docs: module dossier + install/quickstart sync for truthful cutover sprints

- API_CLI_REFERENCE.md, INSTALL_GUIDE.md, quickstart.md, architecture/integrations.md, dev/DEV_ENVIRONMENT_SETUP.md, integrations/LOCAL_SERVICES.md: reflect real-service wiring.
- docs/modules/**: module dossier updates across the modules touched by SPRINT_20260415_001..007 + SPRINT_20260416_003..017 + SPRINT_20260417_018..024 + SPRINT_20260418_025 + SPRINT_20260419_026.
- docs/features/checked/web/**: update feature notes where UI changed.
- docs/qa/feature-checks/runs/web/evidence-presentation-ux/: QA evidence artifacts.
- docs/setup/**, docs/technical/**: align with setup wizard contracts.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

docs/modules/advisory-ai/architecture-detail.md

@@ -49,6 +49,7 @@ Key stages:
 | `GuardrailService` | Applies redaction filters, prompt allowlists, validation schemas, and DSSE sealing. | Shares configuration with Security Guild. |
 | `ProfileRegistry` | Maps profile IDs to runtime implementations (local model, remote connector). | Enforces tenant consent and allowlists. |
 | `AdvisoryOutputStore` | PostgreSQL table storing cached artefacts plus provenance manifest. | TTL defaults 24h; DSSE metadata optional. |
+| `PostgresAiConsentStore` / `PostgresAiAttestationStore` | Durable runtime state for tenant consent and AI attestations. | Bound by `AddAdvisoryAiRuntimePersistence`; fail-fast outside Development/Testing when no DB is configured. |
 | `AdvisoryPipelineWorker` | Background executor for queued jobs (future sprint once 004A wires queue). | Consumes `advisory.pipeline.execute` messages. |
 
 ## 3. Data contracts
@@ -132,6 +133,16 @@ Profile selection is controlled via Authority configuration (`advisoryAi.allowed
 
 Cache misses trigger orchestration and inference; hits return stored artefacts immediately. TTL expiry removes entries unless `forceRefresh` has already regenerated them.
 
+Additional live runtime tables:
+
+| Table | Purpose |
+|-------|---------|
+| `advisoryai.ai_consents` | Authoritative per-tenant, per-user consent records for AI execution and remote-profile opt-in. |
+| `advisoryai.ai_run_attestations` | Durable run-level attestation payloads, content digests, and optional signed envelopes. |
+| `advisoryai.ai_claim_attestations` | Durable claim-level attestation payloads keyed by claim and linked back to their parent run. |
+
+Runtime connection precedence for these tables is `AdvisoryAI:Storage:ConnectionString` -> `ConnectionStrings:Default` -> `Database:ConnectionString`. The web host auto-migrates this schema on startup and refuses live startup outside Development/Testing when no durable store is configured.
+
 ## 7. Telemetry & SLOs
 
 Metrics (registered in Observability backlog):

docs/modules/advisory-ai/architecture.md

@@ -107,6 +107,9 @@ All context references include `content_hash` and `source_id` enabling verifiabl
   - `generated_at`, `model_id`, `profile` (Sovereign/FIPS etc.).
   - `signatures` (optional DSSE if run in deterministic mode).
 - Offline bundle format contains `summary.md`, `citations.json`, `context_manifest.json`, `signatures/`.
+- Runtime consent and attestation state is now durable under `advisoryai.ai_consents`, `advisoryai.ai_run_attestations`, and `advisoryai.ai_claim_attestations`.
+- The remaining mutable runtime state is also durable under `advisoryai.runtime_explanations`, `advisoryai.runtime_policy_intents`, `advisoryai.runtime_runs`, `advisoryai.advisory_chat_settings`, `advisoryai.conversations`, and `advisoryai.turns`.
+- Runtime storage connection resolution is `AdvisoryAI:Storage:ConnectionString` -> `ConnectionStrings:Default` -> `Database:ConnectionString`; outside `Testing` the host fails fast instead of silently falling back to process-local consent, explanation, policy, run, conversation, or chat-settings state.
 
 ## 7) Profiles & sovereignty
 
@@ -136,9 +139,11 @@ All endpoints accept `profile` parameter (default `fips-local`) and return `outp
 
 ## 11) Hosting surfaces
 
-- **WebService** — exposes `/v1/advisory-ai/pipeline/{task}` to materialise plans and enqueue execution messages.
-- **Worker** — background service draining the advisory pipeline queue (file-backed stub) pending integration with shared transport.
+- **WebService** - exposes `/v1/advisory-ai/pipeline/{task}` to materialise plans and enqueue execution messages.
+- **Worker** - background service draining the advisory pipeline queue (file-backed stub) pending integration with shared transport.
 - Both hosts register `AddAdvisoryAiCore`, which wires the SBOM context client, deterministic toolset, pipeline orchestrator, and queue metrics.
+- `advisory-ai-web` registers `AddAdvisoryAiRuntimePersistence`, and `advisory-ai-worker` registers `AddAdvisoryAiCoreRuntimePersistence`; together those paths auto-migrate the `advisoryai` schema and bind durable PostgreSQL-backed consent, attestation, explanation replay, policy-intent, run, conversation, and chat-settings stores whenever runtime database configuration is present.
+- In-memory runtime-state implementations are now reserved for explicit test harnesses or `Testing` fallback. Non-testing hosts fail fast when the runtime database contract is missing.
 - SBOM base address + tenant metadata are configured via `AdvisoryAI:SbomBaseAddress` and propagated through `AddSbomContext`.
 
 ## 12) QA harness & determinism (Sprint 110 refresh)