Language and runtime
Read the public MIND documentation for the Core v1 surface, compiler/runtime boundaries and current roadmap.
Open mindlang.devNAESTRO / DOCUMENTATION
The fastest way to understand NAESTRO is to follow one real question through the system. These public entry points explain the architecture, the language/runtime, persistent memory and evidence without requiring a private deployment.
See how a trajectory holds continuing work.
Trace component roles and their public references.
Distinguish supported behavior from the roadmap.
Read the public MIND documentation for the Core v1 surface, compiler/runtime boundaries and current roadmap.
Open mindlang.devThe language repository and Core specification are the source of truth for behavior that an integration depends on.
Open the Core specificationFollow the proposed IntentSpec to Contract IR boundary and its explicit refusal path in the architecture model.
Read the MIND-Intent roleSee how governance and invariant checks are positioned after establishment and before an admitted action.
Read the 512-MIND roleUse the ecosystem map to distinguish repository and component structure from semantic proof or runtime evidence.
Read the Arch-MIND roleInspect the public memory project for setup, usage, governance and the limits of its retrieval and evidence claims.
Open MIND-MemBring a workflow, its constraints and the decision that deserves review. The next step is a scoped conversation, then a small inspectable trajectory.
Discuss your workflowThe website explorer exposes an illustrative trajectory as a readable JSON sample record. It describes the example’s steps and outcomes for inspection.
MIND’s compiled representation is mic@3, a canonical binary encoding of compiler IR. Its optional MAP (Metadata Attachment Partner) section carries metadata, including evidence when emitted. The proposed mic@4 successor remains roadmap work.
A sample trajectory record and a compiled MIND artifact serve different purposes. Artifact verification applies to the actual compiler output and the evidence it contains.
Read the canonical IR specificationBring a real workflow. Start with a bounded question.
Discuss an evaluation