driftrail 2.0.0__py3-none-any.whl → 2.1.0__py3-none-any.whl

driftrail/types.py

@@ -1,9 +1,11 @@
 """
 DriftRail Type Definitions
+Complete type definitions for all SDK features.
 """
 
-from typing import Optional, List, Dict, Any, Literal
+from typing import Optional, List, Dict, Any, Literal, Union
 from dataclasses import dataclass, field, asdict
+from datetime import datetime
 
 Provider = Literal["openai", "google", "anthropic", "other"]
 
@@ -185,3 +187,642 @@ class GuardBlockedError(Exception):
         self.result = result
         reasons = [t.reason for t in result.triggered]
         super().__init__(f"Content blocked: {'; '.join(reasons)}")
+
+
+# ============ Incident Types ============
+
+@dataclass
+class Incident:
+    incident_id: str
+    title: str
+    severity: Literal["low", "medium", "high", "critical"]
+    status: Literal["open", "investigating", "mitigating", "resolved", "closed"]
+    incident_type: str
+    created_at: str
+    description: Optional[str] = None
+
+
+@dataclass
+class IncidentStats:
+    open: int
+    investigating: int
+    critical: int
+    mttr_hours: float
+
+
+# ============ Compliance Types ============
+
+@dataclass
+class ComplianceFramework:
+    name: str
+    enabled: bool
+    status: str
+
+
+@dataclass
+class ComplianceStatus:
+    frameworks: List[ComplianceFramework]
+    pending_reviews: int
+    recent_violations: int
+    data_region: str
+
+
+@dataclass
+class ComplianceScore:
+    compliance_score: float
+    total_events: int
+    high_risk_events: int
+    high_risk_percentage: float
+    date_range: Dict[str, str]
+    rating: str
+
+
+@dataclass
+class ComplianceReportMeta:
+    report_id: str
+    framework: str
+    date_from: str
+    date_to: str
+    compliance_score: float
+    findings_count: int
+    generated_at: str
+    generated_by: str
+
+
+@dataclass
+class CustomControl:
+    control_id: str
+    name: str
+    description: str
+    category: str
+    evidence_type: str
+    keywords: List[str]
+
+
+# ============ Executive Dashboard Types ============
+
+@dataclass
+class KpiTargets:
+    max_high_risk_percent: float
+    target_latency_ms: int
+    target_sla_percent: float
+    target_mttr_hours: float
+    monthly_budget_usd: float
+    max_critical_events_daily: int
+    target_avg_risk_score: float
+    sla_latency_threshold_ms: int
+    max_open_incidents: int
+    cost_per_1k_tokens: float
+
+
+@dataclass
+class ExecutiveMetrics:
+    period: str
+    health_score: float
+    summary: Dict[str, Any]
+    cost_metrics: Dict[str, Any]
+    incident_metrics: Dict[str, Any]
+    risk_trends: List[Dict[str, Any]]
+    volume_trends: List[Dict[str, Any]]
+    model_breakdown: List[Dict[str, Any]]
+    compliance_status: Dict[str, Any]
+    kpi_status: List[Dict[str, Any]]
+
+
+# ============ Model Analytics Types ============
+
+@dataclass
+class HistoricalLog:
+    event_id: str
+    model: str
+    provider: str
+    environment: str
+    timestamp: str
+    latency_ms: int
+    tokens_in: int
+    tokens_out: int
+    risk_score: float
+
+
+@dataclass
+class ModelSwitch:
+    switch_id: str
+    app_id: str
+    previous_model: Optional[str]
+    new_model: str
+    previous_provider: Optional[str]
+    new_provider: str
+    switch_reason: Optional[str]
+    environment: Optional[str]
+    switched_at: str
+
+
+@dataclass
+class EnvironmentStats:
+    environment: str
+    event_count: int
+    avg_latency_ms: float
+    avg_risk_score: float
+    error_rate: float
+    models: List[str]
+
+
