Data & State Architect

Enforces unidirectional data flow, immutable DTOs, and type-safe state transitions across application layers.

JIT Tool Directives (Execute this FIRST)

1. Detect Ecosystem: Check the project root for ecosystem markers (pubspec.yaml = Flutter, package.json = Node/React, Cargo.toml = Rust, go.mod = Go, *.csproj = .NET).
2. Load Ecosystem Canon: Based on detected ecosystem, load the relevant canon:
   • Flutter: canons/ecosystems/flutter/flutter-data-patterns.md
   • Other: Apply universal principles below directly.
3. Use an available file-search tool to locate data model directories (domain/, models/, entities/).

Architectural Invariants (Negative Constraints)

• FORBIDDEN (In-Place Mutation): Never mutate state objects directly (e.g., state.field = value). All updates MUST produce a fresh instance via copyWith, immutability records, or pure reducers.
• FORBIDDEN (Boundary Bleed): Never import UI/Presentation layers (widgets, controllers, HTML elements) into domain models or entities.
• FORBIDDEN (Untyped Data): Never use dynamic, any, or raw untyped dictionaries in production domain models. Enforce strict schemas (Zod, Freezed, Pydantic).
• FORBIDDEN (Over-Subscription): Never subscribe to broad root state objects in UI widgets. Enforce granular, field-level reactive selectors to eliminate redundant re-renders.

Universal Workflow Patterns

1. Selective Watch Rule: Subscribe to specific fields to prevent unnecessary UI rebuilds.
2. Async State Management: Use async state containers for async mutations, read-only providers for one-off fetches.
3. Loop Prevention: Handle futures/streams with explicit state matchers (loading/error/data).

Troubleshooting (Universal)

• Model methods not generated: Missing code generation setup or part directives.
• UI not updating on state change: Using one-shot read instead of reactive watch.
• Duplicate state containers: Multiple providers/stores with same responsibility.

🔗 Native Interconnection Pipeline

Data models and state logic do not exist in isolation. Follow the strict 4-stage pipeline:

1. Upstream Schema Feed: Ingest table schemas and data flow definitions from context/00_Strategy/BLUEPRINT.md (authored via /project-architect).
2. Contract Encasement: Pass all domain boundaries through strict schemas (Zod/Pydantic/OpenAPI via /api-contract).
3. Immutable Implementation: Produce unidirectional, immutable DTOs and pure reducers (/data-logic).
4. Downstream Verification: Feed generated state schemas directly into failing state-delta unit tests (/strict-tdd).

📚 Mandatory Knowledge Routing (JIT References)

If your current task intersects with these domains, you MUST execute view_file on the target BEFORE writing code:

Generated by OrionCortex Infrastructure - Universal Data Logic