docs(threatscore): add info with Prowler ThreatScore (#8711)

Co-authored-by: Andoni A. <14891798+andoniaf@users.noreply.github.com>
This commit is contained in:
Pedro Martín
2025-09-29 11:17:05 +02:00
committed by GitHub
parent a52697bfdf
commit 7431bab2a7
2 changed files with 487 additions and 0 deletions

View File

@@ -0,0 +1,483 @@
# Prowler ThreatScore Documentation
## Introduction
The **Prowler ThreatScore** is a comprehensive compliance scoring system that provides a unified metric for assessing your organization's security posture across compliance frameworks. It aggregates findings from individual security checks into a single, normalized score ranging from 0 to 100.
### Purpose
- **Unified View**: Get a single metric representing overall compliance health
- **Risk Prioritization**: Understand which areas pose the highest security risks
- **Progress Tracking**: Monitor improvements in compliance posture over time
- **Executive Reporting**: Provide clear, quantifiable security metrics to stakeholders
## How ThreatScore Works
The ThreatScore calculation considers four critical factors for each compliance requirement:
### 1. Pass Rate (`rate_i`)
The percentage of security checks that passed for a specific requirement:
```
Pass Rate = (Number of PASS findings) / (Total findings)
```
### 2. Total Findings (`total_i`)
The total number of checks performed (both PASS and FAIL) for a requirement. This represents the amount of evidence available - more findings provide greater confidence in the assessment.
### 3. Weight (`weight_i`)
A numerical value (1-1000) representing the business importance or criticality of the requirement within your organization's context.
### 4. Risk Level (`risk_i`)
A severity rating (1-5) indicating the potential impact of non-compliance with this requirement.
## Score Interpretation Guidelines
| ThreatScore | Interpretation | Recommended Actions |
|------------------|----------------|-------------------|
| 90-100% | Excellent | Maintain current controls, focus on continuous improvement |
| 80-89% | Good | Address remaining gaps, prepare for compliance audits |
| 70-79% | Acceptable | Prioritize high-risk failures, develop improvement plan |
| 60-69% | Needs Improvement | Immediate attention required, may not pass compliance audit |
| Below 60% | Critical | Emergency response needed, potential regulatory issues |
## Mathematical Formula
The ThreatScore uses a weighted average formula that accounts for all four factors:
```
ThreatScore = (Σ(rate_i × total_i × weight_i × risk_i) / Σ(total_i × weight_i × risk_i)) × 100
```
### Formula Properties
- **Normalization**: Always produces a score between 0 and 100
- **Evidence-weighted**: Requirements with more findings have proportionally greater influence
- **Risk-sensitive**: Higher risk requirements impact the score more significantly
- **Business-aligned**: Weight values allow customization based on organizational priorities
## Parameters Explained
### Weight Values (1-1000)
The weight parameter allows customization of ThreatScore calculation based on organizational priorities and regulatory requirements.
#### Weight Assignment Guidelines
| Weight Range | Priority Level | Use Cases |
|--------------|----------------|-----------|
| 1-100 | Low | Optional or nice-to-have controls |
| 101-300 | Medium | Standard security practices |
| 301-600 | High | Important security controls |
| 601-850 | Critical | Regulatory compliance requirements |
| 851-1000 | Maximum | Mission-critical security controls |
#### Weight Selection Strategy
1. **Regulatory Mapping**: Assign higher weights to controls required by industry regulations
2. **Business Impact**: Consider the potential business impact of control failures
3. **Risk Tolerance**: Align weights with organizational risk appetite
4. **Stakeholder Input**: Involve compliance and business teams in weight decisions
### Risk Levels (1-5)
Risk levels represent the potential security impact of non-compliance with a requirement.
| Risk Level | Severity | Impact Description |
|------------|----------|-------------------|
| 1 | Very Low | Minimal security impact, informational |
| 2 | Low | Limited exposure, low probability of exploitation |
| 3 | Medium | Moderate security risk, potential for limited damage |
| 4 | High | Significant security risk, high probability of impact |
| 5 | Critical | Severe security risk, immediate threat to organization |
#### Risk Level Assessment Criteria
- **Confidentiality Impact**: Data exposure potential
- **Integrity Impact**: Risk of unauthorized data modification
- **Availability Impact**: Service disruption potential
- **Compliance Impact**: Regulatory violation consequences
- **Exploitability**: Ease of exploitation by threat actors
## Security Pillars and Subpillars
Prowler organizes security requirements into a hierarchical structure of pillars and subpillars, providing a comprehensive framework for security assessment and compliance evaluation.
### Security Pillars Overview
The ThreatScore calculation considers requirements organized within the following security pillars:
#### 1. IAM (Identity and Access Management)
**Purpose**: Controls who can access what resources and under what conditions
**Subpillars**:
- **1.1 Authentication**: Verifying user and system identities
- **1.2 Authorization**: Controlling access to resources based on authenticated identity
- **1.3 Privilege Escalation**: Preventing unauthorized elevation of permissions
#### 2. Attack Surface
**Purpose**: Minimizing exposure points that could be exploited by threat actors across network, storage, and application layers
**Subpillars**:
- **2.1 Network**: Network infrastructure security, segmentation, firewall rules, VPC configurations, and traffic controls
- **2.2 Storage**: Data storage systems security, database security, file system permissions, backup security, and storage encryption
- **2.3 Application**: Application-level controls and configurations, application security settings, code security, runtime protections
#### 3. Logging and Monitoring
**Purpose**: Ensuring comprehensive visibility and audit capabilities
**Subpillars**:
- **3.1 Logging**: Capturing security-relevant events and activities
- **3.2 Retention**: Maintaining logs for appropriate time periods
- **3.3 Monitoring**: Active surveillance and alerting on security events
#### 4. Encryption
**Purpose**: Protecting data confidentiality through cryptographic controls
**Subpillars**:
- **4.1 In-Transit**: Encrypting data during transmission
- **4.2 At-Rest**: Encrypting stored data
### Pillar Hierarchy and ThreatScore Impact
#### Hierarchy Structure
```
Security Framework
├── 1. IAM
│ ├── 1.1 Authentication
│ ├── 1.2 Authorization
│ └── 1.3 Privilege Escalation
├── 2. Attack Surface
│ ├── 2.1 Network
│ ├── 2.2 Storage
│ └── 2.3 Application
├── 3. Logging and Monitoring
│ ├── 3.1 Logging
│ ├── 3.2 Retention
│ └── 3.3 Monitoring
└── 4. Encryption
├── 4.1 In-Transit
└── 4.2 At-Rest
Example Requirement Structure:
├── Pillar: 1. IAM
│ ├── Subpillar: 1.1 Authentication
│ │ ├── Requirement: MFA Implementation
│ │ │ ├── Check 1: Admin accounts use MFA
│ │ │ ├── Check 2: Regular users use MFA
│ │ │ └── Check 3: Service accounts use MFA
│ │ └── [Additional Requirements]
│ └── [Additional Subpillars: Authorization, Privilege Escalation]
```
#### Weight and Risk Assignment by Pillar
Different pillars typically receive different weight and risk assignments based on their security impact:
| Pillar | Typical Weight Range | Typical Risk Range | Rationale |
|--------|---------------------|-------------------|-----------|
| 1. IAM | 800-1000 | 4-5 | Critical for access control, high impact if compromised |
| 2. Attack Surface | 500-900 | 3-5 | Highly dependent on exposure and criticality across network, storage, and application layers |
| 3. Logging and Monitoring | 600-800 | 3-4 | Important for detection and compliance, moderate direct impact |
| 4. Encryption | 700-950 | 4-5 | Essential for data protection, regulatory compliance |
**Subpillar Weight Considerations**:
- **2.1 Network (Attack Surface)**: 500-800, Risk 3-4 - Network perimeter defense
- **2.2 Storage (Attack Surface)**: 600-900, Risk 4-5 - Data exposure impact
- **2.3 Application (Attack Surface)**: 400-700, Risk 2-4 - Varies by application criticality
### Pillar-Specific Scoring Considerations
#### High-Impact Pillars (1. IAM, 4. Encryption)
- **Characteristics**: Direct impact on data protection and access control
- **ThreatScore Impact**: Failures in these pillars significantly lower overall score
- **Weight Strategy**: Assign maximum weights (800-1000) to critical requirements
- **Risk Strategy**: Most requirements rated 4-5 due to severe consequences
