Metrics Reference

This document provides a comprehensive reference for all metrics exported by the Cursor Admin API Exporter.

Overview

The exporter provides metrics across several categories:

Team Members: Team composition and roles
Daily Usage: Code activity and AI assistance
Spending: Cost and billing information
Usage Events: Detailed usage tracking
Exporter Health: System performance metrics

Team Members Metrics

`cursor_team_members_total`

Type: Gauge
Description: Total number of team members
Labels: None

# HELP cursor_team_members_total Total number of team members
# TYPE cursor_team_members_total gauge
cursor_team_members_total 15

`cursor_team_members_by_role`

Type: Gauge
Description: Number of team members by role
Labels: role

# HELP cursor_team_members_by_role Number of team members by role
# TYPE cursor_team_members_by_role gauge
cursor_team_members_by_role{role="admin"} 3
cursor_team_members_by_role{role="member"} 10
cursor_team_members_by_role{role="viewer"} 2

Daily Usage Metrics

`cursor_daily_lines_added_total`

Type: Gauge
Description: Total lines of code added per day
Labels: date

# HELP cursor_daily_lines_added_total Total lines of code added per day
# TYPE cursor_daily_lines_added_total gauge
cursor_daily_lines_added_total{date="2024-01-20"} 1250

`cursor_daily_lines_deleted_total`

Type: Gauge
Description: Total lines of code deleted per day
Labels: date

# HELP cursor_daily_lines_deleted_total Total lines of code deleted per day
# TYPE cursor_daily_lines_deleted_total gauge
cursor_daily_lines_deleted_total{date="2024-01-20"} 430

`cursor_daily_suggestion_acceptance_rate`

Type: Gauge
Description: AI suggestion acceptance rate per day (0-1)
Labels: date

# HELP cursor_daily_suggestion_acceptance_rate AI suggestion acceptance rate per day
# TYPE cursor_daily_suggestion_acceptance_rate gauge
cursor_daily_suggestion_acceptance_rate{date="2024-01-20"} 0.75

`cursor_daily_tabs_used_total`

Type: Gauge
Description: Total tab completions used per day
Labels: date

# HELP cursor_daily_tabs_used_total Total tab completions used per day
# TYPE cursor_daily_tabs_used_total gauge
cursor_daily_tabs_used_total{date="2024-01-20"} 1850

`cursor_daily_composer_used_total`

Type: Gauge
Description: Total composer usage per day
Labels: date

# HELP cursor_daily_composer_used_total Total composer usage per day
# TYPE cursor_daily_composer_used_total gauge
cursor_daily_composer_used_total{date="2024-01-20"} 45

`cursor_daily_chat_requests_total`

Type: Gauge
Description: Total chat requests per day
Labels: date

# HELP cursor_daily_chat_requests_total Total chat requests per day
# TYPE cursor_daily_chat_requests_total gauge
cursor_daily_chat_requests_total{date="2024-01-20"} 120

`cursor_daily_model_usage`

Type: Gauge
Description: Most used model per day (1 if this model was most used, 0 otherwise)
Labels: date, model

# HELP cursor_daily_model_usage Most used model per day
# TYPE cursor_daily_model_usage gauge
cursor_daily_model_usage{date="2024-01-20",model="gpt-4"} 1
cursor_daily_model_usage{date="2024-01-20",model="claude-3"} 0

`cursor_daily_extension_usage`

Type: Gauge
Description: Most used extension per day (1 if this extension was most used, 0 otherwise)
Labels: date, extension

# HELP cursor_daily_extension_usage Most used extension per day
# TYPE cursor_daily_extension_usage gauge
cursor_daily_extension_usage{date="2024-01-20",extension="python"} 1
cursor_daily_extension_usage{date="2024-01-20",extension="javascript"} 0

Spending Metrics

`cursor_spending_total_cents`

Type: Gauge
Description: Total spending in cents
Labels: None

# HELP cursor_spending_total_cents Total spending in cents
# TYPE cursor_spending_total_cents gauge
cursor_spending_total_cents 125000

`cursor_spending_by_member_cents`

Type: Gauge
Description: Spending by team member in cents
Labels: member_email, date

