docs(ledger): Wave 1 session log + Orientation refresh

Records the 2026-04-29 audit findings, the Codex review, and the three
memory-write-path fixes shipped on this branch. Refreshes Orientation
with live state verified against Dalidou: build 7042eae, harness 20/20,
1054 interactions, 315/1091 dashboard/integrity memory counts (the
sampling bug behind the discrepancy is fixed in the prior commit; the
two will agree post-deploy).

Also captures the unregistered emerging-projects list (apm=63, openclaw=9,
lead-space=2, drill=1, optiques-fullum=1) so the Wave 1 follow-up
(one-click registration proposal) doesn't get lost.

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): `f44a211` (verified 2026-04-24T14:48:44Z post audit-improvements deploy; status=ok)
+- **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**: 2026-04-24 by Codex (retrieval boundary deployed; project_id metadata branch started)
+- **last_updated**: 2026-04-29 by Claude (Wave 1 debt-pay branch open; Codex review of plan applied)
-- **main_tip**: `f44a211`
+- **main_tip**: `7042eae`
-- **test_count**: 567 on `codex/project-id-metadata-retrieval` (deployed main baseline: 553)
+- **test_count**: 581 on `claude/wave1-dashboard-counts-and-memory-fixes` (572 on main + 9 Wave 1 regressions)
-- **harness**: `19/20 PASS` on live Dalidou, 0 blocking failures, 1 known content gap (`p04-constraints`)
+- **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**: 290 (`/admin/dashboard` 2026-04-24; note integrity panel reports a separate active_memory_count=951 and needs reconciliation)
+- **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 (triage queue drained)
+- **candidate_memories**: 0 (queue drained, +103 captures since 2026-04-25)
-- **interactions**: 951 (`/admin/dashboard` 2026-04-24)
+- **interactions**: 1054 (claude-code 474, openclaw 576; verified `/admin/dashboard` 2026-04-29)
-- **registered_projects**: atocore, p04-gigabit, p05-interferometer, p06-polisher, atomizer-v2, abb-space (aliased p08)
+- **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 (`/admin/dashboard` 2026-04-24)
+- **project_state_entries**: 128 across registered projects (verified 2026-04-29)
-- **entities**: 66 (up from 35 — V1-0 backfill + ongoing work; 0 open conflicts)
+- **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 (NEW)** → weekly synth/lint Sundays → **retrieval harness (NEW)** → **pipeline summary (NEW)**
+- **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 shows pipeline health, interaction totals by client, all registered projects)
+- **dashboard**: http://dalidou:8100/admin/dashboard
-- **active_track**: Engineering V1 Completion (started 2026-04-22). V1-0 landed (`2712c5d`). V1-A density gate CLEARED (784 active ≫ 100 target as of 2026-04-23). V1-A soak gate at day 5/~7 (F4 first run 2026-04-19; nightly clean 2026-04-19 through 2026-04-23; failures confined to the known p04-constraints content gap). Plan: `docs/plans/engineering-v1-completion-plan.md`. Resume map: `docs/plans/v1-resume-state.md`.
+- **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**: `2026-04-23T03:00:20Z` — harness 17/18, triage promoted=3 rejected=7 human=0, dedup 7 clusters (1 tier1 + 6 tier2 auto-merged), graduation 30-skipped 0-graduated 0-errors, auto-triage drained the queue (0 new candidates 2026-04-22T00:52Z run)
+- **last_nightly_pipeline**: `2026-04-28T03:00:30Z` — harness 20/20, triage promoted=1 rejected=1 human=0
-- **open_branches**: none — R14 squash-merged as `0989fed` and deployed 2026-04-23T15:20:53Z. V1-A is the next scheduled work
+- **open_branches**: `claude/wave1-dashboard-counts-and-memory-fixes` (commit `fb4d55c`) — three memory-write-path bugs surfaced by Codex review; awaiting Codex audit before merge/deploy.
 
 ## Active Plan
 
@@ -170,6 +170,12 @@ One branch `codex/extractor-eval-loop` for Day 1-5, a second `codex/retrieval-ha
 
