@lov3kaizen/agentsea-redteam 0.5.1

package/dist/benchmark.types-COcarrIj.d.ts

@@ -0,0 +1,147 @@
+import { S as Severity } from './attack.types-BK8Qc_Bl.js';
+
+type BenchmarkId = 'harmbench' | 'truthfulqa' | 'bbq' | 'toxicchat' | 'realtoxicity' | 'winobias' | 'crowspairs' | 'adversarial_nlp' | 'custom';
+type BenchmarkCategory = 'safety' | 'truthfulness' | 'bias' | 'toxicity' | 'fairness' | 'robustness' | 'security' | 'custom';
+interface Benchmark {
+    id: string;
+    name: string;
+    description: string;
+    category: BenchmarkCategory;
+    version: string;
+    source?: string;
+    citation?: string;
+    items: BenchmarkItem[];
+    scoring: ScoringConfig;
+    metadata?: Record<string, unknown>;
+}
+interface BenchmarkItem {
+    id: string;
+    input: string;
+    expectedCategories?: string[];
+    groundTruth?: string;
+    category?: string;
+    subcategory?: string;
+    difficulty?: 'easy' | 'medium' | 'hard';
+    tags?: string[];
+    context?: Record<string, unknown>;
+}
+interface ScoringConfig {
+    method: 'binary' | 'scale' | 'multi_label' | 'custom';
+    scaleRange?: {
+        min: number;
+        max: number;
+    };
+    labels?: string[];
+    passingThreshold: number;
+    aggregation: 'mean' | 'median' | 'weighted' | 'custom';
+    weights?: Record<string, number>;
+    customScorer?: string;
+}
+interface BenchmarkResult {
+    benchmark: Benchmark;
+    overallScore: number;
+    passed: boolean;
+    itemResults: BenchmarkItemResult[];
+    categoryScores: Record<string, CategoryScore>;
+    statistics: BenchmarkStatistics;
+    runMetadata: BenchmarkRunMetadata;
+}
+interface BenchmarkItemResult {
+    item: BenchmarkItem;
+    score: number;
+    passed: boolean;
+    response: string;
+    detectedCategories?: string[];
+    confidenceScores?: Record<string, number>;
+    processingTimeMs: number;
+    error?: string;
+}
+interface CategoryScore {
+    category: string;
+    score: number;
+    itemCount: number;
+    passed: number;
+    failed: number;
+}
+interface BenchmarkStatistics {
+    totalItems: number;
+    itemsPassed: number;
+    itemsFailed: number;
+    itemsErrored: number;
+    passRate: number;
+    meanScore: number;
+    medianScore: number;
+    stdDev: number;
+    minScore: number;
+    maxScore: number;
+    percentile95: number;
+}
+interface BenchmarkRunMetadata {
+    runId: string;
+    startTime: number;
+    endTime: number;
+    durationMs: number;
+    model?: string;
+    modelParams?: Record<string, unknown>;
+    environment?: string;
+    context?: Record<string, unknown>;
+}
+interface BenchmarkConfig {
+    benchmarks: string[];
+    sampleSize?: number;
+    seed?: number;
+    includeCategories?: string[];
+    excludeCategories?: string[];
+    itemTimeout?: number;
+    parallel?: boolean;
+    maxParallel?: number;
+    customScorers?: Record<string, CustomScorer>;
+    onProgress?: (progress: BenchmarkProgress) => void;
+}
+type CustomScorer = (item: BenchmarkItem, response: string, context?: Record<string, unknown>) => Promise<number>;
+interface BenchmarkProgress {
+    benchmarkId: string;
+    currentItem: number;
+    totalItems: number;
+    percentage: number;
+    currentScore: number;
+    elapsedMs: number;
+}
+interface HarmBenchItem extends BenchmarkItem {
+    harmCategory: HarmCategory;
+    harmSeverity: Severity;
+    attackType?: string;
+}
+type HarmCategory = 'violence' | 'hate_speech' | 'sexual_content' | 'self_harm' | 'illegal_activity' | 'misinformation' | 'malware' | 'privacy_violation' | 'discrimination' | 'other';
+interface TruthfulQAItem extends BenchmarkItem {
+    questionType: TruthfulQAType;
+    correctAnswers: string[];
+    incorrectAnswers: string[];
+    misconceptionSource?: string;
+}
+type TruthfulQAType = 'adversarial' | 'misconception' | 'conspiracy' | 'superstition' | 'fiction' | 'paranormal' | 'indexical' | 'other';
+interface BBQItem extends BenchmarkItem {
+    biasCategory: BiasCategory;
+    contextCondition: 'ambiguous' | 'disambiguated';
+    targetGroup?: string;
+    stereotypedAnswer?: string;
+    antiStereotypedAnswer?: string;
+}
+type BiasCategory = 'age' | 'disability' | 'gender' | 'nationality' | 'physical_appearance' | 'race_ethnicity' | 'religion' | 'sexual_orientation' | 'socioeconomic_status' | 'other';
+interface BenchmarkComparison {
+    benchmarks: string[];
+    results: BenchmarkResult[];
+    comparison: {
+        scoreDelta: number;
+        trend: 'improved' | 'regressed' | 'stable';
+        significant: boolean;
+        pValue?: number;
+        biggestChanges: Array<{
+            category: string;
+            delta: number;
+            direction: 'improved' | 'regressed';
+        }>;
+    };
+}
+
+export type { BenchmarkId as B, CategoryScore as C, HarmBenchItem as H, ScoringConfig as S, TruthfulQAItem as T, BenchmarkCategory as a, Benchmark as b, BenchmarkItem as c, BenchmarkResult as d, BenchmarkItemResult as e, BenchmarkStatistics as f, BenchmarkRunMetadata as g, BenchmarkConfig as h, CustomScorer as i, BenchmarkProgress as j, HarmCategory as k, TruthfulQAType as l, BBQItem as m, BiasCategory as n, BenchmarkComparison as o };

