Performance Monitoring - Aden Documentation

The AgentLens API allows you to programmatically manage, query, and extend the performance metrics of your AI agents. Use these endpoints to power your own internal dashboards or to feed data into the AgentLens Customer Portal.

Overview

The Performance Dashboard API provides a RESTful interface for accessing agent performance metrics, cost analytics, and custom metric definitions. This API enables developers to programmatically monitor their AI agents, track spending, and define business-specific KPIs. Base URL: https://api.adenhq.com/v1/analytics

Authentication

All API requests require authentication via API key in the header:

Authorization: Bearer sk-ant-xxxxxx
x-api-version: 2025-01-06

Data Models

Core Metric Types

enum MetricType {
  COUNTER     = "counter"      // Cumulative values (requests, tokens)
  GAUGE       = "gauge"        // Point-in-time values (latency, rate)
  HISTOGRAM   = "histogram"    // Distribution data (latency percentiles)
  TIMESERIES  = "timeseries"   // Time-indexed values
}

enum AggregationMethod {
  SUM         = "sum"
  AVG         = "avg"
  MIN         = "min"
  MAX         = "max"
  P50         = "p50"
  P95         = "p95"
  P99         = "p99"
  COUNT       = "count"
  RATE        = "rate"
}

Standard Metric Schema

interface Metric {
  id: string                          // Unique metric identifier
  name: string                        // Human-readable name
  type: MetricType                    // Metric classification
  unit: string                        // "usd", "ms", "tokens", "percent", "count"
  description: string                 // Metric description
  value: number | null                // Current/aggregated value
  timestamp: string                   // ISO 8601 timestamp
  metadata: Record<string, string>    // Additional context
}

Time Series Data Point

interface DataPoint {
  timestamp: string                   // ISO 8601 timestamp
  value: number                       // Metric value at timestamp
  dimensions: Record<string, string>  // Grouping dimensions (agent_id, model, customer_id)
}

interface TimeSeriesResponse {
  metric_id: string
  metric_name: string
  unit: string
  granularity: string                 // "minute", "hour", "day", "week", "month"
  data: DataPoint[]
  aggregation: AggregationMethod
}

Percentile Distribution

interface PercentileDistribution {
  p50: number
  p75: number
  p90: number
  p95: number
  p99: number
  min: number
  max: number
  mean: number
  count: number
}

Standard Metrics Catalog

Cost Metrics

Metric ID	Name	Type	Unit	Description
`cost.total`	Total Cost	counter	usd	Cumulative spend across all agents
`cost.daily`	Daily Cost	timeseries	usd	Cost aggregated by day
`cost.by_model`	Cost by Model	gauge	usd	Spend breakdown by model
`cost.by_agent`	Cost by Agent	gauge	usd	Spend breakdown by agent
`cost.per_completion`	Cost per Completion	gauge	usd	Average cost per successful completion
`cost.cache_savings`	Cache Savings	counter	usd	Money saved via prompt caching
`cost.burn_rate`	Budget Burn Rate	gauge	percent	Current spend rate vs. budget

Usage Metrics

Metric ID	Name	Type	Unit	Description
`usage.requests.total`	Total Requests	counter	count	Total API requests
`usage.requests.daily`	Daily Requests	timeseries	count	Requests aggregated by day
`usage.tokens.total`	Total Tokens	counter	tokens	Total tokens consumed
`usage.tokens.input`	Input Tokens	counter	tokens	Prompt/input tokens
`usage.tokens.output`	Output Tokens	counter	tokens	Completion/output tokens

Performance Metrics

Metric ID	Name	Type	Unit	Description
`perf.latency.avg`	Average Latency	gauge	ms	Mean response time
`perf.latency.distribution`	Latency Distribution	histogram	ms	Full latency percentiles
`perf.latency.p50`	Latency P50	gauge	ms	Median latency
`perf.latency.p95`	Latency P95	gauge	ms	95th percentile latency
`perf.latency.p99`	Latency P99	gauge	ms	99th percentile latency

