Rule groups
A rule group is the unit Measured runs and scores. It collects rules, which live on individual data assets, into one runnable set that can span a single asset or many. Running a rule group produces a result and a health score, and can be triggered manually, on a schedule, via the API, or by your agents over MCP.
Coming from the older model? Rule groups replace checkpoints. Rules now attach directly to a data asset, and a rule group references the rules to run together.
Anatomy of a rule group
Rule group
├── Members (rules drawn from one or more data assets)
├── Health score (computed from the latest run)
├── Schedule (optional cron expression)
└── Remediation (optional alerts and incidents)
Creating a rule group
Via the UI
- Author rules on a data asset (accept AI suggestions or add your own).
- Go to Rule Groups → New rule group.
- Add members, rules from one or more assets.
- (Optional) Attach a schedule and remediation actions.
- Run now, or let the schedule drive it.
Via the API
curl -X POST https://app.measured.cloud/api/v2/organizations/$ORG/workspaces/$WS/rule-groups \
-H "Authorization: Bearer $MEASURED_TOKEN" \
-H "Content-Type: application/json" \
-d '{ "data": { "name": "Portfolio quality" } }'
Scheduling
Rule groups use standard cron syntax (5 fields, UTC):
| Schedule | Expression |
|---|---|
| Every hour | 0 * * * * |
| Every 6 hours | 0 */6 * * * |
| Daily at midnight UTC | 0 0 * * * |
| Weekdays at 8am UTC | 0 8 * * 1-5 |
Health score
Every run scores the rule group from its pass/fail and severity mix, so you can track posture over time and surface it on the dashboard. A failing critical rule weighs more heavily than a failing informational one.
Remediation
Route results to action when a run fails:
- Alerts, fire-and-forget to Email, Slack, Teams, PagerDuty, Opsgenie, ServiceNow, Jira, webhook, and more.
- Incidents, open a tracked incident with an owner, severity, timeline, and resolution.
The internal Create incident action sits at the top of the action list, with external channels below it.
Triggering from CI/CD and agents
Run a rule group from your orchestrator, CI/CD, or an LLM agent via the API and MCP, for example, as a gate after a dbt build, blocking the pipeline if data quality drops.
History
The Results page shows a timeline of every run with pass/fail and health. Open any run for per-rule results, failure counts, and an AI root-cause summary.