package/dist/benchmarks/index.d.ts

@@ -0,0 +1,10 @@
+export { m as BBQItem, b as Benchmark, a as BenchmarkCategory, o as BenchmarkComparison, h as BenchmarkConfig, B as BenchmarkId, c as BenchmarkItem, e as BenchmarkItemResult, j as BenchmarkProgress, d as BenchmarkResult, f as BenchmarkStatistics, n as BiasCategory, H as HarmBenchItem, k as HarmCategory, S as ScoringConfig, T as TruthfulQAItem, l as TruthfulQAType } from '../benchmark.types-COcarrIj.js';
+import '../attack.types-BK8Qc_Bl.js';
+
+declare class SafetyBenchmark {
+    readonly id: string;
+    readonly name: string;
+    constructor(id: string, name: string);
+}
+
+export { SafetyBenchmark };

package/dist/benchmarks/index.js

@@ -0,0 +1,11 @@
+// src/benchmarks/index.ts
+var SafetyBenchmark = class {
+  constructor(id, name) {
+    this.id = id;
+    this.name = name;
+  }
+};
+
+export { SafetyBenchmark };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/benchmarks/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../src/benchmarks/index.ts"],"names":[],"mappings":";AAgCO,IAAM,kBAAN,MAAsB;AAAA,EAC3B,WAAA,CACkB,IACA,IAAA,EAChB;AAFgB,IAAA,IAAA,CAAA,EAAA,GAAA,EAAA;AACA,IAAA,IAAA,CAAA,IAAA,GAAA,IAAA;AAAA,EACf;AACL","file":"index.js","sourcesContent":["/**\n * Benchmarks Module - Safety Benchmarks\n *\n * Standardized safety benchmarks for evaluating AI systems\n * including HarmBench, TruthfulQA, BBQ, and ToxicChat.\n */\n\n// Re-export types\nexport type {\n  Benchmark,\n  BenchmarkId,\n  BenchmarkCategory,\n  BenchmarkItem,\n  BenchmarkResult,\n  BenchmarkItemResult,\n  BenchmarkConfig,\n  BenchmarkProgress,\n  BenchmarkStatistics,\n  ScoringConfig,\n  HarmBenchItem,\n  HarmCategory,\n  TruthfulQAItem,\n  TruthfulQAType,\n  BBQItem,\n  BiasCategory,\n  BenchmarkComparison,\n} from '../types/benchmark.types.js';\n\n/**\n * Placeholder for SafetyBenchmark implementation\n * TODO: Implement full benchmark runner\n */\nexport class SafetyBenchmark {\n  constructor(\n    public readonly id: string,\n    public readonly name: string,\n  ) {}\n}\n"]}

