Getting Started
Next steps
Progress to agents, schedules, topology, remediation, and CI integration.
Overview
After your first successful run, mature teams expand along four axes: automation (agents and schedules), integration (CI/CD and webhooks), observability (Test Health, reports, Topology), and governance (release gates and remediation).
The paths below are ordered by typical enterprise adoption, adjust pace based on compliance requirements and team size.
Step-by-step procedure
Week 1: Stabilize core workflow
Invite team members and assign roles via Admin Center.
Organize cases into suites for smoke and regression.
Configure notification routing in Settings.
Week 2: Automate execution
Create schedules for nightly regression on staging.
Enable relevant AI testing agents approved for your organization.
Register cloud or endpoint agents if desktop or restricted workloads apply.
Week 3: Integrate pipeline
Generate API keys in Admin → Developer.
Add Zof validation step to GitHub Actions, GitLab CI, or Jenkins.
Configure webhooks for run completion notifications.
Month 2: Governance
Configure release gates in Governance → Releases.
Explore Topology for dependency-aware coverage planning.
Evaluate governed remediation workflows if enabled for your organization.
Expand by risk and ownership
After a first successful exercise, select the next workflow based on business impact and current uncertainty. Add coverage where a failure would change a real decision rather than increasing the test count without a purpose.
Assign an owner to each added workflow, its inputs, environment, and follow-up work. Include failure handling and evidence retention in the rollout plan.
Review readiness before production reliance
Confirm access, execution placement, integration permissions, approvals, and operational support with the responsible teams. Test a representative failure and escalation path before a release depends on the outcome.
Document unresolved assumptions and the evidence needed to close them. A successful pilot demonstrates the pilot conditions; it does not automatically validate a different topology, workload, or sensitive-data category.
Was this page helpful?