Start manual or continuous audits and monitor the orchestrator pipeline in real time.
Who this is forOperators running day-to-day repository risk scans.
Prerequisites- At least one connected project.
- Audit quota available on your plan (see Settings → Billing).
Manual scan
- Dashboard → Run audit on a project card, or Projects → trigger from the row action.
- POST /api/audits with projectId and branch enqueues the same job programmatically.
- Poll GET /api/audits/:id for runStatus, agentRuns, and events.
Continuous audit
- Sidebar and dashboard toggles set continuousAuditEnabled for the organization.
- When OFF, no automatic audits run: only manual Run audit triggers jobs.
- When ON, idle projects rotate on a ~90s cycle; active runs block new auto-scans.
- PATCH /api/workspace/runtime updates org metadata and project autoRunOnPush flags.
Runtime controls
- Operations Runtime shows pipeline progress, agent count, and terminal logs.
- Stop all: POST /api/workspace/runtime/stop-all disables continuous audit and cancels queued/running/paused runs.
- POST /api/audits/:id/pause saves a checkpoint and sets runStatus paused (orchestrator cooperative stop).
- POST /api/audits/:id/resume re-enqueues from the last checkpoint phase.
Expected result
A completed or paused audit run with findings, clusters, and optional issue candidates.
Was this article helpful?