Autonomy Metrics

Metric ID	Name	Type	Unit	Description
`autonomy.intervention_rate`	Intervention Rate	gauge	percent	Sessions requiring human handoff
`autonomy.goal_completion_rate`	Goal Completion Rate	gauge	percent	Sessions achieving business intent
`autonomy.steps_to_resolution`	Steps to Resolution	gauge	count	Avg turns to solve a problem
`autonomy.self_correction_rate`	Self-Correction Rate	gauge	percent	Errors auto-recovered without human
`autonomy.deflection_rate`	Deflection Rate	gauge	percent	Requests resolved without human

Quality Metrics

Metric ID	Name	Type	Unit	Description
`quality.tool_success_rate`	Tool Usage Success Rate	gauge	percent	Successful tool/API calls
`quality.hallucination_rate`	Hallucination Rate	gauge	percent	Responses not grounded in context
`quality.loop_rate`	Loop Rate	gauge	percent	Sessions with repeated actions
`quality.sentiment_shift`	Sentiment Shift	gauge	score	Change in user sentiment (-1 to +1)

API Endpoints

1. Get Single Metric (Point-in-Time)

Retrieve the current value of a specific metric.

GET /metrics/{metric_id}

Path Parameters:

Parameter	Type	Required	Description
`metric_id`	string	Yes	The metric identifier (e.g., `cost.total`)

Query Parameters:

Parameter	Type	Required	Default	Description
`at`	string	No	now	ISO 8601 timestamp for point-in-time query
`agent_id`	string	No	all	Filter by specific agent
`model`	string	No	all	Filter by model (e.g., `claude-sonnet-4-5-20250929`)
`customer_id`	string	No	all	Filter by end-customer (multi-tenant)

Request Example:

curl -X GET "https://api.adenhq.com/v1/analytics/metrics/cost.total?agent_id=agent_abc123" \
  -H "Authorization: Bearer sk-ant-xxxxxx" \
  -H "x-api-version: 2025-01-06"

Response:

{
  "id": "cost.total",
  "name": "Total Cost",
  "type": "counter",
  "unit": "usd",
  "value": 1547.82,
  "timestamp": "2025-01-06T14:30:00Z",
  "metadata": {
    "agent_id": "agent_abc123",
    "period_start": "2025-01-01T00:00:00Z"
  }
}

2. Get Metric Time Series

Retrieve metric values over a time range.

GET /metrics/{metric_id}/series

Path Parameters:

Parameter	Type	Required	Description
`metric_id`	string	Yes	The metric identifier

Query Parameters:

Parameter	Type	Required	Default	Description
`start`	string	Yes	—	Start of time range (ISO 8601)
`end`	string	No	now	End of time range (ISO 8601)
`granularity`	string	No	auto	`minute`, `hour`, `day`, `week`, `month`
`aggregation`	string	No	sum	Aggregation method for the granularity
`agent_id`	string	No	all	Filter by agent
`model`	string	No	all	Filter by model
`customer_id`	string	No	all	Filter by end-customer
`group_by`	string	No	none	Dimension to group by: `agent_id`, `model`, `customer_id`
`limit`	integer	No	1000	Max data points returned

Request Example:

curl -X GET "https://api.adenhq.com/v1/analytics/metrics/cost.daily/series?\
start=2025-01-01T00:00:00Z&\
end=2025-01-06T23:59:59Z&\
granularity=day&\
group_by=model" \
  -H "Authorization: Bearer sk-ant-xxxxxx" \
  -H "x-api-version: 2025-01-06"

Response:

{
  "metric_id": "cost.daily",
  "metric_name": "Daily Cost",
  "unit": "usd",
  "granularity": "day",
  "aggregation": "sum",
  "start": "2025-01-01T00:00:00Z",
  "end": "2025-01-06T23:59:59Z",
  "data": [
    {
      "timestamp": "2025-01-01T00:00:00Z",
      "value": 234.56,
      "dimensions": { "model": "claude-sonnet-4-5-20250929" }
    },
    {
      "timestamp": "2025-01-01T00:00:00Z",
      "value": 89.12,
      "dimensions": { "model": "claude-haiku-4-5-20251001" }
    },
    {
      "timestamp": "2025-01-02T00:00:00Z",
      "value": 256.78,
      "dimensions": { "model": "claude-sonnet-4-5-20250929" }
    }
  ]
}

3. Get Percentile Distribution

Retrieve distribution data for histogram-type metrics.

GET /metrics/{metric_id}/distribution

Query Parameters:

Parameter	Type	Required	Default	Description
`start`	string	Yes	—	Start of time range
`end`	string	No	now	End of time range
`agent_id`	string	No	all	Filter by agent
`model`	string	No	all	Filter by model
`buckets`	integer	No	20	Number of histogram buckets

Request Example:

curl -X GET "https://api.adenhq.com/v1/analytics/metrics/perf.latency.distribution/distribution?\
start=2025-01-06T00:00:00Z" \
  -H "Authorization: Bearer sk-ant-xxxxxx" \
  -H "x-api-version: 2025-01-06"

Response:

{
  "metric_id": "perf.latency.distribution",
  "metric_name": "Latency Distribution",
  "unit": "ms",
  "start": "2025-01-06T00:00:00Z",
  "end": "2025-01-06T14:30:00Z",
  "percentiles": {
    "p50": 245,
    "p75": 380,
    "p90": 520,
    "p95": 680,
    "p99": 1250,
    "min": 89,
    "max": 3400,
    "mean": 312,
    "count": 15420
  },
  "histogram": [
    { "bucket": "0-100", "count": 1542 },
    { "bucket": "100-200", "count": 4521 },
    { "bucket": "200-300", "count": 3890 },
    { "bucket": "300-500", "count": 2156 },
    { "bucket": "500-1000", "count": 2845 },
    { "bucket": "1000+", "count": 466 }
  ]
}

4. Batch Get Metrics

Retrieve multiple metrics in a single request.

POST /metrics/batch

Request Body:

{
  "metric_ids": ["cost.total", "usage.requests.total", "perf.latency.avg"],
  "filters": {
    "agent_id": "agent_abc123",
    "start": "2025-01-01T00:00:00Z",
    "end": "2025-01-06T23:59:59Z"
  }
}

Response:

{
  "metrics": [
    {
      "id": "cost.total",
      "name": "Total Cost",
      "type": "counter",
      "unit": "usd",
      "value": 1547.82,
      "timestamp": "2025-01-06T14:30:00Z"
    },
    {
      "id": "usage.requests.total",
      "name": "Total Requests",
      "type": "counter",
      "unit": "count",
      "value": 45230,
      "timestamp": "2025-01-06T14:30:00Z"
    },
    {
      "id": "perf.latency.avg",
      "name": "Average Latency",
      "type": "gauge",
      "unit": "ms",
      "value": 312,
      "timestamp": "2025-01-06T14:30:00Z"
    }
  ],
  "filters_applied": {
    "agent_id": "agent_abc123",
    "start": "2025-01-01T00:00:00Z",
    "end": "2025-01-06T23:59:59Z"
  }
}

5. Get Dashboard Summary

Retrieve all KPIs for dashboard rendering.

GET /dashboard/summary

Query Parameters:

Parameter	Type	Required	Default	Description
`period`	string	No	30d	Predefined period: `24h`, `7d`, `30d`, `90d`, `ytd`
`start`	string	No	—	Custom start (overrides period)
`end`	string	No	now	Custom end
`agent_id`	string	No	all	Filter by agent
`customer_id`	string	No	all	Filter by customer

Response:

{
  "period": {
    "start": "2024-12-07T00:00:00Z",
    "end": "2025-01-06T14:30:00Z",
    "label": "30d"
  },
  "kpis": {
    "total_cost": {
      "value": 4521.67,
      "unit": "usd",
      "change": 12.5,
      "change_period": "vs previous 30d"
    },
    "total_requests": {
      "value": 128450,
      "unit": "count",
      "change": 8.2,
      "change_period": "vs previous 30d"
    },
    "total_tokens": {
      "value": 45200000,
      "unit": "tokens",
      "change": 15.1,
      "change_period": "vs previous 30d"
    },
    "avg_latency": {
      "value": 298,
      "unit": "ms",
      "change": -5.3,
      "change_period": "vs previous 30d"
    },
    "cache_savings": {
      "value": 892.45,
      "unit": "usd",
      "change": 22.1,
      "change_period": "vs previous 30d"
    }
  },
  "top_agents_by_spend": [
    { "agent_id": "support-bot", "cost": 1823.45, "requests": 52340 },
    { "agent_id": "sales-assistant", "cost": 1456.22, "requests": 41200 },
    { "agent_id": "data-analyst", "cost": 892.00, "requests": 23100 }
  ],
  "cost_by_model": [
    { "model": "claude-sonnet-4-5-20250929", "cost": 3200.45, "percentage": 70.8 },
    { "model": "claude-haiku-4-5-20251001", "cost": 1021.22, "percentage": 22.6 },
    { "model": "claude-opus-4-5-20251101", "cost": 300.00, "percentage": 6.6 }
  ]
}

Custom Metrics API

Custom metrics allow developers to define business-specific KPIs using formulas based on standard metrics.

Custom Metric Schema

interface CustomMetricDefinition {
  id: string                          // Auto-generated or user-defined
  name: string                        // Display name
  description: string                 // What this metric measures
  formula: FormulaExpression          // Calculation definition
  unit: string                        // Result unit
  created_at: string                  // ISO 8601 timestamp
  updated_at: string                  // ISO 8601 timestamp
  is_active: boolean                  // Whether metric is being calculated
}

interface FormulaExpression {
  type: "arithmetic" | "conditional" | "aggregate"
  expression: string                  // Formula string
  inputs: FormulaInput[]              // Referenced metrics
  conditions?: ConditionalRule[]      // Optional conditional logic
}

interface FormulaInput {
  variable: string                    // Variable name in formula (e.g., "$a")
  metric_id: string                   // Standard metric to reference
  aggregation?: AggregationMethod     // How to aggregate the input
  filters?: Record<string, string>    // Filters to apply to input
}

interface ConditionalRule {
  condition: string                   // Condition expression
  then_value: string | number         // Value if true
  else_value: string | number         // Value if false
}

6. List Custom Metrics

GET /metrics/custom

Query Parameters:

Parameter	Type	Required	Default	Description
`is_active`	boolean	No	all	Filter by active status
`limit`	integer	No	50	Max results
`offset`	integer	No	0	Pagination offset

Response:

{
  "custom_metrics": [
    {
      "id": "cm_roi_per_session",
      "name": "ROI per Session",
      "description": "Revenue generated minus cost per agent session",
      "unit": "usd",
      "is_active": true,
      "created_at": "2025-01-02T10:00:00Z",
      "updated_at": "2025-01-05T14:30:00Z"
    },
    {
      "id": "cm_efficiency_score",
      "name": "Agent Efficiency Score",
      "description": "Composite score of completion rate, speed, and cost",
      "unit": "score",
      "is_active": true,
      "created_at": "2025-01-03T09:15:00Z",
      "updated_at": "2025-01-03T09:15:00Z"
    }
  ],
  "total": 2,
  "limit": 50,
  "offset": 0
}

7. Get Custom Metric Definition

GET /metrics/custom/{custom_metric_id}

Response:

{
  "id": "cm_roi_per_session",
  "name": "ROI per Session",
  "description": "Revenue generated minus cost per agent session",
  "unit": "usd",
  "formula": {
    "type": "arithmetic",
    "expression": "($a - $b) / $c",
    "inputs": [
      {
        "variable": "$a",
        "metric_id": "custom.revenue_attributed",
        "aggregation": "sum"
      },
      {
        "variable": "$b",
        "metric_id": "cost.total",
        "aggregation": "sum"
      },
      {
        "variable": "$c",
        "metric_id": "usage.sessions.total",
        "aggregation": "count"
      }
    ]
  },
  "is_active": true,
  "created_at": "2025-01-02T10:00:00Z",
  "updated_at": "2025-01-05T14:30:00Z"
}