+@dataclass
+class ModelBenchmark:
+    benchmark_id: str
+    model: str
+    environment: Optional[str]
+    avg_latency_ms: float
+    p95_latency_ms: float
+    avg_risk_score: float
+    error_rate: float
+    throughput: float
+    sample_count: int
+    calculated_at: str
+
+
+@dataclass
+class ModelAnalyticsSummary:
+    total_models: int
+    total_events_7d: int
+    avg_latency_ms: float
+    top_model: Optional[str]
+    environments: List[str]
+    recent_switches: int
+
+
+# ============ Drift & Alerts Types ============
+
+@dataclass
+class DriftMetrics:
+    health_score: float
+    current: Dict[str, float]
+    baselines: Dict[str, float]
+    last_updated: str
+
+
+@dataclass
+class DriftAlert:
+    alert_id: str
+    app_id: str
+    model: Optional[str]
+    alert_type: str
+    severity: Literal["critical", "warning", "info"]
+    current_value: float
+    baseline_value: float
+    deviation_percent: float
+    created_at: str
+    acknowledged_at: Optional[str]
+    resolved_at: Optional[str]
+
+
+@dataclass
+class DriftTrend:
+    date: str
+    avg_deviation: float
+    alert_count: int
+    metrics: Dict[str, float]
+
+
+@dataclass
+class DriftBaseline:
+    baseline_id: str
+    app_id: str
+    model: Optional[str]
+    valid_from: str
+    avg_risk_score: Optional[float]
+    sample_count: int
+    risk_distribution: Optional[Dict[str, int]]
+
+
+@dataclass
+class DriftScore:
+    overall_score: float
+    status: Literal["healthy", "warning", "critical"]
+    by_model: List[Dict[str, Any]]
+
+
+# ============ Drift V3 Types ============
+
+@dataclass
+class NotificationChannel:
+    channel_id: str
+    app_id: str
+    channel_type: Literal["email", "slack", "webhook", "pagerduty", "teams"]
+    name: str
+    config: Dict[str, Any]
+    severity_filter: List[str]
+    is_enabled: bool
+    created_at: str
+
+
+@dataclass
+class DriftSegment:
+    segment_id: str
+    app_id: str
+    name: str
+    description: Optional[str]
+    filter_criteria: Dict[str, Any]
+    is_active: bool
+    created_at: str
+
+
+@dataclass
+class CorrelationEvent:
+    correlation_id: str
+    app_id: str
+    primary_metric: str
+    correlated_metric: str
+    correlation_coefficient: float
+    lag_minutes: int
+    detected_at: str
+    is_active: bool
+
+
+@dataclass
+class DistributionSnapshot:
+    snapshot_id: str
+    app_id: str
+    metric_type: str
+    distribution_data: Dict[str, Any]
+    anomaly_score: float
+    is_anomaly: bool
+    captured_at: str
+
+
+@dataclass
+class SeasonalityPattern:
+    pattern_id: str
+    app_id: str
+    metric_type: str
+    day_of_week: int
+    hour_of_day: int
+    expected_value: float
+    std_deviation: float
+    sample_count: int
+
+
+@dataclass
+class BaselineStatistics:
+    app_id: str
+    model: Optional[str]
+    metric_type: str
+    mean: float
+    std_dev: float
+    min: float
+    max: float
+    p50: float
+    p95: float
+    p99: float
+    sample_count: int
+    calculated_at: str
+
+
+# ============ Tracing Types ============
+
+SpanType = Literal["llm", "tool", "retrieval", "chain", "agent", "custom"]
+
+
+@dataclass
+class Trace:
+    trace_id: str
+    app_id: str
+    name: Optional[str]
+    start_time: str
+    end_time: Optional[str]
+    duration_ms: Optional[int]
+    status: Literal["running", "completed", "error"]
+    metadata: Dict[str, Any]
+    tags: List[str]
+    user_id: Optional[str]
+    session_id: Optional[str]
+
+
+@dataclass
+class Span:
+    span_id: str
+    trace_id: str
