Generated Test Cases
Understand generated coverage maturity, platform/core ownership, and system-type boundaries.
Current support
Generated Test Cases expand a testing brief into reusable coverage artifacts. They are an advanced workflow, not the default first-run path.
In the public CLI, older generation command shapes are still present for compatibility, but the product model is Test Cases. If an installed package reports the command as hosted-only, use the platform workflow rather than treating generation as a local smoke test.
| System Type | Current generated support |
|---|---|
| Recommender | Strongest path: generated scenarios, generated personas, and test case execution through the platform/core workflow. |
| Search | Generated scenarios and personas feed audit or compare coverage. Search test case execution parity is not claimed here. |
| Agents | Generated task/environment packs and sidecars are review artifacts, not generated customer agents. |
Recommender and search packs
The command shapes are:
determina generate-scenarios --system-type search --mode fixture --brief "fresh typo and ambiguous query behavior" --output-dir ./search-generated determina generate-personas --system-type search --mode fixture --brief "freshness-sensitive and research-heavy users" --population-size 4 --output-dir ./search-generated
Generated packs write sidecars such as generated_coverage_plan.json and generated_coverage_manifest.json. These are compatibility file names; they explain test case provenance, selected cases, validation status, hashes, and deferred cases when generation is available in the configured workflow.
Agent packs
Agents use generated local task/environment packs instead of generic scenario/persona packs.
determina generate-agent-pack --brief "support agent current-info and refusal coverage" --output-dir ./agent-generated