 ## Session Log
 
+- **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).
+
+- **2026-04-25 Codex (p04 constraint gap closed; harness fully green)** Root-caused the remaining `p04-constraints` fixture: the `Zerodur` / `1.2` fact already existed in Trusted Project State (`requirement/key_constraints`), but project-state formatting was category/key ordered and then truncated to the 20% state budget, so contacts/decisions consumed the budget before the relevant requirement. Added query-relevance ranking for Trusted Project State entries before formatting/truncation, with regression coverage in `test_project_state_query_relevance_before_truncation`. Removed the fixture's `known_issue` lane so future p04 constraint regressions are blocking. Cleaned up a duplicate live requirement entry created during diagnosis by invalidating `requirement/mirror-blank-core-constraints`; canonical `requirement/key_constraints` remains active. Verified focused suite: 35 passed. Verified full local suite: 572 passed. Deployed `d3de9f67eaa08dfc5b2d86e8221b8c70fef266d3`; live exact p04 probe now surfaces `[REQUIREMENT] key_constraints` with `1.2` and `Zerodur`. Live retrieval harness: 20/20, 0 known issues, 0 blocking failures.
+
+- **2026-04-25 Codex (project_id backfill + retrieval stabilization closed)** Merged `codex/project-id-metadata-retrieval` into `main` (`867a1ab`) and deployed to Dalidou. Took Chroma-inclusive backup `/srv/storage/atocore/backups/snapshots/20260424T154358Z`, then ran `scripts/backfill_chunk_project_ids.py` per project; populated projects `p04-gigabit`, `p05-interferometer`, `p06-polisher`, `atomizer-v2`, and `atocore` applied cleanly for 33,253 vectors total, with 0 missing/malformed and an immediate final dry-run showing 33,253 already tagged / 0 updates. Post-backfill harness exposed p06 memory-ranking misses (`Tailscale`, `encoder`), so Codex shipped `4744c69` then `a87d984 fix(memory): widen query-time context candidates`. Full local suite: 571 passed. Live `/health` reports `a87d9845a8c34395a02890f0cf22aa7a46afaf62`, vectors=33,253, sources_ready=true. Live retrieval harness: 19/20, 0 blocking failures, 1 known issue (`p04-constraints` missing `Zerodur` / `1.2`). A repeat backfill dry-run after the code-only stabilization deploy was aborted after the one-off container ran too long; the live service stayed healthy and the earlier post-apply idempotency result remains the migration acceptance record. Dalidou HTTP push credentials are still not configured; this session pushed through the Windows credential path.
+
 - **2026-04-24 Codex (retrieval boundary deployed + project_id metadata tranche)** Merged `codex/audit-improvements-foundation` to `main` as `f44a211` and pushed to Dalidou Gitea. Took pre-deploy runtime backup `/srv/storage/atocore/backups/snapshots/20260424T144810Z` (DB + registry, no Chroma). Deployed via `papa@dalidou` canonical `deploy/dalidou/deploy.sh`; live `/health` reports build_sha `f44a2114970008a7eec4e7fc2860c8f072914e38`, build_time `2026-04-24T14:48:44Z`, status ok. Post-deploy retrieval harness: 20 fixtures, 19 pass, 0 blocking failures, 1 known issue (`p04-constraints`). The former blocker `p05-broad-status-no-atomizer` now passes. Manual p05 `context-build "current status"` spot check shows no p04/Atomizer source bleed in retrieved chunks. Started follow-up branch `codex/project-id-metadata-retrieval`: registered-project ingestion now writes explicit `project_id` into DB chunk metadata and Chroma vector metadata; retrieval prefers exact `project_id` when present and keeps path/tag matching as legacy fallback; added dry-run-by-default `scripts/backfill_chunk_project_ids.py` to backfill SQLite + Chroma metadata; added tests for project-id ingestion, registered refresh propagation, exact project-id retrieval, and collision fallback. Verified targeted suite (`test_ingestion.py`, `test_project_registry.py`, `test_retrieval.py`): 36 passed. Verified full suite: 556 passed in 72.44s. Branch not merged or deployed yet.
 
 - **2026-04-24 Codex (project_id audit response)** Applied independent-audit fixes on `codex/project-id-metadata-retrieval`. Closed the nightly `/ingest/sources` clobber risk by adding registry-level `derive_project_id_for_path()` and making unscoped `ingest_file()` derive ownership from registered ingest roots when possible; `refresh_registered_project()` still passes the canonical project id directly. Changed retrieval so empty `project_id` falls through to legacy path/tag ownership instead of short-circuiting as unowned. Hardened `scripts/backfill_chunk_project_ids.py`: `--apply` now requires `--chroma-snapshot-confirmed`, runs Chroma metadata updates before SQLite writes, batches updates, skips/report missing vectors, skips/report malformed metadata, reports already-tagged rows, and turns missing ingestion tables into a JSON `db_warning` instead of a traceback. Added tests for auto-derive ingestion, empty-project fallback, ingest-root overlap rejection, and backfill dry-run/apply/snapshot/missing-vector/malformed cases. Verified targeted suite (`test_backfill_chunk_project_ids.py`, `test_ingestion.py`, `test_project_registry.py`, `test_retrieval.py`): 45 passed. Verified full suite: 565 passed in 73.16s. Local dry-run on empty/default data returns 0 updates with `db_warning` rather than crashing. Branch still not merged/deployed.

docs/current-state.md

@@ -1,11 +1,18 @@
-# AtoCore - Current State (2026-04-24)
+# AtoCore - Current State (2026-04-25)
 
-Update 2026-04-24: audit-improvements deployed as `f44a211`; live harness is
+Update 2026-04-25: project-id chunk/vector metadata is deployed and backfilled,
-19/20 with 0 blocking failures and 1 known content gap. Active follow-up branch
+and the final p04 Trusted Project State budget/ranking gap is closed. Live
-`codex/project-id-metadata-retrieval` is at 567 passing tests.
+Dalidou is on `d3de9f6`; `/health` is ok with 33,253 vectors and sources
+ready. Live retrieval harness is 20/20 with 0 blocking failures and 0 known
+issues. Full local suite: 572 passed.
 