#### Variable-Impact Pillar (2. Attack Surface)
- **Characteristics**: Impact varies significantly across subpillars (Network, Storage, Application)
- **ThreatScore Impact**: Depends on specific subpillar and business context
- **Weight Strategy**:
- 2.1 Network subpillar: 500-800 (perimeter defense importance)
- 2.2 Storage subpillar: 600-900 (data exposure risk)
- 2.3 Application subpillar: 400-700 (application-specific criticality)
- **Risk Strategy**: Wide range (2-5) based on exposure, data sensitivity, and business criticality
#### Monitoring Pillar (3. Logging and Monitoring)
- **Characteristics**: Essential for compliance and incident response
- **ThreatScore Impact**: Moderate influence, critical for audit requirements
- **Weight Strategy**: Consistent weights (600-800) across logging, retention, and monitoring subpillars
- **Risk Strategy**: Moderate risk levels (3-4) with emphasis on compliance impact
### Cross-Pillar Dependencies
#### Authentication ↔ Authorization (IAM)
- Strong authentication enables effective authorization controls
- Weight both subpillars highly as they're interdependent
#### Logging ↔ Monitoring (Logging and Monitoring)
- Logging provides the data that monitoring systems analyze
- Balance weights to ensure both data collection and analysis are prioritized
#### In-Transit ↔ At-Rest (Encryption)
- Comprehensive data protection requires both encryption types
- Consider data flow patterns when assigning relative weights
### Pillar Coverage in ThreatScore
#### Complete Coverage Benefits
- **Comprehensive Assessment**: All security domains represented in score
- **Balanced View**: Prevents over-emphasis on single security aspect
- **Regulatory Alignment**: Covers requirements across major compliance frameworks
#### Partial Coverage Considerations
- **Focused Assessment**: Target specific security domains
- **Resource Optimization**: Concentrate efforts on high-priority areas
- **Gradual Implementation**: Phase in additional pillars over time
## Scoring Examples
### Example 1: Basic Two-Requirement Scenario
Consider a compliance framework with two requirements:
**Requirement 1: Encryption at Rest**
- Findings: 200 PASS, 500 FAIL (total = 700)
- Pass Rate: 200/700 = 0.286 (28.6%)
- Weight: 500 (High priority - data protection)
- Risk Level: 4 (High risk - data exposure)
**Requirement 2: Access Logging**
- Findings: 300 PASS, 100 FAIL (total = 400)
- Pass Rate: 300/400 = 0.75 (75%)
- Weight: 800 (Critical for audit compliance)
- Risk Level: 3 (Medium risk - audit trail)
**Calculation:**
```
Numerator = (0.286 × 700 × 500 × 4) + (0.75 × 400 × 800 × 3)
= (400,400) + (720,000)
= 1,120,400
Denominator = (700 × 500 × 4) + (400 × 800 × 3)
= 1,400,000 + 960,000
= 2,360,000
ThreatScore = (1,120,400 / 2,360,000) × 100 = 47.5%
```
### Example 2: Enterprise Scenario with Pillar Structure
This example demonstrates how pillar organization affects ThreatScore calculation:
| Pillar | Subpillar | Requirement | Pass | Fail | Total | Weight | Risk | Pass Rate |
|--------|-----------|-------------|------|------|-------|--------|------|-----------|
| 1. IAM | 1.2 Authorization | Access Controls | 280 | 120 | 400 | 800 | 4 | 70% |
| 2. Attack Surface | 2.1 Network | Network Segmentation | 150 | 50 | 200 | 750 | 4 | 75% |
| 2. Attack Surface | 2.2 Storage | Backup Security | 200 | 100 | 300 | 600 | 3 | 66.7% |
| 3. Logging and Monitoring | 3.1 Logging | Audit Logging | 350 | 50 | 400 | 700 | 3 | 87.5% |
| 4. Encryption | 4.2 At-Rest | Encryption | 450 | 50 | 500 | 950 | 5 | 90% |
**Step-by-step Calculation:**
1. **Calculate weighted contributions for each requirement:**
```
Numerator = Σ(rate_i × total_i × weight_i × risk_i)
```
- **Access Controls (1.2 Authorization)**: 0.70 × 400 × 800 × 4 = 896,000
- **Network Segmentation (2.1 Network)**: 0.75 × 200 × 750 × 4 = 450,000