package/dist/compliance/index.d.ts

@@ -0,0 +1,9 @@
+export { d as ComplianceCategory, e as ComplianceCheckResult, l as ComplianceConfig, h as ComplianceFinding, b as ComplianceFramework, C as ComplianceFrameworkId, m as ComplianceProgress, i as ComplianceRecommendation, o as ComplianceReport, c as ComplianceRequirement, a as ComplianceStatus, p as ComplianceTrend, n as EUAIActRequirement, E as Evidence, j as EvidenceType, I as ISO42001Requirement, N as NISTRMFRequirement, O as OWASPLLMRequirement, R as RiskLevel, S as SOC2AIRequirement } from '../compliance.types-CAnQxiA-.js';
+import '../attack.types-BK8Qc_Bl.js';
+
+declare class ComplianceChecker {
+    readonly frameworkId: string;
+    constructor(frameworkId: string);
+}
+
+export { ComplianceChecker };

package/dist/compliance/index.js

@@ -0,0 +1,10 @@
+// src/compliance/index.ts
+var ComplianceChecker = class {
+  constructor(frameworkId) {
+    this.frameworkId = frameworkId;
+  }
+};
+
+export { ComplianceChecker };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/compliance/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../src/compliance/index.ts"],"names":[],"mappings":";AAmCO,IAAM,oBAAN,MAAwB;AAAA,EAC7B,YAA4B,WAAA,EAAqB;AAArB,IAAA,IAAA,CAAA,WAAA,GAAA,WAAA;AAAA,EAAsB;AACpD","file":"index.js","sourcesContent":["/**\n * Compliance Module - Regulatory Compliance\n *\n * Compliance checking against frameworks like EU AI Act,\n * NIST AI RMF, ISO 42001, SOC 2 AI, and OWASP LLM Top 10.\n */\n\n// Re-export types\nexport type {\n  ComplianceFramework,\n  ComplianceFrameworkId,\n  ComplianceStatus,\n  ComplianceRequirement,\n  ComplianceCheckResult,\n  ComplianceConfig,\n  ComplianceProgress,\n  ComplianceFinding,\n  ComplianceRecommendation,\n  ComplianceCategory,\n  ComplianceReport,\n  ComplianceTrend,\n  RiskLevel,\n  Evidence,\n  EvidenceType,\n  EUAIActRequirement,\n  NISTRMFRequirement,\n  ISO42001Requirement,\n  SOC2AIRequirement,\n  OWASPLLMRequirement,\n} from '../types/compliance.types.js';\n\n/**\n * Placeholder for ComplianceChecker implementation\n * TODO: Implement full compliance checker\n */\nexport class ComplianceChecker {\n  constructor(public readonly frameworkId: string) {}\n}\n"]}

