docs(ledger): Wave 1 Codex audit closure + amend session log

Records Codex's formal GO WITH CONDITIONS verdict, the two P2 amends
shipped on 3a474f7, and the deployment checklist carried into the
merge plan.

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

DEV-LEDGER.md

@@ -6,26 +6,26 @@
 
 ## Orientation
 
-- **live_sha** (Dalidou `/health` build_sha): `23cdb31` (verified 2026-04-29T17:04:05Z post V1-A deploy; status=ok)
-- **last_updated**: 2026-04-29 by Claude (Wave 1 + Wave 1.5 + V1-A all squash-merged and deployed today)
-- **main_tip**: `23cdb31`
-- **test_count**: 605 on `main` (572 + 14 Wave 1 + 10 Wave 1.5 + 9 V1-A)
-- **harness**: `20/20 PASS` on live Dalidou against `23cdb31`, 0 blocking failures, 0 known issues
+- **live_sha** (Dalidou `/health` build_sha): `7042eae` (verified 2026-04-29T01:19Z; status=ok; deployed 2026-04-25T01:04Z, docs-only on top of `d3de9f6`)
+- **last_updated**: 2026-04-29 by Claude (Wave 1 debt-pay branch open; Codex review of plan applied)
+- **main_tip**: `7042eae`
+- **test_count**: 586 on `claude/wave1-dashboard-counts-and-memory-fixes` (572 on main + 14 Wave 1 regressions; +5 from Codex review amends)
+- **harness**: `20/20 PASS` on live Dalidou, 0 blocking failures, 0 known issues (last harness run nightly 2026-04-28T03:00:30Z)
 - **vectors**: 33,253
-- **active_memories**: 1170 dashboard (live SQL post-fix) vs 1091 integrity-panel snapshot (last nightly 2026-04-28T03:00:30Z; difference is ~80 captures since the snapshot — no longer a sampling artifact). Total memories: 2436. Pre-deploy the dashboard reported 315 due to a confidence-sorted limit=500 sampling bug, now closed.
-- **candidate_memories**: 44 (live as of 2026-04-29T02:00Z; queue accumulated since the 2026-04-28 nightly autotriage)
+- **active_memories**: 315 dashboard / 1091 integrity (verified live 2026-04-29). Discrepancy was a dashboard sampling bug — Wave 1 commit `fb4d55c` replaces it with SQL aggregates; post-deploy the two will agree.
+- **candidate_memories**: 0 (queue drained, +103 captures since 2026-04-25)
 - **interactions**: 1054 (claude-code 474, openclaw 576; verified `/admin/dashboard` 2026-04-29)
-- **registered_projects** (8): atocore, p04-gigabit, p05-interferometer, p06-polisher, atomizer-v2, abb-space (aliased p08), **apm** (registered 2026-04-29 — Atomaste Parts Manager I03), **openclaw** (registered 2026-04-29 — external LLM agent harness, read-only pull integration; alias `clawd`). `/admin/projects/proposals?min_active=10` is now empty.
+- **registered_projects**: atocore, p04-gigabit, p05-interferometer, p06-polisher, atomizer-v2, abb-space (aliased p08). **Auto-detected, unregistered**: apm (63 active memories), openclaw (9), lead-space (2), drill (1), optiques-fullum (1) — see Wave 1 follow-up below.
 - **project_state_entries**: 128 across registered projects (verified 2026-04-29)
 - **entities**: 66 (V1-0 backfill complete; 0 open conflicts)
 - **off_host_backup**: `papa@192.168.86.39:/home/papa/atocore-backups/` via cron, verified
 - **nightly_pipeline**: backup → cleanup → rsync → OpenClaw import → vault refresh → extract → auto-triage → auto-promote/expire → weekly synth/lint Sundays → retrieval harness → pipeline summary
 - **capture_clients**: claude-code (Stop hook + cwd project inference), openclaw (before_agent_start + llm_output plugin, verified live)
 - **wiki**: http://dalidou:8100/wiki (browse), /wiki/projects/{id}, /wiki/entities/{id}, /wiki/search
-- **dashboard**: http://dalidou:8100/admin/dashboard (now reports SQL-aggregate counts; new fields: `memories.total`, `memories.by_status`)
-- **active_track**: Wave 1 + Wave 1.5 + V1-A closed → V1-B is the next phase (KB-CAD/KB-FEM ingest + D-2 schema docs per the V1 completion plan). Plan: `docs/plans/engineering-v1-completion-plan.md`. Resume map: `docs/plans/v1-resume-state.md`.
+- **dashboard**: http://dalidou:8100/admin/dashboard
+- **active_track**: Wave 1 debt-pay (this session) → V1-A start. V1-A gates have cleared (soak ended 2026-04-26; density 315 ≫ 100). Plan: `docs/plans/engineering-v1-completion-plan.md`. Resume map: `docs/plans/v1-resume-state.md`.
 - **last_nightly_pipeline**: `2026-04-28T03:00:30Z` — harness 20/20, triage promoted=1 rejected=1 human=0
-- **open_branches**: none. Wave 1 (`4c70756`), Wave 1.5 (`b69d2c7`), and V1-A (`23cdb31`) all squash-merged and deployed today; branches cleaned up.
+- **open_branches**: `claude/wave1-dashboard-counts-and-memory-fixes` (tip `3a474f7`) — three memory-write-path bugs + two follow-on P2 fixes from Codex's formal audit (`auto_triage.py` PUT body + `/memory/{id}/supersede` status guard). Awaiting Codex re-review of the amended branch before squash-merge/deploy.
 
 ## Active Plan
 
@@ -170,16 +170,6 @@ One branch `codex/extractor-eval-loop` for Day 1-5, a second `codex/retrieval-ha
 
 ## Session Log
 
-- **2026-04-29 Codex + Claude (V1-A landed and deployed)** Engineering V1 Completion Plan Phase V1-A done. Branch `claude/v1-a-pillar-query-slice` shipped the Q-001 subsystem-scoped shape fix exactly per `engineering-query-catalog.md:71`: new `subsystem_contents()` helper in `queries.py` walking inbound `part_of` edges, new `GET /entities/Subsystem/{id}?expand=contains` route returning `{subsystem, contains: [{id, type, entity_type, name, status}]}`, dual-emit of `type` (spec) and `entity_type` (rest-of-module convention) for compat. The existing project-wide `/engineering/projects/{name}/systems` route stays for Q-004. Aliased under `/v1`. V1-A acceptance integration test runs all four pillar queries (Q-001 subsystem, Q-005 requirements_for, Q-006 orphan_requirements, Q-017 evidence_chain) plus Q-009/Q-011 against a single p05-interferometer seed graph. Codex audited at `b575773` (GO WITH CONDITIONS, two P2s + two P3s) → all four amends folded in at `0e83525` → re-review GO. Squash-merged to main as `23cdb31`. Deployed via canonical script; live `/health` build_sha `23cdb3149fde2185f0eaf1350c5b065b08f80319`, build_time `2026-04-29T17:04:05Z`, status=ok. **Live post-deploy probes** all green: `/entities/Subsystem/{p05-id}?expand=contains` → 200, spec-shaped response; `/v1/entities/Subsystem/{p05-id}?expand=contains` → 200; `?expand=parents` → 400 with informative detail; component-id against the Subsystem route → 404 with informative detail. Live retrieval harness 20/20 against `23cdb31`. Tests on main: 596 → 605 (+9). master-plan-status.md V1-A status flipped to ✅; V1-B is now "next".
-
-- **2026-04-29 Antoine + Claude (openclaw registered, proposals queue drained)** Talked through the framing tension — openclaw is "a tool" but external (we don't own its product), so the 17 memories are really integration notes (plugin architecture, `cleanedBody` hook field, read-only pull contract, hermes-agent comparison, atocore-capture v0.2.0 plugin). Picked Option A from the four options laid out: register `openclaw` straight (zero migration) with description that names the integration framing explicitly. Aliases: `clawd` (matches local repo path on Dalidou + T420). POSTed to `/admin/projects/register-emerging`. Live `/admin/projects/proposals?min_active=10` now returns count=0 — registered project set is 8, no unregistered labels above threshold. Wave 1.5 closed end-to-end: code + audit + deploy + both registrations.
-
-- **2026-04-29 Antoine + Claude (apm registered)** Antoine confirmed apm is a product alongside atomizer-v2 (I03 in the Atomaste tools I-series). Registered via `POST /admin/projects/register-emerging` mirroring atomizer-v2's pattern: aliases `i03`, `atomaste-parts-manager`, `parts-manager`; description "Atomaste Parts Manager — parts/COTS lifecycle tool (I03 in Atomaste tool series alongside I01 Atomizer-v2 and I02 AtoCore)". Default ingest_root `vault:incoming/projects/apm/` (does not yet exist on disk; may want a `/repo` mirror via PUT /projects/apm if a code repo gets staged, like atomizer-v2's `incoming/projects/atomizer-v2/repo`). Live `/admin/projects/proposals?min_active=10` post-registration: count drops 2→1 (apm gone, openclaw remains). 165 active + 22 candidate apm-tagged memories now resolve to a registered canonical id.
-
-- **2026-04-29 Codex + Claude (Wave 1.5 audited, merged, deployed)** Branch `claude/wave1.5-emerging-project-proposals` shipped `GET /admin/projects/proposals?min_active=N` — live, on-demand companion to the nightly `scripts/detect_emerging.py` cache. Each proposal includes project_id, active+candidate counts, sample_memories (3 most recent), suggested_aliases (sibling labels sharing a >=4-char token, case-insensitive), guessed_ingest_root. New `propose_emerging_projects()` helper in `service.py`. Codex audited tip `e8ac8bb`: verdict GO with one P2 (test was trivially true) and one P3 (case-sensitive tokens). Both folded into `f70fa6b` plus a registered-token-leak guard test. Tip `f70fa6b` squash-merged to main as `b69d2c7`. Deployed via canonical script; live `/health` reports build_sha `b69d2c70881b661e6fd8f25194cbc5c20a3947be`, build_time `2026-04-29T02:16:25Z`, status=ok. Live `/admin/projects/proposals?min_active=10` returns the expected 2 proposals: **apm (165 active / 22 candidate, no aliases)** and **openclaw (17 active / 0 candidate, no aliases)**. Hydrotech and lead-space variants below threshold or fewer captures than the dashboard `by_project` showed pre-deploy (live SQL is current). apm sample memories are clearly real APM project content (parts manager, McMaster cleanup workflows). Tests: 586 → 596 (+10). Branches deleted local + remote. **Open Wave 1.5 follow-on (P2)**: Codex recommended a one-step shortcut so the operator doesn't have to copy `suggested_aliases` from the proposal JSON into the register-emerging POST — e.g., `POST /admin/projects/register-emerging` with `from_proposal: {project_id}` that reads server-side. Deferred — the JSON copy is acceptable for the small number of projects in flight. Next: register apm (operator decision pending).
-
-- **2026-04-29 Claude (Wave 1 deployed, post-deploy probes green)** Squash-merged `claude/wave1-dashboard-counts-and-memory-fixes` (tip `9604c3e`) to main as `4c70756` after Codex's GO verdict on the amended branch. Pushed main, deployed via `ssh papa@dalidou "bash /srv/storage/atocore/app/deploy/dalidou/deploy.sh"`. Live `/health` reports build_sha `4c7075650c8187473bc5992b306fb8fa67542074`, build_time `2026-04-29T01:58:59Z`, status=ok. **Post-deploy acceptance** per Codex's checklist: (1) `/admin/dashboard memories.active = 1170` — was 315 pre-deploy, sampling bug closed; new fields `memories.total = 2436`, `memories.by_status` and full `memories.by_project` breakdown live; integrity panel still reports the 22h-old 1091 snapshot from last nightly which is expected behavior. (2) Live retrieval harness 20/20 PASS against `4c70756`, no regression. (3) Supersede/invalidate route branches probed live: unknown id → 404; candidate target → 409 with status-specific detail (`...is candidate; only active memories can be superseded`, `...is candidate; use /reject for candidates`). (4) Auto-triage suggested-project correction path covered by the script-source invariant test plus end-to-end TestClient coverage of the new `MemoryUpdateRequest.project` plumbing — not exercised against live prod state to avoid casual mutation. **Live by-project dashboard breakdown** revealed 14 unregistered projects accumulating memories: apm (165!), openclaw (17), hydrotech-mining variants (13), lead-space variants (5), and ten others. apm is severely overdue. Wave 1.5 starts next: one-click registration proposal endpoint with sample-memory + alias suggestions. **Test count**: 586 on main. **Open branches**: none — Wave 1 closed.
-
 - **2026-04-29 Codex + Claude (Wave 1 formal audit closed + amends)** Codex's formal audit of `fb4d55c` (Wave 1 first commit on `claude/wave1-dashboard-counts-and-memory-fixes`): verdict GO WITH CONDITIONS. Two P1-prior closures confirmed (dashboard count bug + invalidate top-1 lookup). Project-update P2 was only "partially closed" — API/service plumbing fine, but `scripts/auto_triage.py:417` still PUT `{"content": cand["content"]}` so the operational suggested-project correction was unreachable even with `MemoryUpdateRequest.project` in place. Codex also flagged the symmetric supersede-route gap as same-class adjacent surface and recommended pulling it in here, not in Wave 1.5. Plus one P3: cover retarget-to-empty-project against a global active duplicate. Amended on `3a474f7`: (1) auto_triage PUT body now `{"project": suggested}` with a guard test that lints the script source for the new shape; (2) `/memory/{id}/supersede` mirrors the invalidate guard via `get_memory(id)` — 404 unknown / 200 already_superseded / 409 wrong-status / 200 superseded; (3) regression test for project-empty duplicate detection. Test count 581 → 586. Codex's recommended deployment checklist (post-deploy verifications, including the run-or-simulate auto-triage retarget probe) carried into the merge plan. Awaiting Codex re-review of the amended tip before squash-merge.
 
 - **2026-04-29 Claude (Wave 1 debt-pay started; Codex review of state-of-service plan)** Audited live state on Dalidou: `/health` build_sha `7042eae` (4d old), harness 20/20, 33,253 vectors, 1,748 docs, 1054 interactions (+103/4d), dashboard memories.active=315 vs integrity.active_memory_count=1091. Drafted a state-of-service assessment + Wave 1/2/3/4 plan, then asked Codex (gpt-5.5) for an adversarial review via `codex exec`. Codex verdict: assessment MIXED, plan ENDORSE WITH CHANGES. Codex caught two factual corrections (the 315-vs-1091 gap is a *sampling bug* not a definitional gap; my "R9 drops unregistered tags" framing is wrong — `extractor_llm.py:213-233` preserves them) and two new memory-write-path bugs I missed. Branched `claude/wave1-dashboard-counts-and-memory-fixes` from `7042eae`, fixed all three: (1) `/admin/dashboard` now uses a new `get_memory_count_summary()` SQL aggregate helper instead of counting inside a confidence-sorted `get_memories(limit=500)` sample; (2) `MemoryUpdateRequest` and `update_memory()` accept `project` with `resolve_project_name` canonicalization + before/after audit, so `auto_triage.py:407` suggested-project corrections will now actually apply; (3) `POST /memory/{id}/invalidate` replaces the `_get_memories(status="active", limit=1)` lookup (which only saw the highest-confidence active row) with a direct id lookup via new `get_memory(id)` helper. 9 regression tests added across `test_memory.py` and `test_invalidate_supersede.py`. Full local suite: 581 passed (572 → 581). Commit `fb4d55c`. Branch not pushed/deployed yet — awaiting Codex audit per working model. Refreshed Orientation block (live_sha, last_updated, test_count, memory counts, capture cadence, registered/unregistered project breakdown, open_branches). Wave 1 follow-up still open: (W1.2) one-click registration proposal for unregistered projects with ≥10 active memories (apm=63 is overdue); (W1.3 done by this entry) sync ledger; (W1.4) committed measurable-win probe fixture+JSON output. After this branch lands, V1-A is unblocked: gates have effectively cleared (soak ended 2026-04-26; density 315 ≫ 100 target).

docs/master-plan-status.md

@@ -198,8 +198,8 @@ where surfaces are disjoint, pauses when they collide.
 | Phase | Scope | Status |
 |---|---|---|
 | V1-0 | Write-time invariants: F-1 header fields + F-8 provenance enforcement + F-5 hook on every active-entity write + Q-3 flag-never-block | ✅ done 2026-04-22 (`2712c5d`) |
-| V1-A | Minimum query slice: Q-001 subsystem-scoped variant + Q-6 killer-correctness integration test on p05-interferometer | ✅ done 2026-04-29 (`23cdb31`) |
-| V1-B | KB-CAD + KB-FEM ingest (`POST /ingest/kb-cad/export`, `POST /ingest/kb-fem/export`) + D-2 schema docs | next |
+| V1-A | Minimum query slice: Q-001 subsystem-scoped variant + Q-6 killer-correctness integration test on p05-interferometer | 🟡 gated — starts when soak (~2026-04-26) + density (100+ active memories) gates clear |
+| V1-B | KB-CAD + KB-FEM ingest (`POST /ingest/kb-cad/export`, `POST /ingest/kb-fem/export`) + D-2 schema docs | pending V1-A |
 | V1-C | Close the remaining 8 queries (Q-002/003/007/010/012/014/018/019; Q-020 to V1-D) | pending V1-B |
 | V1-D | Full mirror surface (3 spec routes + regenerate + determinism + disputed + curated markers) + Q-5 golden file | pending V1-C |
 | V1-E | Memory→entity graduation end-to-end + remaining Q-4 trust tests | pending V1-D (note: collides with memory extractor; pauses for multi-model triage work) |

src/atocore/api/routes.py

@@ -474,33 +474,6 @@ def api_register_emerging_project(req: RegisterEmergingRequest) -> dict:
     return result
 
 
-@router.get("/admin/projects/proposals")
-def api_project_proposals(min_active: int = 10) -> dict:
-    """Live registration proposals for unregistered projects.
-
-    Reads SQL + the registry directly, so the result is current — unlike
-    `/admin/dashboard.proposals.unregistered_projects` which is the
-    nightly cache from `scripts/detect_emerging.py`. Each proposal
-    includes a guessed ingest root, sibling labels suggested as aliases,
-    and a few sample memories so the operator can sanity-check before
-    POSTing to /admin/projects/register-emerging.
-
-    Query params:
-        min_active: minimum active-memory count for a label to surface
-                    (default 10).
-    """
-    from atocore.memory.service import propose_emerging_projects
-
-    if min_active < 1:
-        raise HTTPException(status_code=400, detail="min_active must be >= 1")
-    proposals = propose_emerging_projects(min_active=min_active)
-    return {
-        "proposals": proposals,
-        "count": len(proposals),
-        "min_active": min_active,
-    }
-
-
 @router.put("/projects/{project_name}")
 def api_project_update(project_name: str, req: ProjectUpdateRequest) -> dict:
     """Update an existing project registration."""
@@ -2393,41 +2366,6 @@ def api_entity_audit(entity_id: str, limit: int = 100) -> dict:
     return {"entity_id": entity_id, "entries": entries, "count": len(entries)}
 
 
-@router.get("/entities/Subsystem/{subsystem_id}")
-def api_subsystem_contents(subsystem_id: str, expand: str = "contains") -> dict:
-    """Q-001 (subsystem-scoped variant) — the spec-shaped invocation.
-
-    Per ``docs/architecture/engineering-query-catalog.md`` Q-001:
-        ``GET /entities/Subsystem/<id>?expand=contains``
-        → ``{ subsystem, contains: [{ id, type, name, status }] }``
-
-    Distinct from the project-wide tree at
-    ``GET /engineering/projects/{name}/systems`` (Q-004), which stays
-    as-is. V1-A only adds this single shape fix; broader query catalog
-    closure is V1-C.
-
-    ``expand`` is currently restricted to ``"contains"``; other expand
-    facets (Q-007 ``constraints``, Q-008 ``decisions``) land in V1-C.
-    """
-    if expand != "contains":
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                f"unsupported expand={expand!r} for /entities/Subsystem; "
-                "V1-A supports only 'contains'."
-            ),
-        )
-    from atocore.engineering.queries import subsystem_contents
-
-    result = subsystem_contents(subsystem_id)
-    if result is None:
-        raise HTTPException(
-            status_code=404,
-            detail=f"Subsystem not found or not a subsystem: {subsystem_id}",
-        )
-    return result
-
-
 @router.get("/entities/{entity_id}")
 def api_get_entity(entity_id: str) -> dict:
     """Get an entity with its relationships and related entities."""

src/atocore/engineering/queries.py

@@ -118,70 +118,6 @@ def system_map(project: str) -> dict:
     return out
 
 
-def subsystem_contents(subsystem_id: str) -> dict | None:
-    """Q-001 subsystem-scoped variant: a single subsystem and its
-    direct ``CONTAINS`` children.
-
-    Spec: ``GET /entities/Subsystem/<id>?expand=contains`` per
-    ``docs/architecture/engineering-query-catalog.md`` Q-001.
-
-    Differs from :func:`system_map` (Q-004) which returns the
-    project-wide tree. The subsystem-scoped form is what individual
-    operator queries actually need: "what's inside this one subsystem?"
-    rather than "show me the whole project."
-
-    The relationship walk uses inbound ``part_of`` edges (the inverse
-    of ``CONTAINS``) so both child Components and child Subsystems
-    surface uniformly. Filters to active children only — superseded
-    or invalid rows do not belong in a "current contents" answer.
-
-    Returns:
-        ``{"subsystem": {id, name, project, status, description},
-           "contains": [{id, entity_type, name, status}]}``
-        or ``None`` when the entity does not exist or is not a subsystem.
-    """
-    with get_connection() as conn:
-        ss = conn.execute(
-            "SELECT * FROM entities WHERE id = ?",
-            (subsystem_id,),
-        ).fetchone()
-        if ss is None or ss["entity_type"] != "subsystem":
-            return None
-        rows = conn.execute(
-            "SELECT e.id, e.entity_type, e.name, e.status "
-            "FROM relationships r "
-            "JOIN entities e ON e.id = r.source_entity_id "
-            "WHERE r.relationship_type = 'part_of' "
-            "AND r.target_entity_id = ? "
-            "AND e.status = 'active' "
-            "ORDER BY e.entity_type, e.name",
-            (subsystem_id,),
-        ).fetchall()
-
-    return {
-        "subsystem": {
-            "id": ss["id"],
-            "name": ss["name"],
-            "project": ss["project"],
-            "status": ss["status"],
-            "description": ss["description"] or "",
-        },
-        "contains": [
-            {
-                "id": r["id"],
-                # V1-A spec uses `type` per engineering-query-catalog.md Q-001;
-                # `entity_type` is duplicated for parity with the rest of
-                # this module's response shape (see `_entity_dict`).
-                "type": r["entity_type"],
-                "entity_type": r["entity_type"],
-                "name": r["name"],
-                "status": r["status"],
-            }
-            for r in rows
-        ],
-    }
-
-
 def decisions_affecting(project: str, subsystem_id: str | None = None) -> dict:
     """Q-008: decisions that affect a subsystem (or whole project).
 

src/atocore/main.py

@@ -66,8 +66,6 @@ _V1_PUBLIC_PATHS = {
     "/entities/{entity_id}/invalidate",
     "/entities/{entity_id}/supersede",
     "/entities/{entity_id}/audit",
-    # V1-A: Q-001 subsystem-scoped variant per engineering-query-catalog
-    "/entities/Subsystem/{subsystem_id}",
     "/relationships",
     "/ingest",
     "/ingest/sources",

src/atocore/memory/service.py

@@ -424,126 +424,6 @@ def get_memory_count_summary() -> dict:
     return summary
 
 
-def propose_emerging_projects(min_active: int = 10) -> list[dict]:
-    """Return live, on-demand registration proposals for unregistered projects.
-
-    Differs from the nightly ``scripts/detect_emerging.py`` cache (which
-    is fresh once a day and lives in ``project_state.proposals``) by
-    reading current SQL and the registry directly. Each proposal is
-    operator-ready: a guessed ingest root, sibling labels suggested as
-    aliases, and a few sample memories so the operator can sanity-check
-    the bucket before committing it.
-
-    Args:
-        min_active: minimum number of active memories required for a
-            label to surface as a proposal. Defaults to 10 — anything
-            smaller is too noisy to register without more signal.
-
-    Returns:
-        list of proposal dicts, sorted by active_count desc:
-            {
-                "project_id": str,
-                "active_count": int,
-                "candidate_count": int,
-                "suggested_aliases": list[str],
-                "guessed_ingest_root": {"source": "vault", "subpath": ...},
-                "sample_memories": [{id, content_preview, updated_at}, ...],
-            }
-    """
-    from atocore.projects.registry import load_project_registry
-
-    # Build the set of names already known to the registry (canonical + aliases),
-    # lowercased. Anything in this set is "registered" and not a proposal.
-    registered_names: set[str] = set()
-    try:
-        for project in load_project_registry():
-            registered_names.add(project.project_id.lower())
-            for alias in project.aliases:
-                registered_names.add(alias.lower())
-    except Exception:
-        # Fail-open: if the registry can't load, assume nothing is
-        # registered and let the proposal surface everything.
-        pass
-
-    with get_connection() as conn:
-        # Active counts per project (excluding empty/null project — that's
-        # the global bucket, not a proposal candidate).
-        active_rows = conn.execute(
-            "SELECT project, count(*) AS c FROM memories "
-            "WHERE status = 'active' AND project IS NOT NULL AND project != '' "
-            "GROUP BY project"
-        ).fetchall()
-        cand_rows = conn.execute(
-            "SELECT project, count(*) AS c FROM memories "
-            "WHERE status = 'candidate' AND project IS NOT NULL AND project != '' "
-            "GROUP BY project"
-        ).fetchall()
-
-    cand_counts = {r["project"]: r["c"] for r in cand_rows}
-
-    # Filter to unregistered labels above threshold
-    unregistered: list[tuple[str, int, int]] = []  # (project, active_n, candidate_n)
-    for r in active_rows:
-        proj = r["project"]
-        if proj.lower() in registered_names:
-            continue
-        if r["c"] < min_active:
-            continue
-        unregistered.append((proj, r["c"], cand_counts.get(proj, 0)))
-
-    # Sibling alias detection: two unregistered labels are siblings if
-    # they share a non-trivial token (length >= 4 after splitting on
-    # '-' and '_'). Cheap, defensible, and the operator gets to veto.
-    def _tokens(label: str) -> set[str]:
-        parts = label.lower().replace("_", "-").split("-")
-        return {p for p in parts if len(p) >= 4}
-
-    label_tokens = {label: _tokens(label) for label, _a, _c in unregistered}
-
-    proposals = []
-    for proj, active_n, candidate_n in sorted(
-        unregistered, key=lambda t: (-t[1], t[0])
-    ):
-        siblings = [
-            other
-            for other in label_tokens
-            if other != proj and (label_tokens[proj] & label_tokens[other])
-        ]
-        siblings.sort()
-
-        # Sample memories: top 3 active by updated_at desc
-        with get_connection() as conn:
-            sample_rows = conn.execute(
-                "SELECT id, content, updated_at FROM memories "
-                "WHERE status = 'active' AND project = ? "
-                "ORDER BY updated_at DESC LIMIT 3",
-                (proj,),
-            ).fetchall()
-        samples = [
-            {
-                "id": r["id"],
-                "content_preview": (r["content"] or "")[:160],
-                "updated_at": r["updated_at"],
-            }
-            for r in sample_rows
-        ]
-
-        proposals.append(
-            {
-                "project_id": proj,
-                "active_count": active_n,
-                "candidate_count": candidate_n,
-                "suggested_aliases": siblings,
-                "guessed_ingest_root": {
-                    "source": "vault",
-                    "subpath": f"incoming/projects/{proj}/",
-                },
-                "sample_memories": samples,
-            }
-        )
-    return proposals
-
-
 def update_memory(
     memory_id: str,
     content: str | None = None,

tests/test_emerging_project_proposals.py

@@ -1,217 +0,0 @@
-"""Wave 1.5 — live emerging-project registration proposals.
-
-The nightly `scripts/detect_emerging.py` writes a stale cache to
-`project_state.proposals.unregistered_projects`. This endpoint provides
-the on-demand alternative that operators can hit before deciding which
-unregistered project to register via `/admin/projects/register-emerging`.
-"""
-
-import json
-
-import pytest
-from fastapi.testclient import TestClient
-
-import atocore.config as config
-from atocore.main import app
-from atocore.memory.service import create_memory
-from atocore.models.database import init_db
-
-
-@pytest.fixture
-def env(tmp_data_dir, tmp_path, monkeypatch):
-    """Fresh DB + a registry holding a single registered project so we
-    can prove the proposals endpoint excludes registered names."""
-    registry_path = tmp_path / "registry.json"
-    registry_path.write_text(
-        json.dumps(
-            {
-                "projects": [
-                    {
-                        "id": "p04-gigabit",
-                        "aliases": ["p04", "gigabit"],
-                        "description": "test",
-                        "ingest_roots": [
-                            {"source": "vault", "subpath": "incoming/projects/p04-gigabit"}
-                        ],
-                    }
-                ]
-            }
-        ),
-        encoding="utf-8",
-    )
-    monkeypatch.setenv("ATOCORE_PROJECT_REGISTRY_PATH", str(registry_path))
-    config.settings = config.Settings()
-    init_db()
-    yield tmp_data_dir
-
-
-def test_proposals_excludes_registered_project_and_its_aliases(env):
-    """Memories tagged on a registered canonical id or any of its
-    aliases must not appear as a registration proposal."""
-    # Registered: p04-gigabit (aliases p04, gigabit)
-    for i in range(15):
-        create_memory("knowledge", f"p04 fact {i}", project="p04-gigabit")
-    for i in range(15):
-        create_memory("knowledge", f"alias fact {i}", project="p04")  # alias
-
-    # Unregistered, above threshold
-    for i in range(12):
-        create_memory("knowledge", f"apm fact {i}", project="apm")
-
-    client = TestClient(app)
-    body = client.get("/admin/projects/proposals?min_active=10").json()
-    ids = [p["project_id"] for p in body["proposals"]]
-    assert "apm" in ids
-    assert "p04-gigabit" not in ids
-    assert "p04" not in ids
-    assert "gigabit" not in ids
-
-
-def test_proposals_threshold_filters_low_count_labels(env):
-    create_memory("knowledge", "single one-off", project="discrawl")
-    for i in range(3):
-        create_memory("knowledge", f"low-volume {i}", project="drill")
-    for i in range(11):
-        create_memory("knowledge", f"high-volume {i}", project="apm")
-
-    client = TestClient(app)
-    proposals = client.get("/admin/projects/proposals?min_active=10").json()["proposals"]
-    ids = [p["project_id"] for p in proposals]
-    assert "apm" in ids
-    assert "drill" not in ids
-    assert "discrawl" not in ids
-
-
-def test_proposals_suggest_sibling_aliases_via_shared_tokens(env):
-    """Lead-space + lead-space-exploration-ltd + space-exploration-ltd
-    should cluster: each proposes the others as suggested_aliases via
-    shared non-trivial tokens (length >= 4)."""
-    for label in ("lead-space", "lead-space-exploration-ltd", "space-exploration-ltd"):
-        for i in range(11):
-            create_memory("knowledge", f"{label} content {i}", project=label)
-
-    client = TestClient(app)
-    proposals = {p["project_id"]: p for p in client.get("/admin/projects/proposals").json()["proposals"]}
-
-    # All three appear and each suggests at least one of the others
-    for label in ("lead-space", "lead-space-exploration-ltd", "space-exploration-ltd"):
-        assert label in proposals
-        siblings = set(proposals[label]["suggested_aliases"])
-        # Every label shares "space" (and others share "exploration"/"lead")
-        # so at least one sibling must be present.
-        assert siblings & {"lead-space", "lead-space-exploration-ltd", "space-exploration-ltd"} - {label}
-
-
-def test_proposals_short_token_does_not_match(env):
-    """Per Codex Wave 1.5 P2: previously this test only asserted apm
-    and drill have empty siblings, which is trivially true because they
-    share no tokens at all. The real risk is an accidental relaxation
-    that lets <4-char tokens trigger clustering. Construct a setup where
-    that would matter:
-      - 'apm' and 'apm-fpga': only the 3-char 'apm' is shared. They must
-        NOT cluster, because 'apm' is too short.
-      - 'foo-fpga' and 'bar-fpga': the 4-char 'fpga' is shared. They
-        MUST cluster.
-    """
-    for label in ("apm", "apm-fpga", "foo-fpga", "bar-fpga"):
-        for i in range(11):
-            create_memory("knowledge", f"{label} fact {i}", project=label)
-
-    client = TestClient(app)
-    proposals = {p["project_id"]: p for p in client.get("/admin/projects/proposals").json()["proposals"]}
-
-    # Negative: short-token match must not happen
-    assert "apm-fpga" not in proposals["apm"]["suggested_aliases"], (
-        "'apm' (3 chars) is below the 4-char minimum; 'apm' and 'apm-fpga' "
-        "must not cluster via the 'apm' token."
-    )
-
-    # Positive: long-token match must happen — both directions
-    assert "bar-fpga" in proposals["foo-fpga"]["suggested_aliases"]
-    assert "foo-fpga" in proposals["bar-fpga"]["suggested_aliases"]
-    # And 'apm-fpga' clusters with the others via 'fpga'
-    assert "apm-fpga" in proposals["foo-fpga"]["suggested_aliases"]
-
-
-def test_proposals_clustering_is_case_insensitive(env):
-    """Token comparison must be case-insensitive so labels captured
-    with mixed casing still cluster. Codex Wave 1.5 P3."""
-    for label in ("HydroTech-Mining", "hydrotech-split-tank"):
-        for i in range(11):
-            create_memory("knowledge", f"{label} fact {i}", project=label)
-
-    client = TestClient(app)
-    proposals = {p["project_id"]: p for p in client.get("/admin/projects/proposals").json()["proposals"]}
-    assert "hydrotech-split-tank" in proposals["HydroTech-Mining"]["suggested_aliases"]
-    assert "HydroTech-Mining" in proposals["hydrotech-split-tank"]["suggested_aliases"]
-
-
-def test_proposals_registered_token_does_not_leak_into_sibling_set(env, monkeypatch):
-    """Registered project ids must be filtered BEFORE clustering so a
-    registered token doesn't get suggested as an alias for an
-    unregistered sibling. p04-gigabit is registered in env; an
-    unregistered 'gigabit-other' must not list 'p04-gigabit' as alias."""
-    for i in range(15):
-        create_memory("knowledge", f"p04 fact {i}", project="p04-gigabit")
-    for i in range(11):
-        create_memory("knowledge", f"gigabit-other fact {i}", project="gigabit-other")
-
-    client = TestClient(app)
-    proposals = {p["project_id"]: p for p in client.get("/admin/projects/proposals").json()["proposals"]}
-    assert "p04-gigabit" not in proposals
-    assert "gigabit-other" in proposals
-    # And the registered name must not surface as a sibling
-    assert "p04-gigabit" not in proposals["gigabit-other"]["suggested_aliases"]
-
-
-def test_proposals_include_sample_memories_and_guessed_root(env):
-    for i in range(11):
-        create_memory("knowledge", f"sample content {i}", project="apm")
-
-    client = TestClient(app)
-    body = client.get("/admin/projects/proposals").json()
-    apm = next(p for p in body["proposals"] if p["project_id"] == "apm")
-    assert apm["active_count"] == 11
-    assert apm["candidate_count"] == 0
-    assert apm["guessed_ingest_root"] == {
-        "source": "vault",
-        "subpath": "incoming/projects/apm/",
-    }
-    assert len(apm["sample_memories"]) == 3
-    for s in apm["sample_memories"]:
-        assert s["id"]
-        assert "sample content" in s["content_preview"]
-
-
-def test_proposals_count_candidates_separately(env):
-    for i in range(11):
-        create_memory("knowledge", f"active {i}", project="apm")
-    for i in range(4):
-        create_memory("knowledge", f"candidate {i}", project="apm", status="candidate")
-
-    client = TestClient(app)
-    apm = next(
-        p for p in client.get("/admin/projects/proposals").json()["proposals"]
-        if p["project_id"] == "apm"
-    )
-    assert apm["active_count"] == 11
-    assert apm["candidate_count"] == 4
-
-
-def test_proposals_min_active_param_validation(env):
-    client = TestClient(app)
-    r = client.get("/admin/projects/proposals?min_active=0")
-    assert r.status_code == 400
-
-
-def test_proposals_sorted_by_active_count_desc(env):
-    for i in range(20):
-        create_memory("knowledge", f"big {i}", project="apm")
-    for i in range(11):
-        create_memory("knowledge", f"small {i}", project="openclaw")
-
-    client = TestClient(app)
-    proposals = client.get("/admin/projects/proposals").json()["proposals"]
-    ids = [p["project_id"] for p in proposals]
-    assert ids[0] == "apm"
-    assert ids[1] == "openclaw"

tests/test_v1_a_pillar_queries.py

@@ -1,278 +0,0 @@
-"""V1-A — minimum query slice that proves the entity model end-to-end.
-
-Per ``docs/plans/engineering-v1-completion-plan.md`` Phase V1-A:
-the four pillar queries (Q-001 subsystem-scoped, Q-005, Q-006, Q-017)
-must run against a single seed graph and report the expected shapes.
-
-This file is intentionally separate from ``test_engineering_queries.py``
-so the V1-A acceptance is auditable in isolation. The seed graph mirrors
-what the V1-A plan called for: one satisfying Component, one orphan
-Requirement, one supported ValidationClaim, one unsupported one, one
-Decision on a flagged Assumption.
-"""
-
-from __future__ import annotations
-
-import pytest
-from fastapi.testclient import TestClient
-
-from atocore.engineering.queries import (
-    evidence_chain,
-    orphan_requirements,
-    requirements_for,
-    risky_decisions,
-    subsystem_contents,
-    unsupported_claims,
-)
-from atocore.engineering.service import (
-    create_entity,
-    create_relationship,
-    init_engineering_schema,
-    invalidate_active_entity,
-)
-from atocore.main import app
-from atocore.models.database import init_db
-
-
-@pytest.fixture
-def v1a_seed(tmp_data_dir):
-    """Seed the Q-6 integration data exactly as V1-A's plan describes:
-    one satisfying Component + one orphan Requirement + one Decision on
-    a flagged Assumption + one supported ValidationClaim + one
-    unsupported ValidationClaim, plus the Subsystem they live under."""
-    init_db()
-    init_engineering_schema()
-
-    # Subsystem (Q-001 target)
-    optics = create_entity("subsystem", "Optics", project="p05-interferometer")
-
-    # Components — one with PART_OF the subsystem and SATISFIES a requirement,
-    # one without parents (orphan_components in Q-004; not the focus here).
-    primary = create_entity("component", "Primary Mirror", project="p05-interferometer")
-    diverger = create_entity("component", "Diverger Lens", project="p05-interferometer")
-    create_relationship(primary.id, optics.id, "part_of")
-    create_relationship(diverger.id, optics.id, "part_of")
-
-    # Requirements — one satisfied by primary, one orphan (Q-006)
-    req_satisfied = create_entity(
-        "requirement", "Surface figure < 25 nm RMS", project="p05-interferometer"
-    )
-    req_orphan = create_entity(
-        "requirement", "Calibration repeatable to lambda/20", project="p05-interferometer"
-    )
-    create_relationship(primary.id, req_satisfied.id, "satisfies")
-
-    # Decision on a flagged Assumption (Q-009 surface — not asserted in V1-A
-    # acceptance but useful as background data)
-    flagged_assumption = create_entity(
-        "parameter",
-        "Vendor lead time = 6 weeks",
-        project="p05-interferometer",
-        properties={"flagged": True},
-    )
-    risky = create_entity(
-        "decision", "Pre-order CGH from external vendor", project="p05-interferometer"
-    )
-    create_relationship(risky.id, flagged_assumption.id, "based_on_assumption")
-
-    # Validation claims — one supported by a Result, one unsupported (Q-017
-    # touches the supported one; Q-011 surfaces the unsupported one)
-    fea_result = create_entity(
-        "result", "FEA thermal sweep 2026-04", project="p05-interferometer"
-    )
-    claim_supported = create_entity(
-        "validation_claim", "Thermal margin is adequate", project="p05-interferometer"
-    )
-    claim_unsupported = create_entity(
-        "validation_claim", "Vibration isolation passes spec", project="p05-interferometer"
-    )
-    create_relationship(fea_result.id, claim_supported.id, "supports")
-
-    return {
-        "subsystem": optics,
-        "primary": primary,
-        "diverger": diverger,
-        "req_satisfied": req_satisfied,
-        "req_orphan": req_orphan,
-        "claim_supported": claim_supported,
-        "claim_unsupported": claim_unsupported,
-        "fea_result": fea_result,
-        "risky_decision": risky,
-        "flagged_assumption": flagged_assumption,
-    }
-
-
-# ---------------------------------------------------------------------------
-# Q-001 subsystem-scoped — the V1-A code change
-# ---------------------------------------------------------------------------
-
-
-def test_subsystem_contents_returns_spec_shape(v1a_seed):
-    """The shape must match the catalog spec
-    (engineering-query-catalog.md Q-001):
-    ``{subsystem, contains: [{id, type, name, status}]}``.
-    The implementation also emits ``entity_type`` for parity with the
-    rest of this module's response style — both must be present."""
-    result = subsystem_contents(v1a_seed["subsystem"].id)
-
-    assert result is not None
-    assert set(result.keys()) == {"subsystem", "contains"}
-    ss = result["subsystem"]
-    assert ss["id"] == v1a_seed["subsystem"].id
-    assert ss["name"] == "Optics"
-    assert ss["project"] == "p05-interferometer"
-    assert ss["status"] == "active"
-
-    contained = {c["name"] for c in result["contains"]}
-    assert contained == {"Primary Mirror", "Diverger Lens"}
-    for child in result["contains"]:
-        # Spec requires `type`; we also include `entity_type` for parity.
-        assert "type" in child
-        assert "entity_type" in child
-        assert child["type"] == child["entity_type"]
-        assert set(child.keys()) >= {"id", "type", "entity_type", "name", "status"}
-        assert child["entity_type"] == "component"
-        assert child["status"] == "active"
-
-
-def test_subsystem_contents_returns_none_for_missing_id(tmp_data_dir):
-    init_db()
-    init_engineering_schema()
-    assert subsystem_contents("no-such-id") is None
-
-
-def test_subsystem_contents_returns_none_when_entity_is_not_a_subsystem(v1a_seed):
-    """Calling the subsystem-scoped query against a Component must
-    return None, not the component dressed up as a subsystem."""
-    assert subsystem_contents(v1a_seed["primary"].id) is None
-
-