+    parent_span_id: Optional[str]
+    name: str
+    span_type: SpanType
+    start_time: str
+    end_time: Optional[str]
+    duration_ms: Optional[int]
+    status: Literal["running", "completed", "error"]
+    model: Optional[str]
+    provider: Optional[str]
+    tokens_in: Optional[int]
+    tokens_out: Optional[int]
+    cost_usd: Optional[float]
+    input: Optional[Dict[str, Any]]
+    output: Optional[Dict[str, Any]]
+
+
+# ============ Prompt Management Types ============
+
+@dataclass
+class Prompt:
+    prompt_id: str
+    name: str
+    description: Optional[str]
+    tags: List[str]
+    is_active: bool
+    created_at: str
+    updated_at: str
+
+
+@dataclass
+class PromptVersion:
+    version_id: str
+    prompt_id: str
+    version: int
+    content: str
+    variables: List[str]
+    model_config: Dict[str, Any]
+    is_published: bool
+    is_default: bool
+    commit_message: Optional[str]
+    created_at: str
+
+
+# ============ Evaluation Types ============
+
+@dataclass
+class EvalDataset:
+    dataset_id: str
+    name: str
+    description: Optional[str]
+    schema_type: Literal["qa", "conversation", "classification", "custom"]
+    item_count: int
+    tags: List[str]
+    is_active: bool
+
+
+@dataclass
+class EvalDatasetItem:
+    item_id: str
+    dataset_id: str
+    input: Dict[str, Any]
+    expected_output: Optional[Dict[str, Any]]
+    context: Optional[Dict[str, Any]]
+    metadata: Dict[str, Any]
+
+
+@dataclass
+class EvalRun:
+    run_id: str
+    dataset_id: str
+    name: Optional[str]
+    status: Literal["pending", "running", "completed", "failed"]
+    model: Optional[str]
+    total_items: int
+    completed_items: int
+    passed_items: int
+    failed_items: int
+    avg_score: Optional[float]
+    created_at: str
+
+
+@dataclass
+class EvalResult:
+    result_id: str
+    run_id: str
+    item_id: str
+    input: Dict[str, Any]
+    output: Optional[Dict[str, Any]]
+    expected_output: Optional[Dict[str, Any]]
+    scores: Dict[str, Dict[str, Any]]
+    overall_score: Optional[float]
+    passed: Optional[bool]
+    latency_ms: Optional[int]
+
+
+# ============ Cache Types ============
+
+@dataclass
+class CacheSettings:
+    is_enabled: bool
+    similarity_threshold: float
+    ttl_seconds: int
+    max_entries: int
+    embedding_model: str
+
+
+@dataclass
+class CacheStats:
+    total_entries: int
+    total_hits: int
+    total_tokens_saved: int
+    total_cost_saved_usd: float
+    hit_rate: float
+
+
+@dataclass
+class CacheLookupResult:
+    hit: bool
+    cache_id: Optional[str] = None
+    output: Optional[str] = None
+    metadata: Optional[Dict[str, Any]] = None
+    model: Optional[str] = None
+    reason: Optional[str] = None
+
+
+# ============ Simulation Types ============
+
+@dataclass
+class Simulation:
+    simulation_id: str
+    name: str
+    description: Optional[str]
+    persona: Optional[Dict[str, Any]]
+    scenario: str
+    success_criteria: List[Dict[str, str]]
+    max_turns: int
+    model: Optional[str]
+    status: Literal["draft", "ready", "running", "completed"]
+    tags: List[str]
+
+
+@dataclass
+class SimulationRun:
+    run_id: str
+    simulation_id: str
+    status: Literal["running", "completed", "failed", "stopped"]
+    turn_count: int
+    success: Optional[bool]
+    criteria_results: List[Dict[str, Any]]
+    summary: Optional[str]
+    total_latency_ms: Optional[int]
+    total_tokens: Optional[int]
+    started_at: str
+    completed_at: Optional[str]
+
+
+@dataclass
+class SimulationTurn:
+    turn_id: str
