Testing

Test generation

Generate scenarios and test cases from requirements with human review.

Overview

Test generation transforms structured requirements into scenarios and executable test cases through governed AI-assisted workflows in Zof Console. Generation accelerates inventory creation; it does not remove human review before release-critical reliance.

The generation pipeline respects selected test types, organizational policy packs, and linked specification content. Output quality correlates directly with requirement clarity, invest in acceptance criteria before tuning generation parameters.

Operators should expect iterative passes: initial generation for breadth, targeted regeneration after requirement updates, and manual library edits for domain nuance automation cannot infer.

Who should read this

  • QA engineers, SREs, platform teams, and developers operating Zof Console and APIs.

Prerequisites

  • Linked or uploaded requirements with requirement IDs
  • Test types selected for the project aligned to risk profile
  • Review permission for scenarios and cases in the test library

When to use this workflow

  • Onboarding new team members to Zof terminology and workflows
  • Authoring internal runbooks aligned with Console labels
  • Designing CI/CD or webhook integrations against documented behavior

Governed generation pipeline

Governed generation pipeline

Generation is intentionally gated by review before execution enters production paths.

Step-by-step procedure

Confirm prerequisites

Verify requirements processing completed and test types reflect compliance needs (not every enabled type by default).

Open the project or specification record to confirm the correct document revision is linked.

Align stakeholders on review SLA before generation at enterprise scale.

Generate scenarios

Start scenario generation from the project wizard or project detail Generation area.

Review scenario titles for naming convention SCN-<REQID>-<short-slug> and reject orphans without requirement mapping.

Merge overlapping scenarios before case generation to reduce duplicate cases.

Generate test cases

Trigger case generation per scenario or in batch according to UI workflow.

Inspect steps for observable actions and expected results, edit vague "verify correct behavior" language.

Set priority P0-P3 consistent with release gating policy (P0 smoke, P3 edge).

Human review and approval

Mark cases reviewed only after domain expert validation, treat unreviewed AI output as draft.

Record review notes in change tickets when regulated industries require evidence.

Return cases to draft status when major product changes invalidate steps.

Organize into suites

Group reviewed cases into smoke and regression suites for Operate schedules and CI triggers.

Cap smoke suite duration for pipeline friendliness, defer breadth to scheduled regression.

Publish suite-to-case ID mapping for service teams consuming APIs.

Validate with pilot runs

Execute staging smoke runs before adding cases to production release gates.

Compare results against manual exploratory findings for the same requirement IDs.

Feed failures back into requirements or case edits, not repeated blind regeneration.

Key concepts

Organization scope
All Zof Console and API operations are isolated to your authenticated tenant.
Governed execution
Agent output and remediation follow policy packs with human approval when configured.

Best practices

  • Regenerate narrowly after requirement diffs instead of full-project regeneration when possible
  • Keep a golden set of manually authored P0 cases alongside generated inventory
  • Log generation batch IDs in change records when auditing AI-assisted testing
  • Reject scenarios that reference requirement IDs not in the linked specification

Common issues

Generation stalls or returns empty output
Insufficient requirement text, wrong test types, or processing incomplete. Add acceptance criteria detail and retry.
Duplicate cases across scenarios
Overlapping requirements or repeated generation passes. Merge duplicates and tighten scenario boundaries.

Was this page helpful?

Test generation | Zof AI Documentation