-Live deploy: `2b86543` · Dalidou health: ok · Harness: 18/20 with 1 known
+The project-id backfill was applied per populated project after a
-content gap and 1 current blocking project-bleed guard · Tests: 553 passing.
+Chroma-inclusive backup at
+`/srv/storage/atocore/backups/snapshots/20260424T154358Z`. The immediate
+post-apply dry-run reported 33,253 already tagged, 0 updates, 0 missing, and 0
+malformed. A later repeat dry-run after the code-only ranking deploy was
+aborted because the one-off container ran too long; the earlier post-apply
+idempotency result remains the migration acceptance record.
 
 ## V1-0 landed 2026-04-22
 
@@ -69,10 +76,10 @@ Last nightly run (2026-04-19 03:00 UTC): **31 promoted · 39 rejected · 0 needs
 | 7G | Re-extraction on prompt version bump | pending |
 | 7H | Chroma vector hygiene (delete vectors for superseded memories) | pending |
 
-## Known gaps (honest, refreshed 2026-04-24)
+## Known gaps (honest, refreshed 2026-04-25)
 
 1. **Capture surface is Claude-Code-and-OpenClaw only.** Conversations in Claude Desktop, Claude.ai web, phone, or any other LLM UI are NOT captured. Example: the rotovap/mushroom chat yesterday never reached AtoCore because no hook fired. See Q4 below.
-2. **Project-scoped retrieval guard is deployed and passing.** The April 24 p05 broad-status bleed guard now passes on live Dalidou. The active follow-up branch adds explicit `project_id` chunk/vector metadata so the deployed path/tag heuristic can become a legacy fallback.
+2. **Project-scoped retrieval guard is deployed and passing.** Explicit `project_id` chunk/vector metadata is now present in SQLite and Chroma for the 33,253-vector corpus. Retrieval prefers exact metadata ownership and keeps path/tag matching as a legacy fallback.
 3. **Human interface is useful but not yet the V1 Human Mirror.** Wiki/dashboard pages exist, but the spec routes, deterministic mirror files, disputed markers, and curated annotations remain V1-D work.
-4. **Harness known issue:** `p04-constraints` wants "Zerodur" and "1.2"; live retrieval surfaces related constraints but not those exact strings. Treat as content/state gap until fixed.
+4. **Harness is currently green.** The former `p04-constraints` known issue is closed; query-relevant Trusted Project State entries now rank before state-budget truncation.
 5. **Formal docs lag the ledger during fast work.** Use `DEV-LEDGER.md` and `python scripts/live_status.py` for live truth, then copy verified claims into these docs.

docs/master-plan-status.md

@@ -131,10 +131,12 @@ This sits implicitly between Phase 8 (OpenClaw) and Phase 11
 (multi-model). Memory-review and engineering-entity commands are
 deferred from the shared client until their workflows are exercised.
 
-## What Is Real Today (updated 2026-04-24)
+## What Is Real Today (updated 2026-04-25)
 
-- canonical AtoCore runtime on Dalidou (`2b86543`, deploy.sh verified)
+- canonical AtoCore runtime on Dalidou (`d3de9f6`, deploy.sh verified)
-- 33,253 vectors across 6 registered projects
+- 33,253 vectors across 6 registered projects, with explicit `project_id`
+  metadata backfilled into SQLite and Chroma after snapshot
+  `/srv/storage/atocore/backups/snapshots/20260424T154358Z`
 - 951 captured interactions as of the 2026-04-24 live dashboard; refresh
   exact live counts with
   `python scripts/live_status.py`
@@ -149,10 +151,13 @@ deferred from the shared client until their workflows are exercised.
 - 290 active memories and 0 candidate memories as of the 2026-04-24 live
   dashboard
 - context pack assembly with 4 tiers: Trusted Project State > identity/preference > project memories > retrieved chunks
-- query-relevance memory ranking with overlap-density scoring
+- query-relevance memory ranking with overlap-density scoring and widened
-- retrieval eval harness: 20 fixtures; current live has 19 pass, 1 known
+  query-time candidate pools so older exact-intent project memories can rank
-  content gap, and 0 blocking failures after the audit-improvements deploy
+  ahead of generic high-confidence notes
-- 567 tests passing on the active `codex/project-id-metadata-retrieval` branch
+- query-relevance Trusted Project State ranking before state-budget truncation
+- retrieval eval harness: 20 fixtures; current live has 20 pass, 0 known
+  issues, and 0 blocking failures
+- 572 tests passing on `main`
 - nightly pipeline: backup → cleanup → rsync → OpenClaw import → vault refresh → extract → triage → **auto-promote/expire** → weekly synth/lint → **retrieval harness** → **pipeline summary to project state**
 - Phase 10 operational: reinforcement-based auto-promotion (ref_count ≥ 3, confidence ≥ 0.7) + stale candidate expiry (14 days unreinforced)
 - pipeline health visible in dashboard: interaction totals by client, pipeline last_run, harness results, triage stats
@@ -173,9 +178,10 @@ These are the current practical priorities.
    Target: 100+ active memories.
 3. **Multi-model triage** (Phase 11 entry) — switch auto-triage to a
    different model than the extractor for independent validation
-4. **Fix p04-constraints harness failure** — retrieval doesn't surface
+4. **Fix Dalidou Git credentials** — the host checkout can fetch but cannot
-   "Zerodur" for p04 constraint queries. Investigate if it's a missing
+   push to Gitea over HTTP in non-interactive SSH sessions. Prefer switching
-   memory or retrieval ranking issue.
+   the deploy checkout to a Gitea SSH key; PAT-backed `credential.helper store`
+   is the fallback.
 