# HELP cursor_spending_by_member_cents Spending by team member in cents
# TYPE cursor_spending_by_member_cents gauge
cursor_spending_by_member_cents{member_email="john@example.com",date="2024-01-20"} 8500
cursor_spending_by_member_cents{member_email="jane@example.com",date="2024-01-20"} 12000

`cursor_premium_requests_by_member_total`

Type: Gauge
Description: Premium requests by team member
Labels: member_email, date

# HELP cursor_premium_requests_by_member_total Premium requests by team member
# TYPE cursor_premium_requests_by_member_total gauge
cursor_premium_requests_by_member_total{member_email="john@example.com",date="2024-01-20"} 45
cursor_premium_requests_by_member_total{member_email="jane@example.com",date="2024-01-20"} 67

`cursor_premium_requests_total`

Type: Gauge
Description: Total premium requests
Labels: None

# HELP cursor_premium_requests_total Total premium requests
# TYPE cursor_premium_requests_total gauge
cursor_premium_requests_total 450

Usage Events Metrics

`cursor_usage_events_total`

Type: Gauge
Description: Total number of usage events
Labels: None

# HELP cursor_usage_events_total Total number of usage events
# TYPE cursor_usage_events_total gauge
cursor_usage_events_total 2850

`cursor_usage_events_by_type_total`

Type: Gauge
Description: Number of usage events by type
Labels: event_type

# HELP cursor_usage_events_by_type_total Number of usage events by type
# TYPE cursor_usage_events_by_type_total gauge
cursor_usage_events_by_type_total{event_type="completion"} 1200
cursor_usage_events_by_type_total{event_type="chat"} 450
cursor_usage_events_by_type_total{event_type="edit"} 800

`cursor_usage_events_by_user_total`

Type: Gauge
Description: Number of usage events by user
Labels: user_email

# HELP cursor_usage_events_by_user_total Number of usage events by user
# TYPE cursor_usage_events_by_user_total gauge
cursor_usage_events_by_user_total{user_email="john@example.com"} 350
cursor_usage_events_by_user_total{user_email="jane@example.com"} 425

`cursor_usage_events_by_model_total`

Type: Gauge
Description: Number of usage events by model
Labels: model

# HELP cursor_usage_events_by_model_total Number of usage events by model
# TYPE cursor_usage_events_by_model_total gauge
cursor_usage_events_by_model_total{model="gpt-4"} 1850
cursor_usage_events_by_model_total{model="claude-3"} 650
cursor_usage_events_by_model_total{model="gpt-3.5-turbo"} 350

`cursor_tokens_consumed_total`

Type: Gauge
Description: Total tokens consumed
Labels: None

# HELP cursor_tokens_consumed_total Total tokens consumed
# TYPE cursor_tokens_consumed_total gauge
cursor_tokens_consumed_total 1250000

`cursor_tokens_consumed_by_model_total`

Type: Gauge
Description: Tokens consumed by model
Labels: model

# HELP cursor_tokens_consumed_by_model_total Tokens consumed by model
# TYPE cursor_tokens_consumed_by_model_total gauge
cursor_tokens_consumed_by_model_total{model="gpt-4"} 850000
cursor_tokens_consumed_by_model_total{model="claude-3"} 300000
cursor_tokens_consumed_by_model_total{model="gpt-3.5-turbo"} 100000

`cursor_tokens_consumed_by_user_total`

Type: Gauge
Description: Tokens consumed by user
Labels: user_email

# HELP cursor_tokens_consumed_by_user_total Tokens consumed by user
# TYPE cursor_tokens_consumed_by_user_total gauge
cursor_tokens_consumed_by_user_total{user_email="john@example.com"} 125000
cursor_tokens_consumed_by_user_total{user_email="jane@example.com"} 180000

Exporter Health Metrics

`cursor_exporter_scrape_duration_seconds`

Type: Histogram
Description: Time spent scraping the Cursor Admin API
Labels: None