8. Create Custom Metric

POST /metrics/custom

Request Body:

{
  "id": "cm_cost_per_deflection",
  "name": "Cost per Deflection",
  "description": "Average cost to successfully deflect a support ticket",
  "unit": "usd",
  "formula": {
    "type": "arithmetic",
    "expression": "$a / ($b * $c)",
    "inputs": [
      {
        "variable": "$a",
        "metric_id": "cost.total",
        "aggregation": "sum",
        "filters": { "agent_id": "support-bot" }
      },
      {
        "variable": "$b",
        "metric_id": "usage.sessions.total",
        "aggregation": "count",
        "filters": { "agent_id": "support-bot" }
      },
      {
        "variable": "$c",
        "metric_id": "autonomy.deflection_rate",
        "aggregation": "avg",
        "filters": { "agent_id": "support-bot" }
      }
    ]
  }
}

Response:

{
  "id": "cm_cost_per_deflection",
  "name": "Cost per Deflection",
  "description": "Average cost to successfully deflect a support ticket",
  "unit": "usd",
  "formula": { ... },
  "is_active": true,
  "created_at": "2025-01-06T15:00:00Z",
  "updated_at": "2025-01-06T15:00:00Z"
}

9. Update Custom Metric

PATCH /metrics/custom/{custom_metric_id}

Request Body (partial update):

{
  "name": "Cost per Ticket Deflection",
  "description": "Updated: Average cost to deflect a support ticket without human intervention",
  "is_active": true
}

Response:

{
  "id": "cm_cost_per_deflection",
  "name": "Cost per Ticket Deflection",
  "description": "Updated: Average cost to deflect a support ticket without human intervention",
  "unit": "usd",
  "formula": { ... },
  "is_active": true,
  "created_at": "2025-01-06T15:00:00Z",
  "updated_at": "2025-01-06T15:45:00Z"
}

10. Delete Custom Metric

DELETE /metrics/custom/{custom_metric_id}

Response:

{
  "deleted": true,
  "id": "cm_cost_per_deflection"
}

11. Evaluate Custom Metric

Query the calculated value of a custom metric.

GET /metrics/custom/{custom_metric_id}/evaluate

Query Parameters: Same as standard metric time series endpoint (start, end, granularity, etc.) Response:

{
  "metric_id": "cm_cost_per_deflection",
  "metric_name": "Cost per Ticket Deflection",
  "unit": "usd",
  "value": 0.42,
  "timestamp": "2025-01-06T15:00:00Z",
  "formula_breakdown": {
    "$a": { "metric_id": "cost.total", "value": 1823.45 },
    "$b": { "metric_id": "usage.sessions.total", "value": 52340 },
    "$c": { "metric_id": "autonomy.deflection_rate", "value": 0.083 }
  },
  "period": {
    "start": "2025-01-01T00:00:00Z",
    "end": "2025-01-06T15:00:00Z"
  }
}

12. Validate Custom Metric Formula

Test a formula before creating a metric.

POST /metrics/custom/validate

Request Body:

{
  "formula": {
    "type": "arithmetic",
    "expression": "$a / $b * 100",
    "inputs": [
      { "variable": "$a", "metric_id": "autonomy.goal_completion_rate" },
      { "variable": "$b", "metric_id": "autonomy.steps_to_resolution" }
    ]
  },
  "test_period": {
    "start": "2025-01-01T00:00:00Z",
    "end": "2025-01-06T00:00:00Z"
  }
}

Response:

{
  "valid": true,
  "test_result": {
    "value": 23.5,
    "inputs_resolved": {
      "$a": 94.0,
      "$b": 4.0
    }
  },
  "warnings": []
}

Error Response (invalid formula):

{
  "valid": false,
  "errors": [
    {
      "code": "INVALID_METRIC_REFERENCE",
      "message": "Metric 'invalid.metric.id' does not exist",
      "variable": "$b"
    }
  ]
}

Formula Expression Language

Custom metrics use a simple expression language for calculations.