 ## Active — Engineering V1 Completion Track (started 2026-04-22)
 

scripts/retrieval_eval_fixtures.json

@@ -27,8 +27,7 @@
     "expect_absent": [
       "polisher suite"
     ],
-    "known_issue": true,
+    "notes": "Regression guard: query-relevant Trusted Project State requirements must survive the project-state budget cap."
-    "notes": "Known content gap as of 2026-04-24: live retrieval surfaces related constraints but not the exact Zerodur / 1.2 strings. Keep visible, but do not make nightly harness red until the source/state gap is fixed."
   },
   {
     "name": "p04-short-ambiguous",

src/atocore/api/routes.py

@@ -303,6 +303,7 @@ class MemoryUpdateRequest(BaseModel):
     memory_type: str | None = None
     domain_tags: list[str] | None = None
     valid_until: str | None = None
+    project: str | None = None
 
 
 class ProjectStateSetRequest(BaseModel):
@@ -636,6 +637,7 @@ def api_update_memory(memory_id: str, req: MemoryUpdateRequest) -> dict:
             memory_type=req.memory_type,
             domain_tags=req.domain_tags,
             valid_until=req.valid_until,
+            project=req.project,
         )
     except ValueError as e:
         raise HTTPException(status_code=400, detail=str(e))
@@ -794,33 +796,25 @@ def api_invalidate_memory(
     req: MemoryInvalidateRequest | None = None,
 ) -> dict:
     """Retract an active memory (Issue E — active → invalid)."""
-    from atocore.memory.service import get_memories as _get_memories, invalidate_memory
+    from atocore.memory.service import get_memory, invalidate_memory
 
     reason = req.reason if req else ""
-    # Quick existence/status check for a clean 404 vs 409.
+    # Direct id lookup — earlier code used get_memories(status='active', limit=1)
-    existing = [
+    # which only saw the highest-confidence active row, so any other active
-        m for m in _get_memories(status="active", limit=1)
+    # memory would 404 here even though it existed.
-        if m.id == memory_id
+    target = get_memory(memory_id)
-    ]
+    if target is None:
-    if not existing:
-        # Fall through to generic not-active if the id exists in another status.
-        all_match = [
-            m for m in _get_memories(status="candidate", limit=5000)
-            + _get_memories(status="invalid", limit=5000)
-            + _get_memories(status="superseded", limit=5000)
-            if m.id == memory_id
-        ]
-        if all_match:
-            if all_match[0].status == "invalid":
-                return {"status": "already_invalid", "id": memory_id}
-            raise HTTPException(
-                status_code=409,
-                detail=(
-                    f"Memory {memory_id} is {all_match[0].status}; "
-                    "use /reject for candidates"
-                ),
-            )
         raise HTTPException(status_code=404, detail=f"Memory not found: {memory_id}")
+    if target.status == "invalid":
+        return {"status": "already_invalid", "id": memory_id}
+    if target.status != "active":
+        raise HTTPException(
+            status_code=409,
+            detail=(
+                f"Memory {memory_id} is {target.status}; "
+                "use /reject for candidates"
+            ),
+        )
 
     success = invalidate_memory(memory_id, actor="api-http", reason=reason)
     if not success:
@@ -1280,16 +1274,20 @@ def api_dashboard() -> dict:
     health beyond the basic /health endpoint.
     """
     import json as _json
-    from collections import Counter
     from datetime import datetime as _dt, timezone as _tz
 
-    all_memories = get_memories(active_only=False, limit=500)
+    from atocore.memory.service import get_memory_count_summary
-    active = [m for m in all_memories if m.status == "active"]
-    candidates = [m for m in all_memories if m.status == "candidate"]
 
-    type_counts = dict(Counter(m.memory_type for m in active))
+    # SQL-backed counts. Earlier code derived these by sampling the top
-    project_counts = dict(Counter(m.project or "(none)" for m in active))
+    # 500 rows of get_memories() ordered by confidence — anything past
-    reinforced = [m for m in active if m.reference_count > 0]
+    # the cap was invisible, so /admin/dashboard silently undercounted
+    # active memories once the corpus crossed ~500 active rows.
+    counts = get_memory_count_summary()
+    active_total = counts["active"]["total"]
+    candidate_total = counts["by_status"].get("candidate", 0)
+    type_counts = counts["active"]["by_type"]
+    project_counts = counts["active"]["by_project"]
+    reinforced_total = counts["active"]["reinforced"]
 
     # Interaction stats — total + by_client from DB directly
     interaction_stats: dict = {"most_recent": None, "total": 0, "by_client": {}}
@@ -1402,13 +1400,13 @@ def api_dashboard() -> dict:
 
     # Triage queue health
     triage: dict = {
-        "pending": len(candidates),
+        "pending": candidate_total,
         "review_url": "/admin/triage",
     }
-    if len(candidates) > 50:
+    if candidate_total > 50:
-        triage["warning"] = f"High queue: {len(candidates)} candidates pending review."
+        triage["warning"] = f"High queue: {candidate_total} candidates pending review."
-    elif len(candidates) > 20:
+    elif candidate_total > 20:
-        triage["notice"] = f"{len(candidates)} candidates awaiting triage."
+        triage["notice"] = f"{candidate_total} candidates awaiting triage."
 
     # Recent audit activity (Phase 4 V1) — last 10 mutations for operator
     recent_audit: list[dict] = []
@@ -1420,11 +1418,13 @@ def api_dashboard() -> dict:
 
     return {
         "memories": {
-            "active": len(active),
+            "active": active_total,
-            "candidates": len(candidates),
+            "candidates": candidate_total,
             "by_type": type_counts,
             "by_project": project_counts,
-            "reinforced": len(reinforced),
+            "reinforced": reinforced_total,
+            "by_status": counts["by_status"],
+            "total": counts["total"],
         },
         "project_state": {
             "counts": ps_counts,

src/atocore/context/builder.py

@@ -11,7 +11,7 @@ from dataclasses import dataclass, field
 from pathlib import Path
 
 import atocore.config as _config
-from atocore.context.project_state import format_project_state, get_state
+from atocore.context.project_state import ProjectStateEntry, format_project_state, get_state
 from atocore.memory.service import get_memories_for_context
 from atocore.observability.logger import get_logger
 from atocore.engineering.service import get_entities, get_entity_with_context
@@ -116,6 +116,11 @@ def build_context(
     if canonical_project:
         state_entries = get_state(canonical_project)
         if state_entries:
+            state_entries = _rank_project_state_entries(
+                state_entries,
+                query=user_prompt,
+                project=canonical_project,
+            )
             project_state_text = format_project_state(state_entries)
             project_state_text, project_state_chars = _truncate_text_block(
                 project_state_text,
@@ -284,6 +289,55 @@ def get_last_context_pack() -> ContextPack | None:
     return _last_context_pack
 
 
+def _rank_project_state_entries(
+    entries: list[ProjectStateEntry],
+    query: str,
+    project: str,
+) -> list[ProjectStateEntry]:
+    """Promote query-relevant trusted state before the state band is truncated."""
+    if not query or len(entries) <= 1:
+        return entries
+
+    from atocore.memory.reinforcement import _normalize, _tokenize
+
+    query_text = _normalize(query.replace("_", " "))
+    query_tokens = set(_tokenize(query_text))
+    query_tokens -= {
+        "how",
+        "what",
+        "when",
+        "where",
+        "which",
+        "who",
+        "why",
+        "current",
+        "status",
+        "project",
+    }
+    for part in (project or "").lower().replace("_", "-").split("-"):
+        query_tokens.discard(part)
+    if not query_tokens:
+        return entries
+
+    scored: list[tuple[int, float, float, int, ProjectStateEntry]] = []
+    for index, entry in enumerate(entries):
+        entry_text = " ".join(
+            [
+                entry.category,
+                entry.key.replace("_", " "),
+                entry.value,
+                entry.source,
+            ]
+        )
+        entry_tokens = _tokenize(_normalize(entry_text))
+        overlap = len(entry_tokens & query_tokens) if entry_tokens else 0
+        density = overlap / len(entry_tokens) if entry_tokens else 0.0
+        scored.append((overlap, density, entry.confidence, -index, entry))
+
+    scored.sort(key=lambda item: (item[0], item[1], item[2], item[3]), reverse=True)
+    return [entry for _, _, _, _, entry in scored]
+
+
 def _rank_chunks(
     candidates: list[ChunkResult],
     project_hint: str | None,

src/atocore/memory/service.py

@@ -347,6 +347,83 @@ def get_memories(
     return [_row_to_memory(r) for r in rows]
 
