:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
import { HogBytecode } from '../cdp/types'
import { PropertyFilter } from '../types'
export type EvaluationStatus = 'active' | 'paused' | 'error'
// Keep in sync with EvaluationStatusReason in products/ai_observability/backend/models/evaluations.py
export type EvaluationStatusReason =
| 'provider_key_required'
| 'provider_key_deleted'
| 'no_default_model'
| 'provider_key_invalid'
| 'provider_key_permission_denied'
| 'provider_key_quota_exceeded'
| 'provider_key_rate_limited'
| 'model_not_found'
| 'model_not_supported'
| 'hog_error'
export type EvaluationTarget = 'generation' | 'trace' | 'session'
export type EvaluationSettleStrategy = 'fixed_window' | 'inactivity'
/** Settle config stored on the evaluation. Rows saved before strategies existed have no
* `strategy` key and mean 'fixed_window'. */
export interface EvaluationTargetConfig {
strategy?: EvaluationSettleStrategy
window_seconds?: number
quiet_period_seconds?: number
max_age_seconds?: number
}
/** Fully-resolved settle config passed to the workflow (defaults applied, one shape per strategy). */
export type ResolvedSettleConfig =
| { strategy: 'fixed_window'; window_seconds: number }
| { strategy: 'inactivity'; quiet_period_seconds: number; max_age_seconds: number }
export interface Evaluation {
id: string
team_id: number
name: string
description?: string
enabled: boolean
status: EvaluationStatus
status_reason?: EvaluationStatusReason
evaluation_type: string
evaluation_config: Record<string, any>
output_type: string
output_config: Record<string, any>
conditions: EvaluationConditionSet[]
target: EvaluationTarget
target_config: EvaluationTargetConfig
provider_key_id?: string | null
created_at: string
updated_at: string
}
export interface EvaluationConditionSet {
id: string
rollout_percentage: number
properties: PropertyFilter[]
bytecode?: HogBytecode // Compiled on save in Python, embedded in JSON
bytecode_error?: string
}
export type EvaluationInfo = Pick<Evaluation, 'id' | 'team_id'>
/** Shared interface for condition matching — used by both evaluations and taggers. */
export interface Matchable {
enabled: boolean
conditions: EvaluationConditionSet[]
}
export interface Tagger {
id: string
team_id: number
name: string
description?: string
enabled: boolean
tagger_type: string
tagger_config: {
prompt: string
tags: { name: string; description?: string }[]
min_tags: number
max_tags: number | null
}
conditions: EvaluationConditionSet[]
provider_key_id?: string | null
created_at: string
updated_at: string
}
export type TaggerInfo = Pick<Tagger, 'id' | 'team_id'>