Skip to content
Back to skills

2509 Risk Scoring B916321e

ASecurity

Combine multiple factors to calculate a composite risk score (0-100):

  • 9 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 11, 2026
securityrustsqltestingapidatabasesecuritydocumentation

Works with

  • api

Security analysis

A100/100

Scanned October 11, 2026

npx -y skills add tools-only/X-Skills --skill 2509-risk_scoring_b916321e --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of 2509 Risk Scoring B916321e?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for 2509 Risk Scoring B916321e
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tools-only-2509-risk-scoring-b916321e/badge)](https://www.skillsdirectory.com/skills/tools-only-2509-risk-scoring-b916321e)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

SKILL.md
# Risk Scoring Methodology

## Risk Score Calculation

Combine multiple factors to calculate a composite risk score (0-100):

```
Risk Score = (Severity × 0.35) + (Reachability × 0.30) + (Exploitability × 0.20) + (Dependency Criticality × 0.15)
```

### Factor 1: Severity Score (0-100)

Based on CVSS score or severity rating:

| CVSS Score | Severity | Score |
|------------|----------|-------|
| 9.0-10.0   | Critical | 100   |
| 7.0-8.9    | High     | 75    |
| 4.0-6.9    | Medium   | 50    |
| 0.1-3.9    | Low      | 25    |
| 0.0        | None     | 0     |

**Adjustments:**
- If CVE has known exploit: +10 points
- If CVE is trending/actively exploited: +15 points

### Factor 2: Reachability Score (0-100)

Based on whether vulnerable code is actually used:

| Reachability Status | Score | Description |
|---------------------|-------|-------------|
| Direct call         | 100   | Vulnerable function is directly called in application code |
| Indirect call       | 75    | Vulnerable function is called through dependency chain |
| Imported but unused | 25    | Vulnerable code is imported but not executed |
| Not reachable       | 0     | Vulnerable code path is never executed |
| Unknown             | 50    | Reachability analysis not available (assume moderate risk) |

**Detection methods:**
- Static analysis (call graph analysis)
- Dynamic analysis (runtime tracing)
- Dependency tree analysis

### Factor 3: Exploitability Score (0-100)

Based on exploit availability and complexity:

| Exploit Status | Score | Description |
|----------------|-------|-------------|
| Active exploitation in wild | 100 | CVE is being actively exploited |
| Public exploit available | 80 | Exploit code is publicly available (Metasploit, ExploitDB) |
| Proof of concept exists | 60 | PoC code exists but not weaponized |
| Theoretical | 30 | No known exploit, but theoretically exploitable |
| Not exploitable | 0 | Requires physical access or impractical conditions |

**Data sources:**
- CISA KEV (Known Exploited Vulnerabilities)
- Exploit databases (ExploitDB, Metasploit)
- Threat intelligence feeds
- CVE references and advisories

### Factor 4: Dependency Criticality Score (0-100)

Based on how critical the affected dependency is:

| Criticality Level | Score | Description |
|-------------------|-------|-------------|
| Core/Critical     | 100   | Authentication, authorization, data validation, crypto |
| High              | 75    | Database drivers, API frameworks, payment processing |
| Medium            | 50    | Logging, monitoring, utilities |
| Low               | 25    | Development tools, testing frameworks |
| Minimal           | 10    | Documentation generators, linters (dev-only) |

**Indicators of criticality:**
- Handles sensitive data (passwords, PII, financial)
- Exposed to untrusted input (user input, network data)
- Core business logic dependency
- Security-critical functionality
- Production vs development dependency

## Risk Tiers

Based on final risk score:

| Risk Tier | Score Range | Priority | Typical Action |
|-----------|-------------|----------|----------------|
| Critical  | 80-100      | P0       | Immediate action required |
| High      | 60-79       | P1       | Address within days |
| Medium    | 40-59       | P2       | Address within weeks |
| Low       | 20-39       | P3       | Address in next cycle |
| Minimal   | 0-19        | P4       | Monitor or ignore |

## Example Calculations

### Example 1: Critical Risk CVE

```
CVE-2024-1234: SQL Injection in database driver
- Severity: Critical (CVSS 9.8) → 100 points
- Reachability: Direct call → 100 points
- Exploitability: Public exploit available → 80 points
- Dependency Criticality: Core (database driver) → 100 points

Risk Score = (100 × 0.35) + (100 × 0.30) + (80 × 0.20) + (100 × 0.15)
           = 35 + 30 + 16 + 15
           = 96 (Critical Risk)
```

**Recommendation**: Immediate upgrade required

### Example 2: Low Risk CVE

```
CVE-2024-5678: XSS in development-only documentation tool
- Severity: Medium (CVSS 5.4) → 50 points
- Reachability: Not reachable (dev dependency) → 0 points
- Exploitability: Theoretical → 30 points
- Dependency Criticality: Minimal (dev tool) → 10 points

Risk Score = (50 × 0.35) + (0 × 0.30) + (30 × 0.20) + (10 × 0.15)
           = 17.5 + 0 + 6 + 1.5
           = 25 (Low Risk)
```

**Recommendation**: Ignore with justification (dev-only, not reachable)

### Example 3: Medium Risk CVE

```
CVE-2024-9012: Denial of Service in logging library
- Severity: High (CVSS 7.5) → 75 points
- Reachability: Indirect call → 75 points
- Exploitability: Proof of concept exists → 60 points
- Dependency Criticality: Medium (logging) → 50 points

Risk Score = (75 × 0.35) + (75 × 0.30) + (60 × 0.20) + (50 × 0.15)
           = 26.25 + 22.5 + 12 + 7.5
           = 68.25 (High Risk)
```

**Recommendation**: Upgrade within days, or apply rate limiting mitigation

## Special Considerations

### Zero-Day CVEs
- Treat as Critical (score ≥ 80) if reachable
- Immediate action even if no exploit available yet

### Transitive Dependencies
- Score based on actual usage, not just presence
- Consider upgrade path complexity

### False Positives
- Reduce score by 20-30 points if confirmed false positive
- Document justification in report

### Compensating Controls
- Reduce score by 10-20 points if effective mitigations exist
- Examples: WAF rules, network segmentation, input validation

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.