City of Hats - Exposure Intelligence Platform
🧠 Intelligence API

Risk Scoring API

Real-World Identity Risk Based on Attacker Behavior

The Risk Scoring API transforms raw exposure signals into a single, trustworthy 0–100 identity risk score β€” powered by AI correlation across breach data, dark-web telemetry, and attacker behavior.

Instead of asking:

"Has this identity been leaked?"

β†’
Your teams finally know:

"How risky is this user β€” right now?"

Trusted by:
🏦 Banks πŸ’³ Fintech πŸ›’ Marketplaces πŸ“± Telcos ☁️ SaaS Platforms

Risk Scoring In The Real World

See how the Risk Scoring API transforms suspicious activity into actionable intelligence.

Risk Score Velocity Review Example - Login from new device with Risk Score 78

Example: Login from new device β†’ Risk Score = 78 β†’ Trigger: MFA + velocity risk review

1

Event Detected

Login attempt from unfamiliar device + location

β†’
2

Risk Analysis

Score = 78 (High Risk) based on velocity + device signals

β†’
3

Action Triggered

Step-up MFA enforced + analyst velocity review

What The Risk Score Represents

Each user lookup produces a continuous risk score from 0–100, designed for policy engines, fraud models, IAM, and SOC.

Score Range Meaning Typical Action
0–24 Low Risk Allow
25–59 Elevated Risk Monitor / Step-Up
60–79 High Risk Step-Up / Review
80–100 Critical Risk Block / Investigate

Inputs Into The Model

Risk scoring combines multiple intelligence dimensions into one unified score.

πŸ”

Credential Exposure Intelligence

  • Password reuse likelihood
  • Breach recency
  • Exposure depth
  • Hashed / plaintext classification
  • Attacker interest trends
πŸ•΅οΈ

Dark-Web & Criminal Market Signals

  • Reposts & combo list circulation
  • Trade volume
  • Mention intelligence
  • "Ready-to-use" credential flags
  • Marketplace activity
πŸ§‘

Identity Trust Indicators

  • Email age
  • Alias / burner / masked
  • Corporate vs consumer domain
  • Validation integrity
  • Passive risk footprint
πŸ€–

Bot & Abuse Risk

  • Behavioral anomalies
  • Known fraud vectors
  • Throw-away lifecycle patterns
  • Velocity signals
  • Registration timing

Real-World Threat Funnel Stage

We don't just check if credentials leaked β€” we analyze where they are in the criminal process.

CLEAN 0%
β†’
LEAKED 15%
β†’
REPOSTED 35%
β†’
TRADED 55%
β†’
TESTED 75%
β†’
ATTACKED 90%
β†’
MONETIZED 100%

Risk increases as the funnel progresses β€” from passive exposure to active criminal monetization.

Output You Receive

Each API response includes everything your system needs to decide automatically.

πŸ“Š

Risk Score

0–100 continuous scale

🎯

Confidence Level

low / medium / high

πŸ“

Threat Funnel Stage

Attack lifecycle position

🏷️

Risk Reason Codes

Explainable factors

⚑

Suggested Action

allow / step_up / block

API Reference

Simple REST API with JSON request/response.

POST /api/v1/risk/score
cURL
curl -X POST "https://api.cityofhats.com/api/v1/risk/score" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "email": "user@example.com"
  }'
Response 200 OK
{
  "email": "user@example.com",
  "risk_score": 84,
  "confidence": "high",
  "threat_stage": "TESTED",
  "exposed": true,
  "exposure_count": 8,
  "last_seen": "2025-01-02",
  "risk_reasons": [
    "recent credential exposure",
    "observed on criminal trade channels",
    "suspected password reuse",
    "active testing detected"
  ],
  "recommended_action": "step_up_auth",
  "model_version": "2.1"
}

Response Field Reference

Field Type Description
risk_score integer 0–100 risk score (higher = more risky)
confidence string low | medium | high β€” model confidence level
threat_stage string CLEAN | LEAKED | REPOSTED | TRADED | TESTED | ATTACKED | MONETIZED
exposed boolean Whether email appears in breach data
exposure_count integer Number of breach/exposure sources
last_seen string ISO 8601 date of most recent exposure
risk_reasons array Human-readable explainable risk factors
recommended_action string allow | step_up_auth | block | review
model_version string Risk model version for audit/compliance

Why Risk Score Instead of Binary Exposure?

Traditional Tools Answer:

"Breached? Yes / No."

❌ false positives ❌ user impact ❌ alert fatigue ❌ no context
VS

City of Hats Answers:

"How likely is harm?"

βœ… confidence βœ… automation βœ… prevention βœ… explainability
🧠

Model Integrity & Explainability

Every score includes explainable reason codes, so:

πŸ›‘οΈ SOC teams understand
πŸ“‹ Auditors trust the model
βš™οΈ Product teams tune flows
πŸ“Š Fraud analysts optimize controls

Built For Modern Security Stacks

IAM / Authentication
SIEM / SOAR
Fraud & Trust Platforms
Customer Screening
Risk-Based Policy Engines

Ideal For

πŸ›‘οΈ Account protection
🚫 Fraud prevention
πŸ†” Identity verification
βš–οΈ Trust & safety
πŸ’³ High-risk transactions
πŸ” Preference-based MFA

πŸ” Privacy & Compliance

Your platform operates with confidence.

  • Never stores passwords
  • Never resells identity data
  • Anonymizes telemetry
  • Honors enterprise governance
  • SOC 2 Type II compliant
  • GDPR / PDPA aligned

⚑ Performance

Built for real-time decisioning at scale.

  • Real-time lookups
  • Enterprise-scale throughput
  • Global delivery
  • 99.9% uptime SLA
  • <200ms p95 latency
  • Unlimited lookups (Enterprise)

The Risk Scoring API converts raw exposure intelligence into a single, reliable decision signal β€” so your platform can reduce friction for good users, and stop attackers early.

Start Using Risk Scoring API

Transform exposure signals into actionable risk scores. Free tier available β€” no credit card required.

AI-Powered Explainable 99.9% Uptime Enterprise SLA