Principles, Requirements, Governance & Change

The Preliminary phase's principles and tailoring, the requirements catalog that sits at the centre of the wheel with traceability to components and gates, the risk register, and Phases G and H expressed as what the platform itself enforces.

Preliminary · Architecture principlesSix absolute principles, four supporting — each with rationale and implications

absolute cannot be waived by configuration or by a reviewer · supporting may be tailored per domain with a recorded decision.

P01 · Subgraph, never the schema absolute

StatementGeneration agents receive a retrieved, cited subgraph (≤30 tables by default), never the full schema.

RationaleSchema size must not change the cost or precision of a use case; models pick wrong identifiers from large candidate sets.

ImplicationsSemantic catalog is a prerequisite; groundedness gate rejects uncited tables; subgraph budget enforced by the supervisor.

P02 · Contracts, not generated code absolute

StatementAgents emit schema slices and UI specs; resolvers are compiled, screens are rendered. Code generation exists only as a reviewed exception into a registry.

RationaleA thousand generated codebases cannot be secured or audited; contracts can be verified deterministically.

ImplicationsDSL and compiler are platform assets; escape-hatch rate is a tracked metric; no agent has a code-execution tool.

P03 · Authorization declared once, compiled in absolute

StatementClassified fields carry a policy on the schema; entitlements are composed into the SQL from validated claims before execution. Post-filtering is prohibited.

RationaleA control every workflow must remember is not a control; recall collapses under post-filtering for narrow-permission users.

ImplicationsPolicy catalog as data; 100% coverage gate; compiler refuses roles it cannot scope; edge never holds business authorization.

P04 · Reads publish, writes propose absolute

StatementNo generated artifact writes to a system of record directly. Write actions create proposals; approval is enforced by the platform from the registration.

RationaleThe structural bound against prompt injection is the write path, not a classifier.

ImplicationsProposal service is mandatory; PUBLISH_WRITE unreachable without an approval event; autonomous-safe targets are a policy decision, not a composer decision.

P05 · The release unit is the whole tuple absolute

StatementCatalog version, slice, UI spec, policy version, model deployment, prompt template versions and component versions are released and pinned together.

RationaleA change to any element changes behaviour; a provider rolling a model is a release you did not perform.

ImplicationsModel versions pinned; drift triggers re-verification; reviewers approve a tuple, not a file.

P06 · A human approves first publish and every write absolute

StatementGates decide what can be measured; a person decides the rest, on evidence, for every first publish and every write path.

RationaleJudgment is the scarce resource; the architecture exists to spend it on approving contracts rather than writing them.

ImplicationsReview workspace is core, not optional; auto-publish applies only to read-only re-publishes and only by registration.

P07 · Generation never touches production data supporting

StatementAgents see catalog cards, masked profiles and synthetic sample data. Runtime is the only path to production rows.

ImplicationsSandbox environment; PII-aware profiling; environment rule = data rule.

P08 · Deterministic wherever possible supporting

StatementSupervisor, compiler and gates are deterministic; the model contributes proposals and prose, never decisions.

ImplicationsTyped events only; same input → same SQL; gate results are not overridable by model output.

P09 · Refuse rather than guess supporting

StatementAmbiguity in requests or schema is returned to a person; retries are capped; dead ends park.

ImplicationsClarification loops in intake; ambiguity events from the Scout; glossary learns from resolutions.

P10 · Everything traceable to its sources supporting

StatementEvery table cites a card; every screen names its release unit; every query is logged with its predicate set.

ImplicationsTrace store is a first-class component; any bad screen is replayable into a golden case.

Tailoring decisions (Preliminary)

  • Phases A–D produce diagrams and catalogs for a product; no enterprise-wide landscape is attempted.
  • Phases G and H are realised by the platform's own gates and drift handling rather than a separate governance board — the platform is the compliance mechanism.
  • Requirements Management uses one catalog with IDs traced to components and gates; no separate requirements tool.
  • Building blocks: ABBs are the capabilities (B1); SBBs are the components (C5) and technology choices (D4).
  • Iteration: the ADM is run once per transition architecture (TA-0 … TA-4), re-entering at Phase E each time.

Requirements Management · Requirements catalogFunctional and non-functional requirements with traceability to components and gates

IDRequirementTypeRealised byVerified byPriority
FR-01A prose request becomes a typed use-case spec, or a clarifying question, or a refusalFunctionalIntake service · archetype librarySpec schema validation · M1 exitMust
FR-02Retrieval returns a subgraph ≤30 tables with join paths, each citing a catalog cardFunctionalCatalog service · ScoutGroundedness gate · M0 precision/recallMust
FR-03Composer emits a federation-compatible slice with policies on every classified field and persisted operationsFunctionalContract composer · policy serviceComposition gate · policy coverage gateMust
FR-04UI composer emits a valid UI spec bound only to persisted operations; unexpressible needs become custom-component requestsFunctionalUI composer · DSL validatorBinding validity gate · a11y gateMust
FR-05Write actions are proposals; approval enforced from the registrationFunctionalProposal service · supervisorAbsence test: no direct write path in CIMust
FR-06Reviewer sees spec, subgraph rationale, SDL diff, policy and gate reports, and a live preview on sample dataFunctionalReview workspace · artifact storeM4 exit walkthroughMust
FR-07Catalog drift re-verifies affected use cases automatically and degrades on failureFunctionalCrawler · event workers · registryM5 exit: renamed-column scenarioMust
FR-08Custom components reach the registry only through sandboxed build, scan, screenshot diff and human reviewFunctionalSandbox runner · registry UIM5 exit: escape-hatch scenarioShould
NFR-01Zero unscoped queries reachable at runtime; every operation compiles scoped for every actor roleSecurityQuery compiler · policy serviceEntitlement-compile gate · CI absence testMust
NFR-02100% of classified fields carry a policy; protected fields nullable; deny-by-defaultSecurityContract composerPolicy coverage gateMust
NFR-03Only persisted operations are executable; query text is rejected at the edge/gatewaySecuritySupergraph gateway · persisted storeM2 exit: unknown id rejectedMust
NFR-04Generation never reads production data; sandbox holds synthetic data onlySecurity / privacyEnvironment design (D2)Network policy review · PII scan gateMust
NFR-05Any screen's SQL is reproducible from the trace, with its predicate setAuditabilityTrace store · compiler loggingM6: reproduction drillMust
NFR-06Persisted operations run under a cost ceiling with actual arguments; N+1 batchedPerformanceQuery compiler · cacheCost & complexity gateMust
NFR-07Response cache cannot serve across entitlement fingerprints and busts on catalog generation changeSecurity / correctnessResponse cache keyingGolden tests: narrow-permission user casesMust
NFR-08Per-run budgets (steps, tokens, wall-clock, cost) fail closedCostSupervisor · LLM gatewayM1 exit: budget exhaustion parksMust
NFR-09Machine time per use case in minutes; reviewer throughput ~10/hour (targets)Performance / usabilityArchetypes · review workspaceM6 metricsShould
NFR-10Screens meet accessibility baseline (labels, contrast, keyboard order, table semantics)AccessibilityComponent registry · DSL lintAccessibility gateShould
NFR-11Model deployments pinned; upgrades are dated releases re-verified against golden setsOperabilityLLM gateway · release unitGolden tests on model changeMust
NFR-12Every artifact and decision is versioned; nothing hard-deletedAuditabilityRegistry · platform DBCRUD matrix review (C4)Must

