Skip to main content

​
Overview

Rogue automatically maps red team findings to multiple industry compliance frameworks. This enables organizations to understand their security posture in the context of established standards and regulations.

​
Supported Frameworks

The OWASP Top 10 for LLM Applications 2025 covers the most critical security risks in LLM systems.
CategoryMapped Vulnerabilities
LLM01: Prompt Injectionprompt-extraction, prompt-override, indirect-injection, ascii-smuggling
LLM02: Sensitive Info Disclosurepii-direct, pii-api-db, pii-session, pii-social, cross-session-leakage
LLM03: Supply Chainip-violations
LLM04: Data & Model Poisoningmemory-poisoning, rag-poisoning
LLM05: Improper Output Handlingsql-injection, shell-injection, malicious-code
LLM06: Excessive Agencyexcessive-agency, rbac, bola, bfla
LLM07: System Prompt Leakageprompt-extraction
LLM08: Vector & Embedding Weaknessesrag-exfiltration
LLM09: Misinformationhallucination, unverifiable-claims, misinformation-disinformation
LLM10: Unbounded Consumptionunbounded-consumption, reasoning-dos, divergent-repetition
The MITRE Adversarial Threat Landscape for AI Systems provides a comprehensive taxonomy of AI-specific attacks.
Attack CategoryMapped Vulnerabilities
Prompt/Input Attacksascii-smuggling, prompt-extraction, prompt-override, indirect-injection
Privacy Attacksprivacy-violation, pii-api-db, pii-direct, pii-session, pii-social
System Compromiseexcessive-agency, ip-violations
Content Generationhate-speech, harassment, child-exploitation, cybercrime, extremism
The NIST AI Risk Management Framework addresses governance, technical, and ethical considerations.
Risk DomainMapped Vulnerabilities
Excessive Agencyexcessive-agency
Information Integritymisinformation-disinformation
Privacyprivacy-violation, pii-direct, pii-api-db, pii-session, pii-social
Safetywmd-content, weapons-content, dangerous-activity, cybercrime
Technical Securityshell-injection, sql-injection, bfla, bola, rbac
Content Safetyharassment, hate-speech, personal-attacks
The ISO/IEC 42001 AI Management System standard for organizational AI governance.
DomainMapped Vulnerabilities
Governanceexcessive-agency, overreliance
Bias & Non-Discriminationbias-disability, bias-gender, bias-age, bias-race, bias-political, bias-religion
Privacyprivacy-violation, pii-direct, pii-api-db, pii-session, pii-social
Securityascii-smuggling, prompt-extraction, shell-injection, sql-injection
Information Qualityhallucination, unverifiable-claims
The European Union Artificial Intelligence Act regulatory framework.
RequirementMapped Vulnerabilities
High-Risk Requirementsexcessive-agency, misinformation-disinformation, overreliance
Privacy (GDPR Alignment)pii-direct, pii-session, privacy-violation, pii-api-db
Technical Safetyshell-injection, sql-injection, ssrf
Non-Discriminationhate-speech
Information Qualityhallucination
The General Data Protection Regulation for data protection and privacy.
ArticleMapped Vulnerabilities
Data Protectionprivacy-violation, pii-api-db, pii-direct, pii-session, pii-social, cross-session-leakage
Non-Discriminationbias-disability, bias-gender, bias-age, bias-race, hate-speech
Access Controlrbac, bola, bfla
Securityprompt-extraction, shell-injection, sql-injection, debug-access, ssrf
The OWASP API Security Top 10 for API-related vulnerabilities.
CategoryMapped Vulnerabilities
API1: Broken Object Level Authorizationbola
API2: Broken Authenticationrbac
API3: Broken Object Property Level Authorizationbfla
API4: Unrestricted Resource Consumptionunbounded-consumption
API5: Broken Function Level Authorizationexcessive-agency
API7: Server Side Request Forgeryssrf
API8: Security Misconfigurationdebug-access
A minimal security baseline for essential checks.
CategoryVulnerabilities
Prompt Securityprompt-extraction, prompt-override
PII Protectionpii-direct
Technicalsql-injection, shell-injection
Agencyexcessive-agency

​
Compliance Scoring

Rogue calculates compliance scores for each framework based on tested vulnerabilities: 
# Compliance calculation
tested_vulns = [v for v in framework.vulnerabilities if v in results]
passed_vulns = [v for v in tested_vulns if results[v].passed]

compliance_score = (len(passed_vulns) / len(tested_vulns)) * 100

​
Score Interpretation

Score RangeStatusMeaning
80-100%✅ ExcellentStrong security posture
60-79%⚠️ GoodSome vulnerabilities need attention
0-59%❌ PoorSignificant security gaps

​
Framework Coverage Cards

Rogue generates coverage cards showing compliance status: 
{
  "framework_id": "owasp-llm",
  "framework_name": "OWASP LLM Top 10",
  "compliance_score": 75.0,
  "tested_count": 8,
  "total_count": 25,
  "passed_count": 6,
  "status": "good"
}

​
Default Framework Selection

Frameworks are automatically selected based on scan type:
Scan TypeDefault Frameworks
Basicbasic-security
Fullowasp-llm, mitre-atlas, basic-security
CustomUser-specified

​
Using Frameworks

​
In Configuration

from rogue.server.red_teaming import RedTeamConfig, ScanType

config = RedTeamConfig(
    scan_type=ScanType.CUSTOM,
    vulnerabilities=["prompt-extraction", "pii-direct", "excessive-agency"],
    attacks=["base64", "roleplay"],
    frameworks=["owasp-llm", "gdpr", "eu-ai-act"]
)

​
Accessing Framework Data

from rogue.server.red_teaming.catalog.framework_mappings import (
    get_framework,
    get_all_frameworks,
    get_vulnerabilities_for_framework,
    get_frameworks_for_vulnerability
)

# Get OWASP LLM framework
owasp = get_framework("owasp-llm")
print(f"{owasp.name}: {len(owasp.vulnerabilities)} vulnerabilities")

# Find which frameworks cover a specific vulnerability
frameworks = get_frameworks_for_vulnerability("prompt-extraction")
# Returns: ["owasp-llm", "mitre-atlas", "iso-42001", "gdpr", ...]

​
Report Generation

Compliance reports include:
  1. Compliance Score: Overall percentage for each framework
  2. Vulnerability Breakdown: Per-vulnerability pass/fail status
  3. Recommendations: Prioritized remediation guidance
  4. Framework Mapping: Which controls are affected
## Framework Compliance

### ✅ OWASP LLM Top 10

**Compliance Score:** 85.0%
**Tested:** 17 / 25

**Recommendations:**

- [HIGH] System Prompt Disclosure: Implement prompt guards
- [MEDIUM] Hallucination: Add output validation

​
Adding Custom Frameworks

Custom frameworks can be defined with vulnerability mappings: 
from rogue.server.red_teaming.models import FrameworkDef

custom_framework = FrameworkDef(
    id="custom-security",
    name="Custom Security Framework",
    description="Organization-specific security requirements",
    vulnerabilities=[
        "prompt-extraction",
        "pii-direct",
        "excessive-agency",
        "sql-injection"
    ]
)