For universities
Use all thirteen runnable pilot missions at ML-D depth with the existing quantitative channels, provenance, configuration, evidence-quality, limitation, and verification-and-validation prompts. There is no separate university runtime.
Inspect visible channels, their provenance roles, deterministic run hashes, and the claim each artifact can support.
Compare bounded configurations, evaluate evidence sufficiency, and state what a longer or more detailed software run still cannot establish.
Local practice has zero measured channels and is not verified recognition. Institutional and Challenge authority remain separate workflows.
Engineering surface
Configure bounded spacecraft definitions, run the shared model core, inspect Mission Control views, compare variants and export evidence with model and channel provenance visible. Mission Control is a capability inside the Digital Twin—not a separate product.
Bounded developer access
The versioned Developer API and TypeScript, Python and Jupyter source clients support explicitly enabled local/private evaluation. A stable production-public or browser API is not available, and MissionLab does not host arbitrary learner Python, JavaScript or shell execution.