Supported Operators

Operator	Description	Example
`+`	Addition	`$a + $b`
`-`	Subtraction	`$a - $b`
`*`	Multiplication	`$a * $b`
`/`	Division	`$a / $b`
`%`	Modulo	`$a % $b`
`^`	Power	`$a ^ 2`
`()`	Grouping	`($a + $b) / $c`

Built-in Functions

Function	Description	Example
`abs(x)`	Absolute value	`abs($a - $b)`
`min(a, b)`	Minimum of two values	`min($a, $b)`
`max(a, b)`	Maximum of two values	`max($a, $b)`
`round(x, n)`	Round to n decimals	`round($a / $b, 2)`
`if(cond, then, else)`	Conditional	`if($b > 0, $a / $b, 0)`
`coalesce(a, b)`	First non-null	`coalesce($a, 0)`

Example Formulas

Efficiency Score (composite):

round((($gcr * 0.4) + ((1 - $intervention) * 0.3) + (1 / $steps * 0.3)) * 100, 1)

Cost per Successful Task:

if($success_count > 0, $total_cost / $success_count, 0)

Normalized Latency Score (0-100):

max(0, 100 - (($p95_latency - 200) / 10))

Webhooks & Alerts (Future)

Alert Rule Schema

interface AlertRule {
  id: string
  name: string
  metric_id: string                   // Standard or custom metric
  condition: {
    operator: "gt" | "lt" | "gte" | "lte" | "eq"
    threshold: number
    duration: string                  // e.g., "5m", "1h" - sustained period
  }
  actions: AlertAction[]
  is_enabled: boolean
}

interface AlertAction {
  type: "webhook" | "email" | "slack"
  target: string                      // URL, email, or channel
}

Error Responses

All errors follow a consistent format:

{
  "error": {
    "type": "invalid_request_error",
    "code": "INVALID_TIME_RANGE",
    "message": "Start time must be before end time",
    "param": "start",
    "documentation_url": "https://docs.adenhq.com/errors/INVALID_TIME_RANGE"
  }
}

Error Codes

HTTP Status	Error Type	Description
400	`invalid_request_error`	Malformed request or invalid parameters
401	`authentication_error`	Invalid or missing API key
403	`permission_error`	Insufficient permissions for resource
404	`not_found_error`	Metric or resource not found
409	`conflict_error`	Resource already exists (duplicate ID)
429	`rate_limit_error`	Too many requests
500	`api_error`	Internal server error

Rate Limits

Endpoint Category	Requests per Minute
Read (GET)	300
Write (POST/PATCH/DELETE)	60
Batch operations	30

Rate limit headers are included in all responses:

X-RateLimit-Limit: 300
X-RateLimit-Remaining: 287
X-RateLimit-Reset: 1704553200

SDK Examples

Python SDK

from anthropic import AnalyticsClient

client = AnalyticsClient(api_key="sk-ant-xxxxxx")

# Get dashboard summary
summary = client.dashboard.summary(period="30d")
print(f"Total cost: ${summary.kpis.total_cost.value}")

# Query time series
cost_trend = client.metrics.series(
    metric_id="cost.daily",
    start="2025-01-01",
    end="2025-01-06",
    granularity="day",
    group_by="model"
)

for point in cost_trend.data:
    print(f"{point.timestamp}: ${point.value} ({point.dimensions['model']})")

# Create custom metric
custom_metric = client.metrics.custom.create(
    id="cm_efficiency",
    name="Agent Efficiency",
    unit="score",
    formula={
        "type": "arithmetic",
        "expression": "($gcr * $deflection) / $steps",
        "inputs": [
            {"variable": "$gcr", "metric_id": "autonomy.goal_completion_rate"},
            {"variable": "$deflection", "metric_id": "autonomy.deflection_rate"},
            {"variable": "$steps", "metric_id": "autonomy.steps_to_resolution"}
        ]
    }
)