Risk registerTop risks, likelihood × impact, mitigation and owner

IDRiskL × IMitigation (structural first)OwnerResidual
R-01Retrieval selects a wrong join path; screen shows plausible wrong numbersM × HQuery-log-mined paths ranked above inferred; golden tests per archetype incl. identifier queries; review preview on sample dataPlatform engLow-medium
R-02Classification gaps let a sensitive column ship without a policyM × HClassification coverage is a prerequisite gate per domain; PII scan on sample results; deny-by-default for unknown fieldsData owners · SecurityLow
R-03Prompt injection via catalog descriptions or request text steers a composerM × MAgents have no write or execution tools; typed outputs; gates are deterministic; write path unreachable without approvalSecurityLow
R-04Escape-hatch usage climbs; governance load moves to code reviewM × MEscape-hatch rate tracked; DSL grows by adding components; registry reusePlatform engMedium
R-05Reviewer throughput becomes the bottleneckH × MBy design; evidence bundle optimised; auto-publish for read-only re-publishes by registrationDomain ownersAccepted
R-06Generated queries overload systems of recordM × HCost ceilings with real arguments; batching; entitlement-aware cache; read replicasPlatform eng · DBAsLow
R-07Model provider changes behaviour silentlyM × MPinned deployments; upgrade as a dated release; golden sets rerunPlatform engLow
R-08Supergraph composition slows or breaks at ~1,000 slicesL × MNamespaced, small slices; incremental composition; retirement schedulePlatform engLow
R-09Ambiguous enterprise schemas stall intake with repeated questionsH × LGlossary learns from resolutions; owner backlog of naming decisions surfaced as a reportData ownersAccepted
R-10Reference-implementation numbers are read as production claimsM × MEvery artifact labelled "target / reference"; metrics published with caveats at M6AuthorLow

Phase G · Implementation governanceCompliance checkpoints are the platform's gates; the architecture contract is the registration

Compliance checkpoints (automated)

  • Composition — no conflicts, no breaking change to live operations
  • Policy coverage 100% — every classified field, nullable, deny-by-default
  • Entitlement compile — every operation scoped for every actor role
  • Cost & complexity — under ceiling with real arguments; batched
  • Binding validity — every UI binding and action resolves
  • Golden tests — archetype-derived, on sample data
  • PII exposure scan — sample results vs. classification
  • Accessibility lint — DSL-level
  • Groundedness — every table and join cites a card

Failures route back at most twice, then park. Gates never auto-override.

Compliance checkpoints (human)

  • First publish of any use case — domain owner
  • Every write path — owner of the target system (e.g. Finance)
  • Every custom component — platform engineering
  • Every domain enablement — security (classification coverage) and data owner (descriptions reviewed)

Decisions are immutable and reference the release unit they approved.

The architecture contract = the registration

A use case's registration is the contract between the owning team and the platform: actors, allowed actions, autonomous-safe targets (never for regulated data), auto-publish rules for read-only re-publishes, review routing, budgets. The platform enforces it; the team cannot bypass it per screen.

Dispensations

Absolute principles P01–P06 admit no dispensation. Supporting principles may be tailored per domain with a recorded decision, an expiry, and a re-verification on expiry. Every dispensation is a row in the platform DB, not an email.

Phase H · Architecture change managementChange triggers, the automatic response, and when a human is called

TriggerDetected byAutomatic responseHuman involvementState
Schema drift (table/column added, renamed, dropped; FK change)Crawler diff → new catalog version → eventEvery use case whose subgraph touches the change is re-verified (gates only). Pass → re-pinned. Fail → degraded: banner, slice pinned to last passing version, owner pagedOwner fixes or re-runs generation; reviewer approves the new release unitlive / degraded
Policy change (classification or entitlement)Policy version eventAffected slices re-checked for coverage and entitlement compile; failures degradeSecurity signs the policy version; domain owners re-approve affected write pathsas above
Model deployment retirement / upgradeLLM gateway deployment registryGolden sets re-run on the candidate deployment; release units re-pinned only on passPlatform engineering schedules the dated releasescheduled release
Prompt template or archetype changeTemplate registry versionSeeded use cases regenerated in sandbox and diffed; regressions block the templatePlatform engineering reviews diffsgated
DSL / runtime versionRuntime releaseRuntime supports N-1 spec versions; specs migrated by a validated transformPlatform engineering; domain owners informedcompatibility window
Component version updateRegistry eventScreens referencing the component re-verified (bindings, a11y, screenshot diff)Reviewer approves visual change if diff exceeds thresholdgated
Use-case retirementUsage telemetry below threshold for N periodsProposed for retirement; slice removed from supergraph on approval; release unit archivedOwner confirmsretired
New domain enablementRequestClassification coverage report; owner-review queue for descriptionsSecurity + data owner sign; ADM re-entered at Phase E for the domainenabled

The Architecture Development Method is re-entered at Phase E for every transition architecture and for every new domain; Phases A–D are revisited only when an absolute principle would change — which is a new architecture, not a change.