Use Cases
- Daily sign-in monitoring — Check for brute force patterns every morning
- Weekly threat hunting — Search for lateral movement indicators across your environment
- Monthly compliance checks — Verify security baselines and audit log completeness
- One-time delayed investigation — Schedule an investigation to run at a specific future time
Creating a Schedule
From the UI
Navigate to Schedules in the sidebar and click Create Schedule. Fill in the investigation query, persona, frequency, and preferred run time.From the API
Validate the query
Before creating a schedule, validate that your query is specific enough for recurring execution:The validation checks that the query has enough specificity (concrete entities, clear time scoping, measurable conditions) for consistent recurring execution.
Frequency Options
| Frequency | Description | Example |
|---|---|---|
once | Runs a single time at the scheduled time | One-time overnight scan |
daily | Runs every day at the scheduled time | Morning sign-in review |
weekly | Runs once per week | Weekly threat hunt |
monthly | Runs once per month | Monthly compliance check |
custom | Raw cron expression for advanced scheduling | 0 */6 * * * (every 6 hours) |
For
once schedules, the status automatically changes to COMPLETED after the single run finishes. All other frequencies continue running until manually paused or deleted.Schedule Lifecycle
Managing Schedules
List Schedules
Update a Schedule
Modify the query, frequency, or scheduled time. The next run time is automatically recalculated.Trigger an Immediate Run
Run a scheduled investigation immediately without waiting for the next scheduled time:Delete a Schedule
Soft-deletes the schedule. It will no longer run.Monitoring Runs
View Run History
See the last 50 runs for a schedule, including status, duration, and linked investigation sessions:Run Statuses
| Status | Description |
|---|---|
running | Investigation is currently in progress |
completed | Investigation finished successfully |
failed | Investigation encountered an error |
How It Works
Scheduled investigations run as automated background tasks:- Checking — SoCMate checks every minute for schedules past their
next_run_attime - Triggering — For each due schedule, a new investigation session is started automatically
- Monitoring — Run statuses are checked every 5 minutes for completed or failed runs
- Notifications — Completion and failure notifications are sent through the notification system
- Next run — After each run, the
next_run_atis recalculated based on the frequency
Notifications
Scheduled investigations generate notifications on:- Completion — Investigation finished with results
- Failure — Investigation encountered an error
Failure Handling
- Individual run failures are recorded in the run history with error details
- After 5 consecutive failures, the schedule is automatically disabled with
status=FAILED - Failed schedules can be re-enabled after resolving the underlying issue
- Common failure causes: Sentinel connectivity issues, query timeout, LLM provider errors
