All documentation

Overview

Base URL, authentication, decision types, risk scores, and rate limits.

Overview

Getting Started

Naiza provides a RESTful JSON API for real-time fraud detection. Submit events with customer, device, and session context to receive instant ALLOW / BLOCK / REVIEW decisions with risk scores and matched rules.

Base URL

https://api.naiza.ai/api/v1

Authentication

All API requests require an API key passed via the x-api-key header. Keys follow the format naiza_api_sk_live_... for production or naiza_api_sk_test_... for testing.

curl -X POST https://api.naiza.ai/api/v1/events \
  -H "x-api-key: naiza_api_sk_live_abc123..." \
  -H "Content-Type: application/json" \
  -d '{"eventName": "user.login", ...}'

Decision Types

ALLOW
No fraud signals detected
Proceed normally
REVIEW
Suspicious indicators found
Require additional verification (e.g., MFA)
BLOCK
High-confidence fraud detected
Deny access immediately

Decisions are determined by your configured rules. BLOCK rules short-circuit immediately. REVIEW is triggered when a review rule matches or the accumulated score reaches the threshold.

Risk Score

Risk scores accumulate from SCORE-type rules during evaluation. A score of 100 or above triggers a REVIEW decision.

RangeLevelDescription
0-25LowNormal behavior, minimal fraud indicators
26-50MediumSome suspicious indicators present
51-75HighMultiple fraud indicators detected
76-100+CriticalStrong fraud signals — score >= 100 triggers REVIEW

Rate Limits

API requests are subject to quota limits per tenant. Exceeding the limit returns a 429 status with a Retry-After header.

Default quota
10,000 req/hr
Window
1 hour (rolling)

Response Time

Typical
50-200ms
99th percentile
<500ms