# HELP cursor_exporter_scrape_duration_seconds Time spent scraping Cursor Admin API
# TYPE cursor_exporter_scrape_duration_seconds histogram
cursor_exporter_scrape_duration_seconds_bucket{le="0.1"} 0
cursor_exporter_scrape_duration_seconds_bucket{le="0.5"} 12
cursor_exporter_scrape_duration_seconds_bucket{le="1"} 45
cursor_exporter_scrape_duration_seconds_bucket{le="2.5"} 48
cursor_exporter_scrape_duration_seconds_bucket{le="5"} 48
cursor_exporter_scrape_duration_seconds_bucket{le="10"} 48
cursor_exporter_scrape_duration_seconds_bucket{le="+Inf"} 48
cursor_exporter_scrape_duration_seconds_sum 18.5
cursor_exporter_scrape_duration_seconds_count 48

`cursor_exporter_scrape_errors_total`

Type: Counter
Description: Total number of scrape errors
Labels: None

# HELP cursor_exporter_scrape_errors_total Total number of scrape errors
# TYPE cursor_exporter_scrape_errors_total counter
cursor_exporter_scrape_errors_total 2

Metric Labels

Common Labels

Label	Description	Example Values
`date`	Date in YYYY-MM-DD format	`2024-01-20`
`member_email`	Team member email address	`john@example.com`
`user_email`	User email address	`jane@example.com`
`role`	Team member role	`admin`, `member`, `viewer`
`model`	AI model name	`gpt-4`, `claude-3`, `gpt-3.5-turbo`
`extension`	File extension	`python`, `javascript`, `go`
`event_type`	Type of usage event	`completion`, `chat`, `edit`

Metric Collection

Collection Frequency

Default: Every 30 seconds (configurable via Prometheus)
Recommended: 30-60 seconds for production
Minimum: 15 seconds (to avoid API rate limits)

Data Freshness

Team Members: Real-time
Daily Usage: Updated daily at midnight UTC
Spending: Updated hourly
Usage Events: Updated every 5 minutes

Querying Examples

PromQL Queries

Team Productivity

# Average suggestion acceptance rate over 7 days
avg_over_time(cursor_daily_suggestion_acceptance_rate[7d])

# Total lines of code added per week
sum(increase(cursor_daily_lines_added_total[7d]))

# Most active users by token consumption
topk(5, cursor_tokens_consumed_by_user_total)

Cost Analysis

# Daily spending trend
cursor_spending_total_cents / 100

# Top spenders by team member
topk(10, cursor_spending_by_member_cents / 100)

# Premium request utilization
cursor_premium_requests_total / cursor_team_members_total

Model Usage

# Most used AI models
topk(5, cursor_tokens_consumed_by_model_total)

# Model usage distribution
(cursor_tokens_consumed_by_model_total / cursor_tokens_consumed_total) * 100

Grafana Queries

Dashboard Variables

# Team members for filtering
label_values(cursor_usage_events_by_user_total, user_email)

# Available models
label_values(cursor_tokens_consumed_by_model_total, model)

# Date range for daily metrics
label_values(cursor_daily_lines_added_total, date)

Panels

# Lines of code trend
cursor_daily_lines_added_total - cursor_daily_lines_deleted_total

# Suggestion acceptance rate trend
cursor_daily_suggestion_acceptance_rate * 100

# Cost per team member
cursor_spending_total_cents / cursor_team_members_total / 100

Alerting Rules

Suggested Alerts

# Low suggestion acceptance rate
- alert: CursorLowSuggestionAcceptance
  expr: cursor_daily_suggestion_acceptance_rate < 0.3
  for: 1h
  labels:
    severity: warning
  annotations:
    summary: "Low AI suggestion acceptance rate"

# High spending
- alert: CursorHighSpending
  expr: increase(cursor_spending_total_cents[24h]) > 10000
  for: 15m
  labels:
    severity: critical
  annotations:
    summary: "High daily spending detected"

# API scrape errors
- alert: CursorExporterErrors
  expr: rate(cursor_exporter_scrape_errors_total[5m]) > 0.1
  for: 5m
  labels:
    severity: warning
  annotations:
    summary: "Cursor exporter experiencing errors"

Troubleshooting Metrics

Missing Metrics

Check API connectivity: cursor_exporter_scrape_errors_total
Verify authentication: Look for 401 errors in logs
Check rate limits: High scrape error rates
Validate permissions: Ensure API token has admin access

Incorrect Values

Time zone issues: Daily metrics use UTC
API delays: Some metrics may have delays
Caching: API responses may be cached
Permissions: Limited data based on token permissions

Related Documentation: