01 — Live Substrate Readiness (Hard Gate 0)

All values via read-only query_pg @ directus, role context_pack_readonly, 2026-06-02. No mutation.

1.1 Hard Gate 0 result — PASS

Check	Result
All Phase-1 substrate objects exist	✅ 8 tables + 6 views present (§1.2)
Candidate/owner table row counts	✅ all 0 except scope=6 (§1.2)
Governance event types active=false	✅ 5 governance types, 0 active
event_outbox governance emit	✅ 0
Worker cursor state	✅ gov_worker_cursor=0 rows (no tail position)
Approval / e-sign counts	✅ approval_requests=211, apr_approvals=42, os_proposal_approvals=0
dot_tools count	✅ 309 (no DOT registered)
Idle transactions	✅ idle_in_transaction=0
Ownership seam objects/views	✅ all present; both integration seams WHERE false (0 rows)
Source inventories available	✅ all present (§1.4)

1.2 Substrate object/row census

Tables (8): governance_build_authorization=0 · governance_ruleset=0 · gov_worker_cursor=0 · governance_candidate_state=0 · governance_responsibility_scope=6 · governance_object_ownership=0 · candidate_scan_run=0 · governance_candidate_object=0.

Views (6), all 0 rows: v_build_auth_valid · v_object_effective_owner (9 columns — recursive form) · v_object_owner_gap · v_object_owner_conflict · v_governance_object_containment (inert seam) · v_governance_object_inventory (inert seam).

The 6 responsibility scopes are reference data, not coverage state: {approval, audit, execution, health, policy, render}.

1.3 What is empty / seeded / inactive / unsafe

Category	Items	Note
Empty (by design)	gba, ruleset, cursor, candidate_state, ownership, candidate_scan_run, candidate_object	nothing to scan/own/grant yet
Seeded (reference data only)	governance_responsibility_scope=6; apr_action_types=14 (8 gov unimplemented)	vocabulary, not activation
Inactive	5 governance event types active=false; 8 gov action-types handler_ref='unimplemented'	register-before-emit + no handler
Inert seams	v_governance_object_containment, v_governance_object_inventory	resolve WHERE false
Unsafe to touch	os_proposal_approvals (sovereign e-sign), approval_requests/apr_approvals (live L2 spine), birth_registry (1.08M, organic), event_outbox, governance_relations, dot_tools	production surfaces; any write = activation/forgery

1.4 Source inventories (read-only census — the dry-run universe)

Source	Rows	Coverage relevance
birth_registry	1,079,549	born-object tail; key collection:entity_code; cursor source for SB-13
collection_registry	168	per-collection coverage verdict ledger (coverage_status, governance_role)
meta_catalog	169	object/class drift (orphan_count, actual_count vs baseline_count)
information_unit	219	IU container/content refs (containment source)
iu_relation	60	IU edges (containment / relation source)
taxonomy	58	topic nodes (status, facet_id)
taxonomy_facets	10	axis/facet surfaces (incl. FAC-08 topic facet)
entity_labels	795,841	label projections (axis assignment reconcile)
pivot_definitions	37	pivot hierarchy (parent_code containment)
system_issues	198,442	existing issue buckets (issue ride-along target)
evolution_snapshots	1	snapshot ref for candidate_scan_run.source_snapshot_ref
event_type_registry	45 total	5 governance, 0 active

1.5 collection_registry coverage profile (sizes the dry-run)

coverage_status	governance_role	collections
BIRTH_REQUIRED	governed	35
BIRTH_REQUIRED	observed	36
BIRTH_REQUIRED	locked	3
BIRTH_DEFERRED_NEEDS_REVIEW	excluded	51
BIRTH_DEFERRED_NEEDS_REVIEW	observed	7
BIRTH_EXEMPT_* (junction/log/cache)	observed/excluded/law_artifact/governed	35

The 35 governed + BIRTH_REQUIRED collections are the natural maximal in-scope universe; the smallest useful dry-run (doc 03) carves a 1–3 collection subset from these.

meta_catalog: 164 active (Σ orphan_count = 161), 3 published (0 orphans), 2 log (0 orphans). Orphan drift is real and non-zero — exactly the signal a coverage scan would surface, and the reason the dry-run must be tightly bounded.

1.6 Readiness verdict

READY for dry-run design; NOT ready for dry-run execution. The substrate is present and inert; the source universe is enumerable and bounded; but the two SB-2 seams resolve WHERE false and the axis layer is unbuilt — so a scan today would see nothing. Wiring (doc 02) + seeds (doc 04) are prerequisites, and each is a ratified follow-up, not pure engineering.