Risk Analyzer

This security layer acts as the "brain" analyzing suspicious behavior and the "filter" preventing data-level attacks.

1. Risk Analyzer (Behavioral Analysis)

File: hierachain/risk_management/risk_analyzer.py

Risk analysis system based on operational data:

• Anomaly Detection: Uses Z-score algorithm to detect sudden changes in request frequency or error rates.
• Threat Intelligence: Identifies common attack patterns (such as Brute force API key, Replay attack).
• Scoring System: Evaluates Risk Score for each Peer or API Key to apply corresponding countermeasures.

2. Input Sanitization & Validation

File: hierachain/security/sanitization.py

Defense layer against data-level attacks:

• Injection Protection: Sanitizes input data to prevent SQL Injection, NoSQL Injection, and Command Injection.
• Nested Bomb Protection: Limits JSON payload depth to prevent denial-of-service attacks through complex nested data structures.
• Type Strictness: Ensures input data fully matches the defined schema, rejecting any redundant or malformed fields.

Risk Response Matrix

The system automatically executes actions based on risk score:

Sanitization Flow

graph LR
    A[Raw Request Body] --> B[Size Limit Check]
    B --> C[JSON Schema Validation]
    C --> D[Recursive Type Checking]
    D --> E[Injection Pattern Scan]
    E --> F[Sanitized Data for Logic]

Related

• Risk management
• Alert system