Skip to content

Target Architecture Progress Snapshot

Date: 2026-05-28 Source of criteria: docs/TARGET_ARCHITECTURE.md Evidence anchors: docs/PROJECT_STATUS.md, runtime/tooling/docs state

Method

  1. Each Definition-of-Done criterion is scored in [0.0, 1.0].
  2. Pillar % = average(criteria) * 100.
  3. Overall % = average(pillar %).

This is a planning score, not a replacement for CI pass/fail gates.

Pillar Scoring

1) Durability (Temporal-inspired)

Criteria scoring:

  1. Survive restart without semantic drift: 0.60
  2. Retry/timer/resume explicit and test-covered: 0.80
  3. Checkpoint/replay API backward compatibility: 0.80

Pillar completion: 73%

Main gaps:

  1. Durability contract baseline is now formalized and CI-verified, but workflow-history durability is not yet Temporal-grade.
  2. Process restart continuity is covered at baseline level, but not yet expanded to full canonical durable orchestration semantics.

2) Contracts, Artifacts, Promotion (Dagster/dbt-inspired)

Criteria scoring:

  1. Orchestrated reproducible pipeline stages: 0.60
  2. Traceable lineage source -> artifacts: 0.50
  3. CI/local equivalent governance outcomes: 0.80

Pillar completion: 63%

Main gaps:

  1. Pipeline orchestration is strong but still custom-script centric.
  2. End-to-end lineage is present in pieces, not yet unified as a first-class pipeline lineage model.

3) Policy and Tenancy (Kubernetes/OPA-inspired)

Criteria scoring:

  1. Externally explainable and auditable policy decisions: 0.89
  2. Tenant isolation consistency: 0.94
  3. Existing runtime safety controls preserved: 0.90

Pillar completion: 92%

Main gaps:

  1. OPA adapter wiring and tenant-scoped policy-bundle lifecycle verification are in place, including explicit environment promotion controls in manifest governance checks.
  2. same_tenant runtime enforcement, 8-capability rollout, channel-level probes, and HTTP/MCP transport probes are in place; broader registry rollout is still pending.

4) Developer Experience and Distribution (Stripe-like/Vercel-like)

Criteria scoring:

  1. Time-to-first-success measured and improved: 0.55
  2. Docs/examples parity-validated: 0.78
  3. Environment promotion flow documented/reproducible: 0.65
  4. OSS baseline without paid SaaS: 1.00

Pillar completion: 75%

Main gaps:

  1. DX metrics trend and initial SLO evaluation are now in place, but thresholds still need progressive hardening.
  2. Preview/promote/rollback environment UX is now formalized at baseline; remaining work is operational polish and broader adoption evidence.
  3. Cross-workflow release gate and lineage are implemented at baseline; remaining work is strategic consolidation depth rather than missing core controls.

Overall Completion

Overall target architecture completion: 75%

Computation:

  1. Durability: 73%
  2. Contracts/artifacts/promotion: 63%
  3. Policy/tenancy: 92%
  4. DX/distribution: 75%
  5. Overall average: 76%

What Is Missing Next (Priority Order)

  1. Temporal-grade durable orchestration layer beyond current durability contract baseline.
  2. Orchestrated delivery lineage model for contract/artifact/promotion flow.
  3. OPA staged enforcement policy hardening and CI/branch-level enforcement for environment promotion controls.
  4. DX SLO hardening and deploy-flow latency instrumentation.

Notes

  1. LangGraph remains optional integration surface and is not scored as a mandatory core pillar.
  2. Score reflects strategic target alignment, not capability coverage (which is already high).
  3. Production operations baseline is now explicitly documented in docs/PRODUCTION_READINESS.md.