 
+def get_memory(memory_id: str) -> Memory | None:
+    """Return a single memory by id, or None if missing.
+
+    Direct id lookup (no LIMIT, no confidence ordering) — the right
+    primitive for routes that need to check a specific memory's status
+    before acting. Avoids the sampling pitfall where ``get_memories``
+    with a small ``limit`` could hide a target row sorted past the cap.
+    """
+    with get_connection() as conn:
+        row = conn.execute(
+            "SELECT * FROM memories WHERE id = ?", (memory_id,)
+        ).fetchone()
+    return _row_to_memory(row) if row else None
+
+
+def get_memory_count_summary() -> dict:
+    """Aggregate memory counts straight from SQL (no sampling).
+
+    Returned shape:
+        {
+            "total": int,                          # all rows
+            "by_status": {status: int, ...},       # full table
+            "active": {
+                "total": int,
+                "reinforced": int,                 # active with reference_count > 0
+                "by_type": {memory_type: int, ...},
+                "by_project": {project_or_none: int, ...},
+            },
+        }
+
+    Distinct from ``get_memories(...)``, which is a row-fetcher with a
+    confidence-sorted LIMIT and is therefore not safe for counting.
+    """
+    summary: dict = {
+        "total": 0,
+        "by_status": {},
+        "active": {
+            "total": 0,
+            "reinforced": 0,
+            "by_type": {},
+            "by_project": {},
+        },
+    }
+
+    with get_connection() as conn:
+        row = conn.execute("SELECT count(*) FROM memories").fetchone()
+        summary["total"] = row[0] if row else 0
+
+        rows = conn.execute(
+            "SELECT status, count(*) FROM memories GROUP BY status"
+        ).fetchall()
+        summary["by_status"] = {r[0]: r[1] for r in rows}
+
+        active_total = summary["by_status"].get("active", 0)
+        summary["active"]["total"] = active_total
+
+        rows = conn.execute(
+            "SELECT memory_type, count(*) FROM memories "
+            "WHERE status = 'active' GROUP BY memory_type"
+        ).fetchall()
+        summary["active"]["by_type"] = {r[0]: r[1] for r in rows}
+
+        rows = conn.execute(
+            "SELECT COALESCE(NULLIF(project, ''), '(none)') AS project, count(*) "
+            "FROM memories WHERE status = 'active' GROUP BY project"
+        ).fetchall()
+        summary["active"]["by_project"] = {r[0]: r[1] for r in rows}
+
+        row = conn.execute(
+            "SELECT count(*) FROM memories "
+            "WHERE status = 'active' AND reference_count > 0"
+        ).fetchone()
+        summary["active"]["reinforced"] = row[0] if row else 0
+
+    return summary
+
+
 def update_memory(
     memory_id: str,
     content: str | None = None,
@@ -355,6 +432,7 @@ def update_memory(
     memory_type: str | None = None,
     domain_tags: list[str] | None = None,
     valid_until: str | None = None,
+    project: str | None = None,
     actor: str = "api",
     note: str = "",
 ) -> bool:
@@ -368,6 +446,10 @@ def update_memory(
 
         next_content = content if content is not None else existing["content"]
         next_status = status if status is not None else existing["status"]
+        next_project = (
+            resolve_project_name(project) if project is not None
+            else (existing["project"] or "")
+        )
         if confidence is not None:
             _validate_confidence(confidence)
 
@@ -375,7 +457,7 @@ def update_memory(
             duplicate = conn.execute(
                 "SELECT id FROM memories "
                 "WHERE memory_type = ? AND content = ? AND project = ? AND status = 'active' AND id != ?",
-                (existing["memory_type"], next_content, existing["project"] or "", memory_id),
+                (existing["memory_type"], next_content, next_project, memory_id),
             ).fetchone()
             if duplicate:
                 raise ValueError("Update would create a duplicate active memory")
@@ -386,6 +468,7 @@ def update_memory(
             "status": existing["status"],
             "confidence": existing["confidence"],
             "memory_type": existing["memory_type"],
+            "project": existing["project"] or "",
         }
         after_snapshot = dict(before_snapshot)
 
@@ -422,6 +505,10 @@ def update_memory(
             updates.append("valid_until = ?")
             params.append(vu)
             after_snapshot["valid_until"] = vu or ""
+        if project is not None:
+            updates.append("project = ?")
+            params.append(next_project)
+            after_snapshot["project"] = next_project
 
         if not updates:
             return False

tests/test_context_builder.py

@@ -143,6 +143,52 @@ def test_project_state_respects_total_budget(tmp_data_dir, sample_markdown):
     assert len(pack.formatted_context) <= 120
 
