Skip to content

ROADMAP — Systems Architecture Spec — Machine-Readable System Contract (SAS v0.1)

Initiative: INIT-SYSTEMSARCHITECTURESPEC-001 Repository: github.com/plexusone/systemspec-architecture

Phase 1 — Core IR & Schema Pipeline

Theme: Lock the typed graph model and the Go → JSON Schema → Zod pipeline before any tooling.

  • [ ] RMI-SYSTEMSARCHITECTURESPEC-001 Core graph types: Architecture, Node (kind taxonomy, technology, owner, criticality), Boundary with multi-membership
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-002 Relationship semantics: transport (protocol/port/encryption), operations vocabulary with generic-verb mappings, data classification, critical path
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-003 Identity and entitlement types: identity type/mechanism, subject-action-resource-conditions entitlements
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-004 Namespaced typed extensions and external reference model (openapi, pidl, threat-model-spec, multi-agent-spec, terraform, otel)
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-005 Schema pipeline: generate JSON Schemas from Go types (invopop/jsonschema), schemakit lint camelCase, go:embed, tools.go guard, Go/JSON/Zod round-trip fixtures
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-001
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-002
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-003
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-004

Phase 2 — Views, Profiles & Validation

Theme: Prove optional-in-core, required-by-profile on the real portfolio, not synthetic fixtures.

  • [ ] RMI-SYSTEMSARCHITECTURESPEC-006 View model: views as queries (includeKinds/includeRelations/includeBoundaries, groupBy, abstraction level)
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-005
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-007 Profile model: development, deployment, security, threat-model, sre profiles declaring required semantics
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-005
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-008 Validation rules engine: referential integrity, tier0 requirements, boundary-crossing authn/encryption rules, profile-conditional severity
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-007
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-009 sas CLI foundation: cobra scaffold, sas validate with profile selection, JSON and console output
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-008
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-010 Fixture corpus and dogfood architectures: valid/invalid examples plus real portfolio systems (horizontal forge app and vertical web app) modeled end to end
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-008

Phase 3 — Rendering & Catalogs

Theme: First payoff — diagrams for the portfolio; one model, many correct projections.

  • [ ] RMI-SYSTEMSARCHITECTURESPEC-011 Mermaid renderer: view-driven flowchart output with boundary grouping
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-006
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-012 D2 renderer: primary generated-diagram target with layout and grouping hints
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-006
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-013 Graphviz DOT renderer for large-graph auto-layout
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-006
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-014 Technology catalogs: AWS, GCP, Kubernetes kind/service mappings, operation specializations, display labels
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-015 sas view command: render any view in any format; three distinct projections of the dogfood architectures verified
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-011
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-012
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-013
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-010

Phase 4 — Threat Modeling, PIDL & Launch Readiness

Theme: Second payoff — going-live security: PIDL bindings, threat-model handoff, launch-readiness checks. Closes v0.1.

  • [ ] RMI-SYSTEMSARCHITECTURESPEC-021 ProtocolBinding: bind PIDL protocol participants to concrete SAS nodes; sas bind resolution checks
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-005
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-026 Threat Model Spec bridge: export system-under-analysis (nodes, boundary crossings, data flows, identities) consumable by threat-model-spec
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-005
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-027 Launch-readiness validation rules: internet-facing exposure checks (public endpoints require authn, TLS, owner, data classification)
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-008
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-022 Assurance references and coverage: per-node/edge test/metric/detection/deployment refs, sas assure coverage report
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-008
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-028 Launch dogfood: model and threat-model the first live web properties end to end (vertical apps going to production)
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-015
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-026
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-027
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-025 v0.1 release: SPEC.md normative document, README with statically-typed-friendly positioning, CI (test/lint/schema-lint/conformance), tag
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-015
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-021
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-022
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-028

Phase 5 — Semantic Diff & Change Impact

Theme: Architecture change as typed, classifiable data — post-v0.1 (v0.2 candidate).

  • [ ] RMI-SYSTEMSARCHITECTURESPEC-016 Semantic diff engine: typed ChangeSet with node/edge/boundary and field-level deltas
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-005
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-017 Change impact classification: new external dependency, authn change, entitlement expansion, new boundary crossing, data classification change
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-016
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-018 Baseline and assessment model: approved immutable baselines, proposed versions, recorded human decisions with evidence
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-016
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-019 FedRAMP change-assessment profile: three-outcome classification with control-mapping hooks and decision-support language
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-017
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-018
  • [ ] RMI-SYSTEMSARCHITECTURESPEC-020 sas diff command: human and JSON output, CI-gate exit codes, four canonical scenarios verified on fixtures
  • Depends on: RMI-SYSTEMSARCHITECTURESPEC-019

Out of scope for this initiative: CALM interop adapters are demand-driven — CALM is not yet widely adopted and is FINOS/finance-domain-centric, so adapters will be scoped as new RMIs only if and when someone asks for CALM.