walks through the same call step by step.

5. Learn to read the answer

A score on its own will only take you so far. You also need the reason codes, because they explain why the score came out the way it did.

Read Understanding Scores and Reason Codes.

The recipe Understand a score decision using reason codes shows how to do this against a real response.

Which Integration Method Should I Choose?

Your situationUse this
I want a score during signup, in real time.REST API
I want to score a list of past customers.Uploads
I want to look at one suspicious person.Run Check
I want to catch bots on my signup page.JavaScript SDK
I want my AI agent to do the checking.MCP Server
I only want to try Trustfull out.Run Check, then the Sandbox. Your free credit covers the Run Check.

Nothing stops you using several at once. A common arrangement is the REST API for live traffic and Uploads for older data that you want to catch up on.

Recipes

A recipe is a short worked example that you can follow from start to finish. These are the ones worth knowing about.

RecipeWhat it shows you
Make a sandbox callYour first request, with no credit spent.
Assess the risk of a mobile phone numberA complete phone lookup, from request to answer.
Understand a score decision using reason codesWhy a score came out the way it did.
Approve or reject a signupTurning a score into a yes or no at registration.
Create a custom scorecardScoring on your own rules rather than the standard ones.
Tune API latency with max_enrichment_timeTrading a little accuracy for a faster answer.
How to access a datasetDownloading the bulk datasets from S3.

FAQ

📘

Is there a free trial?

Yes. Every new account gets one, and it runs for 15 days.

Trustfull also adds 50 € of credit when the account opens, so you can start making real lookups immediately.

📘

What can I spend the 50 € on?

Anything you like. Any lookup, through any of the integration methods on this page.

Trial credit is ordinary credit, so nothing is held back from you.

📘

Does the trial cost me anything?

No. Trustfull puts the 50 € on your account, and you are not charged for it.

📘

The 50 € has not appeared on my account. What should I do?

Open the dashboard and click Ask AI in the header, then explain what has happened.

Our AI assistant will either sort it out there and then, or open a support ticket so that somebody can look into it for you.

📘

How do I get an API key?

You will usually find one already waiting in the Dashboard, under API Keys.

If there is nothing there, click Ask AI in the dashboard header and explain what you need. Keys cannot be created by hand from your side, so the assistant will either issue one or pass the request on to somebody who can.

📘

Can I try Trustfull without writing any code?

Yes. Use Run Check in the dashboard. You type in the data by hand, and you get a full result back.

You still need an account, an API key and credit. A Run Check costs the same as an API call.

📘

Can I test without spending credit?

Yes. Use the Sandbox.

Send your requests to sandbox.trustfull.com instead of api.fido.id. You can also keep the normal address and add the header demo: 1.

The sandbox returns made-up data. It is there to test your code, not to score real people.

📘

What is the base URL?

It is https://api.fido.id. All traffic goes over HTTPS.

📘

How do I authenticate?

Put your API key in the x-api-key header. Every request needs it.

There is no login step and no token to refresh. The key is enough.

📘

Can I use more than one integration method?

Yes. They all use the same account, the same key and the same credit balance.

A common pattern is the REST API for live traffic, and Uploads for a backlog of past customers.

📘

Which integration method is fastest?

The REST API. It answers in real time, so you can use it while a user waits.

Uploads is not built for speed. It is built for volume, and it runs in the background.

📘

Do I need a separate key for each product?

No. One API key covers every product your account is allowed to use.

If a product returns nothing at all, the likely explanation is that it is not switched on for your account. Click Ask AI in the dashboard header and ask, and the assistant will check for you.

📘

What happens if a request fails?

The API returns an HTTP status code that explains the problem. See HTTP Status Codes.

Related


Did this page help you?