 
+def test_project_state_query_relevance_before_truncation(tmp_data_dir, sample_markdown):
+    """Relevant trusted state should survive the project-state budget cap."""
+    init_db()
+    init_project_state_schema()
+    ingest_file(sample_markdown)
+
+    set_state(
+        "p04-gigabit",
+        "contact",
+        "abb-space",
+        "ABB Space is the primary vendor contact for polishing, CCP, IBF, procurement coordination, "
+        "contract administration, interface planning, and delivery discussions.",
+    )
+    set_state(
+        "p04-gigabit",
+        "decision",
+        "back-structure",
+        "Option B selected: conical isogrid back structure with variable rib density. "
+        "Chosen over flat-back for stiffness-to-weight ratio and manufacturability.",
+    )
+    set_state(
+        "p04-gigabit",
+        "decision",
+        "polishing-vendor",
+        "ABB Space selected as polishing vendor. Contract includes computer-controlled polishing "
+        "and ion beam figuring.",
+    )
+    set_state(
+        "p04-gigabit",
+        "requirement",
+        "key_constraints",
+        "The program targets a 1.2 m lightweight Zerodur mirror with filtered mechanical WFE below 15 nm "
+        "and mass below 103.5 kg.",
+    )
+
+    pack = build_context(
+        "what are the key GigaBIT M1 program constraints",
+        project_hint="p04-gigabit",
+        budget=3000,
+    )
+
+    assert "Zerodur" in pack.formatted_context
+    assert "1.2" in pack.formatted_context
+    assert pack.formatted_context.find("[REQUIREMENT]") < pack.formatted_context.find("[CONTACT]")
+
+
 def test_project_hint_matches_state_case_insensitively(tmp_data_dir, sample_markdown):
     """Project state lookup should not depend on exact casing."""
     init_db()

tests/test_invalidate_supersede.py

@@ -192,3 +192,91 @@ def test_v1_aliases_present(env):
         "/v1/memory/{memory_id}/supersede",
     ):
         assert p in paths, f"{p} missing"
+
+
+# ---------------------------------------------------------------------------
+# Wave 1 (2026-04-29) — invalidation route used to do
+# `_get_memories(status='active', limit=1)` and look for the target id
+# inside that single highest-confidence row, so any active memory
+# outside slot 0 fell through as 404. Direct id lookup fixes it.
+# ---------------------------------------------------------------------------
+
+
+def test_api_invalidate_finds_active_memory_outside_top_one(env):
+    """An active memory not at the top of the confidence sort must still
+    be invalidatable via POST /memory/{id}/invalidate."""
+    high = create_memory(
+        memory_type="knowledge",
+        content="high-confidence top row",
+        confidence=0.99,
+    )
+    low = create_memory(
+        memory_type="knowledge",
+        content="lower-confidence target",
+        confidence=0.55,
+    )
+    client = TestClient(app)
+    r = client.post(f"/memory/{low.id}/invalidate", json={"reason": "wave1 regression"})
+    assert r.status_code == 200, r.text
+    assert r.json()["status"] == "invalidated"
+    # And confirm the high-confidence row is untouched
+    assert _get_memory(high.id).status == "active"
+    assert _get_memory(low.id).status == "invalid"
+
+
+def test_api_invalidate_already_invalid_is_idempotent(env):
+    m = create_memory(memory_type="knowledge", content="already invalid")
+    client = TestClient(app)
+    r1 = client.post(f"/memory/{m.id}/invalidate", json={"reason": "first"})
+    assert r1.status_code == 200
+    r2 = client.post(f"/memory/{m.id}/invalidate", json={"reason": "again"})
+    assert r2.status_code == 200
+    assert r2.json()["status"] == "already_invalid"
+
+
+def test_api_invalidate_candidate_returns_409(env):
+    m = create_memory(
+        memory_type="knowledge", content="candidate route", status="candidate"
+    )
+    client = TestClient(app)
+    r = client.post(f"/memory/{m.id}/invalidate", json={"reason": "wrong route"})
+    assert r.status_code == 409
+
+
+def test_api_invalidate_unknown_id_is_404(env):
+    client = TestClient(app)
+    r = client.post("/memory/no-such-id/invalidate", json={"reason": "ghost"})
+    assert r.status_code == 404
+
+
+def test_admin_dashboard_active_count_matches_full_table(env):
+    """/admin/dashboard memories.active must match the SQL aggregate even
+    when there are more active memories than the legacy sample limit (500).
+
+    This guards the Codex finding that the dashboard was deriving counts
+    from a confidence-sorted limit=500 fetch, hiding rows past the cap.
+    We don't need 500 rows in the test — a small corpus that exercises
+    the SQL-aggregate path is enough; the integrity-vs-dashboard equality
+    is the invariant being asserted.
+    """
+    # Mix of statuses to exercise the by_status aggregate
+    create_memory(memory_type="knowledge", content="a")
+    create_memory(memory_type="knowledge", content="b", project="p06-polisher")
+    create_memory(memory_type="project", content="c-cand", status="candidate")
+    cand = create_memory(memory_type="project", content="d-cand", status="candidate")
+    # Invalidate one to seed an "invalid" bucket
+    from atocore.memory.service import invalidate_memory
+    target_id = cand.id
+    # Promote it first via direct DB so invalidate does flip a candidate
+    # to invalid via the service path (mirrors actual API trajectory).
+    invalidate_memory(target_id)
+
+    client = TestClient(app)
+    dash = client.get("/admin/dashboard").json()
+    assert dash["memories"]["active"] == 2
+    assert dash["memories"]["candidates"] == 1
+    assert dash["memories"]["by_status"]["invalid"] == 1
+    assert dash["memories"]["total"] == 4
+    assert dash["memories"]["by_project"].get("p06-polisher") == 1
+    # "(none)" bucket is the COALESCE label for empty/null project
+    assert "(none)" in dash["memories"]["by_project"]

