Hello Rick Huebner,
Thank you for posting your question in the Microsoft Q&A forum.
This rejection highlights a critical conflict between Azure Marketplace's strict validation policies and legitimate infrastructure-as-code security practices. Your approach of marking sensitive outputs as secureString is technically correct in Bicep and ARM to prevent credential exposure in logs.
However, the Marketplace's validation scripts appear to interpret any secure output as a prohibited secret, regardless of its actual necessity or context. To resolve this, you must explicitly demonstrate that these outputs are essential runtime identifiers like resource names or connection strings rather than concealed credentials.
You may engage directly with the Marketplace onboarding team to clarify the requirement, providing a justification for each secure output explaining its operational necessity. They may require you to refactor, potentially moving these values into a child resource or adjusting their sensitivity designation, even if less ideal, to comply with their automated checks while maintaining security through alternative means like managed identities.
Please, let me know the response helps answer your question? If the above answer helped, please do not forget to "Accept Answer" as this may help other community members to refer the info if facing a similar issue. 🙂