FICTIONAL REVIEW SCENARIOS / USAGE ANALYST
Test the actual service.
Five successful workflows and three invalid-input checks, using only fictional data.
Each scenario calls the production MCP endpoint. A negative case passes when the service rejects the invalid input. These checks verify service behavior; they do not simulate a ChatGPT conversation or prove native mobile compatibility.