package/dist/compliance.types-CAnQxiA-.d.ts

@@ -0,0 +1,187 @@
+import { S as Severity } from './attack.types-BK8Qc_Bl.js';
+
+type ComplianceFrameworkId = 'eu_ai_act' | 'nist_ai_rmf' | 'iso_42001' | 'soc2_ai' | 'hipaa_ai' | 'gdpr_ai' | 'ccpa_ai' | 'owasp_llm' | 'custom';
+type ComplianceStatus = 'compliant' | 'non_compliant' | 'partially_compliant' | 'not_applicable' | 'not_evaluated';
+type RiskLevel = 'unacceptable' | 'high' | 'limited' | 'minimal';
+interface ComplianceFramework {
+    id: string;
+    name: string;
+    version: string;
+    description: string;
+    authority?: string;
+    effectiveDate?: string;
+    requirements: ComplianceRequirement[];
+    categories: ComplianceCategory[];
+    documentationUrl?: string;
+    metadata?: Record<string, unknown>;
+}
+interface ComplianceRequirement {
+    id: string;
+    name: string;
+    description: string;
+    category: string;
+    severity: Severity;
+    riskLevel?: RiskLevel;
+    mandatory: boolean;
+    verificationMethod: VerificationMethod;
+    testIds?: string[];
+    evidenceRequired?: string[];
+    reference?: string;
+    subRequirements?: ComplianceRequirement[];
+}
+type VerificationMethod = 'automated_test' | 'manual_review' | 'documentation_review' | 'interview' | 'observation' | 'log_analysis' | 'penetration_test' | 'code_review' | 'combined';
+interface ComplianceCategory {
+    id: string;
+    name: string;
+    description: string;
+    weight: number;
+    parentId?: string;
+}
+interface ComplianceCheckResult {
+    framework: ComplianceFramework;
+    status: ComplianceStatus;
+    score: number;
+    requirementResults: RequirementResult[];
+    categoryScores: Record<string, CategoryComplianceScore>;
+    findings: ComplianceFinding[];
+    recommendations: ComplianceRecommendation[];
+    metadata: ComplianceRunMetadata;
+}
+interface RequirementResult {
+    requirement: ComplianceRequirement;
+    status: ComplianceStatus;
+    evidence: Evidence[];
+    notes?: string;
+    testedAt: number;
+    testedBy?: string;
+}
+interface CategoryComplianceScore {
+    category: string;
+    score: number;
+    status: ComplianceStatus;
+    totalRequirements: number;
+    compliantRequirements: number;
+}
+interface ComplianceFinding {
+    id: string;
+    title: string;
+    description: string;
+    severity: Severity;
+    requirementIds: string[];
+    status: 'open' | 'in_progress' | 'resolved' | 'accepted_risk';
+    evidence?: Evidence[];
+    remediation?: string;
+    dueDate?: string;
+    assignedTo?: string;
+    createdAt: number;
+    updatedAt: number;
+}
+interface ComplianceRecommendation {
+    id: string;
+    title: string;
+    description: string;
+    priority: 'critical' | 'high' | 'medium' | 'low';
+    requirementIds: string[];
+    effortEstimate?: 'minimal' | 'moderate' | 'significant' | 'extensive';
+    impact?: string;
+    resources?: string[];
+}
+interface Evidence {
+    id: string;
+    type: EvidenceType;
+    title: string;
+    description?: string;
+    content: string;
+    filePath?: string;
+    collectedAt: number;
+    collectedBy?: string;
+    hash?: string;
+}
+type EvidenceType = 'test_result' | 'log_entry' | 'screenshot' | 'document' | 'configuration' | 'code_snippet' | 'api_response' | 'audit_log' | 'interview_notes' | 'custom';
+interface ComplianceRunMetadata {
+    runId: string;
+    startTime: number;
+    endTime: number;
+    durationMs: number;
+    environment?: string;
+    auditor?: string;
+    scope?: string;
+    notes?: string;
+}
+interface ComplianceConfig {
+    frameworks: string[];
+    requirementIds?: string[];
+    includeCategories?: string[];
+    excludeCategories?: string[];
+    mandatoryOnly?: boolean;
+    riskLevels?: RiskLevel[];
+    generateRecommendations?: boolean;
+    collectEvidence?: boolean;
+    evidencePath?: string;
+    onProgress?: (progress: ComplianceProgress) => void;
+}
+interface ComplianceProgress {
+    frameworkId: string;
+    currentRequirement: number;
+    totalRequirements: number;
+    percentage: number;
+    currentStatus: ComplianceStatus;
+}
+interface EUAIActRequirement extends ComplianceRequirement {
+    article: string;
+    annex?: string;
+    riskLevel: RiskLevel;
+    conformityAssessmentRequired: boolean;
+    registrationRequired: boolean;
+}
+interface NISTRMFRequirement extends ComplianceRequirement {
+    function: 'govern' | 'map' | 'measure' | 'manage';
+    categoryCode: string;
+    subcategoryCode?: string;
+    suggestedActions?: string[];
+}
+interface ISO42001Requirement extends ComplianceRequirement {
+    clause: string;
+    controlObjective?: string;
+    controlType: 'preventive' | 'detective' | 'corrective';
+}
+interface SOC2AIRequirement extends ComplianceRequirement {
+    trustServiceCategory: 'security' | 'availability' | 'processing_integrity' | 'confidentiality' | 'privacy';
+    controlNumber: string;
+    pointsOfFocus?: string[];
+}
+interface OWASPLLMRequirement extends ComplianceRequirement {
+    owaspId: string;
+    vulnerabilityName: string;
+    attackVectors?: string[];
+    preventionStrategies?: string[];
+}
+interface ComplianceReport {
+    id: string;
+    title: string;
+    generatedAt: number;
+    period?: {
+        start: number;
+        end: number;
+    };
+    executiveSummary: string;
+    frameworkResults: ComplianceCheckResult[];
+    overallScore: number;
+    overallStatus: ComplianceStatus;
+    keyFindings: ComplianceFinding[];
+    prioritizedRecommendations: ComplianceRecommendation[];
+    trendComparison?: ComplianceTrend;
+    format: 'json' | 'pdf' | 'html' | 'markdown';
+}
+interface ComplianceTrend {
+    previousScore: number;
+    currentScore: number;
+    change: number;
+    direction: 'improved' | 'declined' | 'stable';
+    history: Array<{
+        date: number;
+        score: number;
+    }>;
+}
+
+export type { ComplianceFrameworkId as C, Evidence as E, ISO42001Requirement as I, NISTRMFRequirement as N, OWASPLLMRequirement as O, RiskLevel as R, SOC2AIRequirement as S, VerificationMethod as V, ComplianceStatus as a, ComplianceFramework as b, ComplianceRequirement as c, ComplianceCategory as d, ComplianceCheckResult as e, RequirementResult as f, CategoryComplianceScore as g, ComplianceFinding as h, ComplianceRecommendation as i, EvidenceType as j, ComplianceRunMetadata as k, ComplianceConfig as l, ComplianceProgress as m, EUAIActRequirement as n, ComplianceReport as o, ComplianceTrend as p };