-def test_subsystem_contents_route_matches_spec(v1a_seed):
-    """Spec invocation: GET /entities/Subsystem/<id>?expand=contains.
-    Verify the route is registered, the expand=contains path works,
-    and unsupported expand values 400."""
-    client = TestClient(app)
-    sid = v1a_seed["subsystem"].id
-
-    r = client.get(f"/entities/Subsystem/{sid}?expand=contains")
-    assert r.status_code == 200
-    body = r.json()
-    assert body["subsystem"]["id"] == sid
-    names = {c["name"] for c in body["contains"]}
-    assert names == {"Primary Mirror", "Diverger Lens"}
-
-    # Default expand is "contains" so omitting the param should also work
-    r = client.get(f"/entities/Subsystem/{sid}")
-    assert r.status_code == 200
-
-    # Unsupported expand value 400s with an informative message
-    r = client.get(f"/entities/Subsystem/{sid}?expand=parents")
-    assert r.status_code == 400
-    assert "expand" in r.json()["detail"].lower()
-
-    # Missing/wrong-type subsystem 404s
-    r = client.get(f"/entities/Subsystem/{v1a_seed['primary'].id}")
-    assert r.status_code == 404
-
-
-def test_subsystem_contents_v1_alias_is_present():
-    """The subsystem-scoped variant must also be reachable under /v1."""
-    client = TestClient(app)
-    spec = client.get("/openapi.json").json()
-    assert "/v1/entities/Subsystem/{subsystem_id}" in spec["paths"]
-
-
-def test_subsystem_contents_v1_alias_serves_real_response(v1a_seed):
-    """Behavior, not just OpenAPI presence — catches a future
-    route-copy or ordering regression (Codex V1-A P3)."""
-    client = TestClient(app)
-    sid = v1a_seed["subsystem"].id
-
-    r = client.get(f"/v1/entities/Subsystem/{sid}?expand=contains")
-    assert r.status_code == 200
-    body = r.json()
-    assert body["subsystem"]["id"] == sid
-    assert {c["name"] for c in body["contains"]} == {"Primary Mirror", "Diverger Lens"}
-
-
-def test_subsystem_contents_includes_child_subsystems_and_excludes_inactive(tmp_data_dir):
-    """Codex V1-A P3: lock in two intended behaviors that aren't
-    obvious from the data shape:
-      1. Child *Subsystems* (nested) surface in `contains` — the impl
-         walks part_of irrespective of child entity_type.
-      2. Children whose status is not 'active' are filtered out."""
-    init_db()
-    init_engineering_schema()
-
-    parent = create_entity("subsystem", "Parent System", project="p-nested")
-    child_ss = create_entity("subsystem", "Child Subsystem", project="p-nested")
-    child_comp = create_entity("component", "Child Component", project="p-nested")
-    invalid_comp = create_entity("component", "Invalidated Component", project="p-nested")
-    create_relationship(child_ss.id, parent.id, "part_of")
-    create_relationship(child_comp.id, parent.id, "part_of")
-    create_relationship(invalid_comp.id, parent.id, "part_of")
-
-    invalidate_active_entity(invalid_comp.id, reason="v1a regression seed")
-
-    result = subsystem_contents(parent.id)
-    names_by_type = {(c["entity_type"], c["name"]) for c in result["contains"]}
-    assert ("subsystem", "Child Subsystem") in names_by_type
-    assert ("component", "Child Component") in names_by_type
-    # Inactive child must not appear
-    assert all(c["name"] != "Invalidated Component" for c in result["contains"])
-    assert all(c["status"] == "active" for c in result["contains"])
-
-
-# ---------------------------------------------------------------------------
-# V1-A acceptance — the four pillar queries against the single seed graph
-# ---------------------------------------------------------------------------
-
-
-def test_v1a_pillar_queries_run_end_to_end_against_single_seed(v1a_seed):
-    """The V1-A acceptance test. All four pillar queries must report
-    the expected shape against the same seed graph. If this test fails,
-    V1-A is not done — full stop. (Per the plan: "If the four pillar
-    queries don't work, stopping here is cheap.")"""
-    project = "p05-interferometer"
-
-    # Q-001 subsystem-scoped
-    q1 = subsystem_contents(v1a_seed["subsystem"].id)
-    assert q1 is not None
-    assert {c["name"] for c in q1["contains"]} == {"Primary Mirror", "Diverger Lens"}
-
-    # Q-005 — requirements satisfied by Primary Mirror
-    q5 = requirements_for(v1a_seed["primary"].id)
-    assert q5["count"] == 1
-    assert q5["requirements"][0]["name"] == "Surface figure < 25 nm RMS"
-
-    # Q-006 (killer correctness) — orphan requirements in the project
-    q6 = orphan_requirements(project)
-    orphan_names = {r["name"] for r in q6["gaps"]}
-    assert "Calibration repeatable to lambda/20" in orphan_names
-    assert "Surface figure < 25 nm RMS" not in orphan_names
-
-    # Q-017 — evidence chain for the supported ValidationClaim
-    q17 = evidence_chain(v1a_seed["claim_supported"].id)
-    via_supports = [edge for edge in q17["evidence_chain"] if edge["via"] == "supports"]
-    assert any(edge["source_name"] == "FEA thermal sweep 2026-04" for edge in via_supports)
-
-    # And the unsupported claim must have an empty supports chain
-    q17_unsup = evidence_chain(v1a_seed["claim_unsupported"].id)
-    assert not any(edge["via"] == "supports" for edge in q17_unsup["evidence_chain"])
-
-
-def test_v1a_seed_also_exercises_q009_and_q011_killers(v1a_seed):
-    """Per Codex V1-A P2: the seed already includes Q-009 (decision on
-    flagged assumption) and Q-011 (unsupported validation claim) data,
-    so the V1-A integration test should assert those killer queries
-    surface them. Otherwise the seed entities are dead weight."""
-    project = "p05-interferometer"
-
-    q9 = risky_decisions(project)
-    risky_names = {row["decision_name"] for row in q9["gaps"]}
-    assert "Pre-order CGH from external vendor" in risky_names
-
-    q11 = unsupported_claims(project)
-    unsupported_names = {row["name"] for row in q11["gaps"]}
-    assert "Vibration isolation passes spec" in unsupported_names
-    assert "Thermal margin is adequate" not in unsupported_names