Overview

Agents in this category extend governed validation for specialized quality dimensions. Enable only categories aligned to your application risk profile and review AI-generated output before production gates depend on results.

Enable agents per organization in Automation → AI Agents. Human review applies to generated output before release-critical use.

Who should read this

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

Prerequisites

Organization administrator approval to enable agent category
Staging or approved test environment reachable from execution plane
Requirement linkage for audit-oriented teams (recommended)

When to use this workflow

Scheduled validation aligned to release cadence
Targeted runs after changes in related system components
Audit evidence collection for quality reviews

Step-by-step procedure

Review catalog entry

Automation → AI Agents → browse category manifest and capability labels.

Confirm runtime targets (cloud vs endpoint) match your deployment model.

Enable for organization

Toggle enablement after administrator approval.

Document ownership for case review and triage.

Integrate with runs

Add category cases to appropriate suites.

Monitor results in Operate → Runs and Quality → Test Health.

Key concepts

i18n validation scope: Category-specific behaviors documented in agent manifest; Policy-compliant execution on approved environments; Repeatable cases linked to requirements where applicable.
Check 1: Category-specific behaviors documented in agent manifest
Check 2: Policy-compliant execution on approved environments
Check 3: Repeatable cases linked to requirements where applicable

Best practices

Human review required for AI-generated tests before release-critical reliance
Scope agents to relevant applications rather than enabling entire catalog
Use Test Health quarantine for flaky cases instead of disabling agents globally