tests/test_memory.py

@@ -575,3 +575,89 @@ def test_expire_stale_candidates_keeps_reinforced(isolated_db):
     assert mid not in expired
     mem = _get_memory_by_id(mid)
     assert mem["status"] == "candidate"
+
+
+# ---------------------------------------------------------------------------
+# Wave 1 (2026-04-29) — counts come from SQL, not from the top-N sample.
+# Exposed by Codex audit when prod /admin/dashboard reported 315 active
+# while /admin/integrity-check reported 1091. The dashboard was building
+# its counts from a confidence-sorted limit=500 fetch.
+# ---------------------------------------------------------------------------
+
+
+def test_get_memory_count_summary_returns_full_table_aggregates(isolated_db):
+    """Counts come from SQL aggregates, not a sampled fetch."""
+    from atocore.memory.service import (
+        create_memory,
+        get_memory_count_summary,
+        invalidate_memory,
+    )
+
+    # Create more rows than any reasonable sampling LIMIT so any
+    # LIMIT-based counter would visibly disagree with reality.
+    for i in range(120):
+        create_memory(
+            "knowledge",
+            f"fact-{i}",
+            project="p04-gigabit",
+            confidence=0.9,
+            status="active",
+        )
+    for i in range(7):
+        create_memory("knowledge", f"cand-{i}", status="candidate")
+    invalid_obj = create_memory("knowledge", "to-invalidate", status="active")
+    invalidate_memory(invalid_obj.id)
+
+    summary = get_memory_count_summary()
+    assert summary["total"] == 120 + 7 + 1
+    assert summary["by_status"]["active"] == 120
+    assert summary["by_status"]["candidate"] == 7
+    assert summary["by_status"]["invalid"] == 1
+    assert summary["active"]["total"] == 120
+    assert summary["active"]["by_type"] == {"knowledge": 120}
+    assert summary["active"]["by_project"] == {"p04-gigabit": 120}
+
+
+def test_get_memory_returns_single_row_or_none(isolated_db):
+    from atocore.memory.service import create_memory, get_memory
+
+    mem = create_memory("knowledge", "single-row test")
+    fetched = get_memory(mem.id)
+    assert fetched is not None
+    assert fetched.id == mem.id
+    assert get_memory("non-existent-id") is None
+
+
+def test_update_memory_can_change_project_with_canonicalization(
+    isolated_db, project_registry
+):
+    """update_memory(project=...) canonicalizes aliases and writes audit."""
+    project_registry(("p04-gigabit", ("p04", "gigabit")))
+    from atocore.memory.service import (
+        create_memory,
+        get_memory,
+        get_memory_audit,
+        update_memory,
+    )
+
+    mem = create_memory("knowledge", "retargetable fact", project="atocore")
+    ok = update_memory(mem.id, project="p04")  # alias
+    assert ok is True
+
+    refreshed = get_memory(mem.id)
+    assert refreshed.project == "p04-gigabit"  # canonical, not "p04"
+
+    audit_rows = get_memory_audit(mem.id, limit=10)
+    update_rows = [r for r in audit_rows if r.get("action") == "updated"]
+    assert update_rows, f"expected an updated audit row, got {audit_rows}"
+    head = update_rows[0]
+    assert head["before"]["project"] == "atocore"
+    assert head["after"]["project"] == "p04-gigabit"
+
+
+def test_update_memory_project_unchanged_when_not_passed(isolated_db):
+    from atocore.memory.service import create_memory, get_memory, update_memory
+
+    mem = create_memory("knowledge", "untouched project", project="p06-polisher")
+    update_memory(mem.id, content="edited content")
+    assert get_memory(mem.id).project == "p06-polisher"