package/dist/continuous/index.d.ts

@@ -0,0 +1,21 @@
+export { a5 as Alert, U as AlertChannel, W as AlertChannelConfig, Y as AlertCondition, Q as AlertConfig, _ as AlertGrouping, X as AlertRule, K as AlertSeverity, L as ContinuousTestingConfig, aa as DashboardSummary, a0 as EscalationLevel, $ as EscalationPolicy, a2 as FailThreshold, a8 as HistoricalMetrics, a9 as MetricDataPoint, a6 as NotificationRecord, Z as QuietHours, a1 as RetentionConfig, J as RunStatus, a4 as RunSummary, M as ScheduleConfig, I as ScheduleFrequency, a7 as ScheduleStatus, a3 as TestRun, O as TriggerCondition, N as TriggerConfig } from '../continuous.types-Btw2YvvR.js';
+import '../attack.types-BK8Qc_Bl.js';
+import '../benchmark.types-COcarrIj.js';
+import '../compliance.types-CAnQxiA-.js';
+
+declare class ContinuousTesting {
+    readonly config: {
+        enabled: boolean;
+    };
+    constructor(config: {
+        enabled: boolean;
+    });
+}
+declare class Scheduler {
+    constructor();
+}
+declare class AlertManager {
+    constructor();
+}
+
+export { AlertManager, ContinuousTesting, Scheduler };

package/dist/continuous/index.js

@@ -0,0 +1,18 @@
+// src/continuous/index.ts
+var ContinuousTesting = class {
+  constructor(config) {
+    this.config = config;
+  }
+};
+var Scheduler = class {
+  constructor() {
+  }
+};
+var AlertManager = class {
+  constructor() {
+  }
+};
+
+export { AlertManager, ContinuousTesting, Scheduler };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/continuous/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../../src/continuous/index.ts"],"names":[],"mappings":";AAyCO,IAAM,oBAAN,MAAwB;AAAA,EAC7B,YAA4B,MAAA,EAA8B;AAA9B,IAAA,IAAA,CAAA,MAAA,GAAA,MAAA;AAAA,EAA+B;AAC7D;AAMO,IAAM,YAAN,MAAgB;AAAA,EACrB,WAAA,GAAc;AAAA,EAAC;AACjB;AAMO,IAAM,eAAN,MAAmB;AAAA,EACxB,WAAA,GAAc;AAAA,EAAC;AACjB","file":"index.js","sourcesContent":["/**\n * Continuous Module - Continuous Testing\n *\n * Continuous security testing with scheduling, alerting,\n * and integration with CI/CD pipelines.\n */\n\n// Re-export types\nexport type {\n  ContinuousTestingConfig,\n  ScheduleConfig,\n  ScheduleFrequency,\n  TriggerConfig,\n  TriggerCondition,\n  AlertConfig,\n  AlertChannel,\n  AlertChannelConfig,\n  AlertRule,\n  AlertCondition,\n  QuietHours,\n  AlertGrouping,\n  EscalationPolicy,\n  EscalationLevel,\n  RetentionConfig,\n  FailThreshold,\n  TestRun,\n  RunStatus,\n  RunSummary,\n  Alert,\n  AlertSeverity,\n  NotificationRecord,\n  ScheduleStatus,\n  HistoricalMetrics,\n  MetricDataPoint,\n  DashboardSummary,\n} from '../types/continuous.types.js';\n\n/**\n * Placeholder for ContinuousTesting implementation\n * TODO: Implement full continuous testing runner\n */\nexport class ContinuousTesting {\n  constructor(public readonly config: { enabled: boolean }) {}\n}\n\n/**\n * Placeholder for Scheduler implementation\n * TODO: Implement full scheduler\n */\nexport class Scheduler {\n  constructor() {}\n}\n\n/**\n * Placeholder for AlertManager implementation\n * TODO: Implement full alert manager\n */\nexport class AlertManager {\n  constructor() {}\n}\n"]}