- **Backup Security (2.2 Storage)**: 0.667 × 300 × 600 × 3 = 360,060
- **Audit Logging (3.1 Logging)**: 0.875 × 400 × 700 × 3 = 735,000
- **Encryption (4.2 At-Rest)**: 0.90 × 500 × 950 × 5 = 2,137,500
2. **Sum numerator:** 2,137,500 + 896,000 + 735,000 + 360,060 + 450,000 = **4,578,560**
3. **Calculate total weights for each requirement:**
```
Denominator = Σ(total_i × weight_i × risk_i)
```
- **Access Controls (1.2 Authorization)**: 400 × 800 × 4 = 1,280,000
- **Network Segmentation (2.1 Network)**: 200 × 750 × 4 = 600,000
- **Backup Security (2.2 Storage)**: 300 × 600 × 3 = 540,000
- **Audit Logging (3.1 Logging)**: 400 × 700 × 3 = 840,000
- **Encryption (4.2 At-Rest)**: 500 × 950 × 5 = 2,375,000
4. **Sum denominator:** 2,375,000 + 1,280,000 + 840,000 + 540,000 + 600,000 = **5,635,000**
5. **Final ThreatScore calculation:**
```
ThreatScore = (Numerator / Denominator) × 100
ThreatScore = (4,578,560 / 5,635,000) × 100 = 81.2%
```
**Pillar-Level Analysis:**
- **1. IAM pillar (1.2 Authorization)**: Significant impact despite lower pass rate (70%) due to high weight (800)
- **2. Attack Surface pillar (2.1 Network)**: Strong performance (75%) with high weight (750) balances the score
- **2. Attack Surface pillar (2.2 Storage)**: Lowest performance (66.7%) but limited impact due to moderate weight (600)
- **3. Logging and Monitoring pillar (3.1 Logging)**: Moderate contribution with good performance (87.5%)
- **4. Encryption pillar (4.2 At-Rest)**: Highest contribution due to maximum weight (950) and risk (5)
### Example 3: Multi-Pillar Comprehensive Scenario
| Pillar | Subpillar | Requirement | Pass | Fail | Weight | Risk | Pass Rate |
|--------|-----------|-------------|------|------|--------|------|-----------|
| 1. IAM | 1.1 Authentication | MFA Implementation | 180 | 20 | 900 | 5 | 90% |
| 1. IAM | 1.2 Authorization | Least Privilege Access | 150 | 50 | 850 | 4 | 75% |
| 1. IAM | 1.3 Privilege Escalation | Admin Account Controls | 95 | 5 | 950 | 5 | 95% |
| 2. Attack Surface | 2.1 Network | Firewall Configuration | 400 | 100 | 600 | 3 | 80% |
| 2. Attack Surface | 2.1 Network | Public Endpoint Security | 80 | 20 | 700 | 4 | 80% |
| 2. Attack Surface | 2.2 Storage | Data Classification | 300 | 100 | 650 | 3 | 75% |
| 2. Attack Surface | 2.3 Application | Input Validation | 150 | 50 | 500 | 3 | 75% |
| 3. Logging and Monitoring | 3.1 Logging | Transaction Logging | 500 | 50 | 750 | 3 | 90.9% |
| 3. Logging and Monitoring | 3.3 Monitoring | Real-time Alerts | 200 | 50 | 700 | 4 | 80% |
| 4. Encryption | 4.2 At-Rest | Database Encryption | 300 | 20 | 900 | 5 | 93.8% |
| 4. Encryption | 4.1 In-Transit | API/Web Encryption | 250 | 10 | 800 | 4 | 96.2% |
**Pillar Performance Summary**:
- **1. IAM Pillar Average**: ~87% (weighted by findings across Authentication, Authorization, and Privilege Escalation subpillars)
- **2. Attack Surface Pillar Average**: ~77% (weighted across Network, Storage, and Application subpillars)
- 2.1 Network subpillar: ~80% average
- 2.2 Storage subpillar: 75%
- 2.3 Application subpillar: 75%
- **3. Logging and Monitoring Average**: ~87% (weighted by findings across Logging and Monitoring subpillars)
- **4. Encryption Pillar Average**: ~94% (weighted by findings across In-Transit and At-Rest subpillars)
**Overall ThreatScore**: ~85.3%
This comprehensive example demonstrates how:
- High-performing, high-weight pillars (4. Encryption, 1. IAM) significantly boost the score
- The 2. Attack Surface pillar shows how diverse subpillars (Network, Storage, Application) are aggregated
- Multiple requirements within pillars provide detailed granular assessment
- Cross-pillar balance prevents single points of failure in security posture
### Example 4: Impact of Parameter Changes
Using the scenario, let's see how parameter changes affect the score:
#### Scenario A: Increase Encryption Risk Level
Change Encryption risk from 5 to 3:
- **New ThreatScore: 77.8%** (decrease of 3.4 points)
- **Impact**: Lower risk weighting reduces the influence of high-performing critical controls
#### Scenario B: Improve Access Controls Pass Rate
Change Access Controls from 70% to 90% pass rate:
- **New ThreatScore: 85.1%** (increase of 3.9 points)
- **Impact**: Improving performance on high-weight requirements has significant score impact
#### Scenario C: Add New Low-Weight Requirement
Add "Documentation Completeness" (50 PASS, 10 FAIL, weight=100, risk=1):
- **New ThreatScore: 81.3%** (minimal change of 0.1 points)
- **Impact**: Low-weight requirements have minimal impact on overall score
## Implementation Details
### Edge Cases and Special Conditions
#### Zero Findings Scenario
When a requirement has `total_i = 0` (no findings):
- **Behavior**: Requirement is completely excluded from calculation
- **Rationale**: No evidence means no contribution to confidence in the score
- **Impact**: Other requirements receive proportionally more influence
#### Perfect Score Scenario
When all requirements have 100% pass rate:
- **Result**: ThreatScore = 100%
- **Interpretation**: All implemented security checks are passing
#### Zero Pass Rate Scenario
When all requirements have 0% pass rate:
- **Result**: ThreatScore = 0%
- **Interpretation**: Critical security failures across all requirements
#### Single Requirement Framework
For frameworks with only one requirement:
- **Formula simplification**: ThreatScore = pass_rate × 100
- **Impact**: Weight and risk values become irrelevant for score calculation
### Performance Considerations
#### Computational Complexity
- **Time Complexity**: O(n) where n = number of requirements
- **Space Complexity**: O(1) - constant space for accumulation
- **Scalability**: Efficiently handles frameworks with thousands of requirements
#### Calculation Precision
- **Floating Point**: Use double precision for intermediate calculations
- **Rounding**: Final score rounded to 1 decimal place for display
- **Overflow Protection**: Validate that weight × risk × total values don't exceed system limits
### Data Requirements
#### Minimum Data Set
For each requirement, the following data must be available:
- **pass_count**: Number of PASS findings (integer ≥ 0)
- **fail_count**: Number of FAIL findings (integer ≥ 0)
- **weight**: Business importance (integer 1-1000)
- **risk**: Risk level (integer 1-5)
#### Data Validation Rules
```
total_i = pass_i + fail_i
rate_i = pass_i / total_i (when total_i > 0)
1 ≤ weight_i ≤ 1000
1 ≤ risk_i ≤ 5
```
#### Handling Invalid Data
- **Negative values**: Treat as 0 and log warning
- **Out-of-range weights/risk**: Clamp to valid range and log warning
- **Missing data**: Exclude requirement from calculation and log warning
## Best Practices
### Monitoring and Trending
1. **Establish Baseline**
- Record initial ThreatScore after implementing measurement
- Set realistic improvement targets based on organizational capacity
- Track score changes over time to identify trends
2. **Regular Reporting**
- Generate monthly ThreatScore reports for stakeholders
- Highlight significant score changes and their causes
- Include requirement-level breakdowns for detailed analysis
3. **Continuous Improvement**
- Use score trends to identify systematic issues
- Correlate score changes with security incidents or changes
- Adjust weights and risk levels based on lessons learned

View File

@@ -138,6 +138,10 @@ nav:
- MongoDB Atlas:
- Getting Started: tutorials/mongodbatlas/getting-started-mongodbatlas.md
- Authentication: tutorials/mongodbatlas/authentication.md
- Compliance:
- ThreatScore: tutorials/compliance/threatscore.md
- Developer Guide:
- Concepts:
- Introduction: developer-guide/introduction.md