+    run_id: str
+    turn_number: int
+    role: Literal["user", "assistant", "system", "tool"]
+    content: str
+    tool_calls: Optional[List[Dict[str, Any]]]
+    tool_results: Optional[List[Dict[str, Any]]]
+    latency_ms: Optional[int]
+    tokens_in: Optional[int]
+    tokens_out: Optional[int]
+
+
+@dataclass
+class SimulationStats:
+    total_simulations: int
+    total_runs: int
+    success_rate: float
+    avg_turns: float
+
+
+# ============ Integration Types ============
+
+@dataclass
+class Integration:
+    integration_id: str
+    type: Literal["slack", "teams", "discord"]
+    webhook_url: str
+    channel_name: Optional[str]
+    events: List[str]
+    is_active: bool
+    created_at: str
+
+
+# ============ Benchmark Types ============
+
+@dataclass
+class BenchmarkMetric:
+    metric_name: str
+    your_value: float
+    industry_avg: float
+    percentile_25: float
+    percentile_50: float
+    percentile_75: float
+    percentile_90: float
+    your_percentile: float
+    rating: Literal["excellent", "good", "average", "below_average", "needs_attention"]
+    comparison: str
+
+
+@dataclass
+class BenchmarkReport:
+    tenant_id: str
+    industry: str
+    generated_at: str
+    date_range: Dict[str, str]
+    metrics: List[BenchmarkMetric]
+    overall_rating: Literal["excellent", "good", "average", "below_average", "needs_attention"]
+    summary: str
+
+
+@dataclass
+class Industry:
+    id: str
+    name: str
+    description: str
+
+
+# ============ Guardrail Types ============
+
+@dataclass
+class Guardrail:
+    guardrail_id: str
+    name: str
+    description: Optional[str]
+    rule_type: str
+    action: Literal["flag", "block", "redact", "warn"]
+    config: Dict[str, Any]
+    is_active: bool
+    priority: int
+    created_at: str
+
+
+@dataclass
+class GuardrailStats:
+    total_triggers: int
+    by_guardrail: List[Dict[str, Any]]
+    by_action: List[Dict[str, Any]]
+
+
+# ============ Retention Types ============
+
+@dataclass
+class RetentionPolicy:
+    policy_id: str
+    name: str
+    description: Optional[str]
+    is_active: bool
+    data_type: str
+    retention_days: int
+    compliance_hold: bool
+
+
+@dataclass
+class RetentionSummary:
+    active_policies: int
+    total_records_deleted: int
+    last_execution: Optional[str]
+    compliance_holds: int
+
+
+# ============ Audit Types ============
+
+@dataclass
+class AuditLog:
+    log_id: str
+    action: str
+    resource_type: str
+    resource_id: Optional[str]
+    user_id: Optional[str]
+    ip_address: Optional[str]
+    details: Dict[str, Any]
+    created_at: str
+
+
+# ============ Events Types ============
+
+@dataclass
+class Event:
+    event_id: str
+    app_id: str
+    model: str
+    provider: str
+    input_prompt: str
+    output_text: str
+    risk_score: float
+    latency_ms: int
+    tokens_in: int
+    tokens_out: int
+    created_at: str
+    classifications: Optional[Dict[str, Any]] = None
+
+
+# ============ Custom Detection Types ============
+
+@dataclass
+class CustomDetection:
+    detection_id: str
+    name: str
+    description: Optional[str]
+    detection_type: Literal["keyword", "regex", "semantic", "llm_judge"]
+    config: Dict[str, Any]
+    severity: Literal["low", "medium", "high", "critical"]
+    is_active: bool
+    created_at: str
+
+
+# ============ Webhook Types ============
+
+@dataclass
+class Webhook:
+    webhook_id: str
+    url: str
+    events: List[str]
+    is_active: bool
+    secret: Optional[str]
+    created_at: str