# Evaluate custom metric
result = client.metrics.custom.evaluate(
    custom_metric_id="cm_efficiency",
    start="2025-01-01",
    end="2025-01-06"
)
print(f"Efficiency score: {result.value}")

TypeScript SDK

import { AnalyticsClient } from '@anthropic/analytics';

const client = new AnalyticsClient({ apiKey: 'sk-ant-xxxxxx' });

// Get single metric
const totalCost = await client.metrics.get('cost.total', {
  agentId: 'support-bot',
});
console.log(`Total cost: $${totalCost.value}`);

// Get latency distribution
const latency = await client.metrics.distribution('perf.latency.distribution', {
  start: '2025-01-06T00:00:00Z',
});
console.log(`P95 Latency: ${latency.percentiles.p95}ms`);

// Batch query
const kpis = await client.metrics.batch({
  metricIds: ['cost.total', 'usage.requests.total', 'perf.latency.avg'],
  filters: {
    agentId: 'support-bot',
    start: '2025-01-01T00:00:00Z',
  },
});

// Create and evaluate custom metric
const customMetric = await client.metrics.custom.create({
  id: 'cm_cost_per_success',
  name: 'Cost per Successful Completion',
  unit: 'usd',
  formula: {
    type: 'arithmetic',
    expression: '$cost / $successes',
    inputs: [
      { variable: '$cost', metricId: 'cost.total' },
      { variable: '$successes', metricId: 'usage.completions.successful', aggregation: 'count' },
    ],
  },
});

const evaluation = await client.metrics.custom.evaluate('cm_cost_per_success', {
  start: '2025-01-01T00:00:00Z',
});
console.log(`Cost per success: $${evaluation.value.toFixed(2)}`);

Support

For API support, contact contact@adenhq.com.

Getting Started

Use Cases

TypeScript SDK

Python SDK

​Overview

​Authentication

​Data Models

​Core Metric Types

​Standard Metric Schema

​Time Series Data Point

​Percentile Distribution

​Standard Metrics Catalog

​Cost Metrics

​Usage Metrics

​Performance Metrics

​Autonomy Metrics

​Quality Metrics

​API Endpoints

​1. Get Single Metric (Point-in-Time)

​2. Get Metric Time Series

​3. Get Percentile Distribution

​4. Batch Get Metrics

​5. Get Dashboard Summary

​Custom Metrics API

​Custom Metric Schema

​6. List Custom Metrics

​7. Get Custom Metric Definition

​8. Create Custom Metric

​9. Update Custom Metric

​10. Delete Custom Metric

​11. Evaluate Custom Metric

​12. Validate Custom Metric Formula

​Formula Expression Language

​Supported Operators

​Built-in Functions

​Example Formulas

​Webhooks & Alerts (Future)

​Alert Rule Schema

​Error Responses

​Error Codes

​Rate Limits

​SDK Examples

​Python SDK

​TypeScript SDK

​Support

Overview

Authentication

Data Models

Core Metric Types

Standard Metric Schema

Time Series Data Point

Percentile Distribution

Standard Metrics Catalog

Cost Metrics

Usage Metrics

Performance Metrics

Autonomy Metrics

Quality Metrics

API Endpoints

1. Get Single Metric (Point-in-Time)

2. Get Metric Time Series

3. Get Percentile Distribution

4. Batch Get Metrics

5. Get Dashboard Summary

Custom Metrics API

Custom Metric Schema

6. List Custom Metrics

7. Get Custom Metric Definition

8. Create Custom Metric

9. Update Custom Metric

10. Delete Custom Metric

11. Evaluate Custom Metric

12. Validate Custom Metric Formula

Formula Expression Language

Supported Operators

Built-in Functions

Example Formulas

Webhooks & Alerts (Future)

Alert Rule Schema

Error Responses

Error Codes

Rate Limits

SDK Examples

Python SDK

TypeScript SDK

Support