Skip to main content

Overview

The Analytics endpoints expose the same monitoring data that powers your dashboard: pre-computed metric aggregations, raw executions, availability samples, incidents and anomalies. They all require an API key with analytics: read (see Authentication) and are automatically scoped to the connections your key can access. Base URL:

Cursor Pagination

The read endpoints use keyset (cursor) pagination rather than offsets — stable even as new rows arrive. Every response carries a pagination object:
To fetch the next page, pass next_cursor back as cursor. When has_more is false, you’ve reached the end.

Metric Aggregations

Returns pre-computed daily/weekly/monthly rollups.

Parameters

Valid metric_type values: daily_executions, daily_success_rate, daily_avg_duration, weekly_executions, weekly_success_rate, weekly_avg_duration, monthly_executions, monthly_success_rate, monthly_avg_duration, daily_availability, weekly_availability, monthly_availability, monthly_openclaw_cost

Executions

Raw workflow execution records.

Availability

Raw availability check samples.

Availability Summary

Rolled-up uptime instead of raw samples.

Incidents

First-class incidents opened by the monitoring pipeline (kinds: availability, missed_heartbeat, anomaly, budget; lifecycle open → acknowledged → resolved). Supports connection_id, since / until and cursor pagination.

Anomalies

Statistically-detected anomalies (response times, execution failure rates and OpenClaw token spend, against 7-day baselines). Supports connection_id, since / until and cursor pagination.

Analytics in the app

How this data is visualized in the dashboard

Usage & Cost API

Push and query AI token usage and cost