Xobriq

Get started

Quickstart

Run your first Guard fraud score in under five minutes.

1. Create your account

Sign up at /register with a work email. You'll be dropped into the Console with a sandbox environment pre-configured.

2. Grab a sandbox API key

From the Console → API Keys, generate a sandbox key. Keys prefixed with sk_test_ are safe for local development.

Never expose production keys client-side

Sandbox keys can safely live in your dev environment. Production keys (sk_live_) must only be used server-side.

3. Score your first transaction

bash
`curl -X POST https://api.xobriq.com/v1/guard/score \
  -H "Authorization: Bearer sk_test_..." \
  -H "Content-Type: application/json" \
  -d '{
    "transaction_id": "txn_demo_001",
    "amount": 150000,
    "currency": "KES",
    "customer_id": "cust_demo",
    "device_fingerprint": "fp_demo"
  }'`

4. Interpret the response

json
`{
  "risk_score": 0.03,
  "decision": "allow",
  "latency_ms": 42,
  "signals": {
    "velocity": "normal",
    "geo": "verified",
    "device": "trusted"
  }
}`

A risk_score under 0.20 is safe to allow, 0.20–0.70 is worth reviewing, above 0.70 should be blocked or step-up authenticated.

5. Where to go next

  • Read Concepts to understand risk scoring, agents and workflows.
  • Explore the Guard reference for deepfake, identity, and behavioural APIs.
  • Provision a GPU instance on Xobriq Cloud.
  • Deploy a pre-built agent from the Agentic overview.
← WelcomeChoose a workflow →