mavweb: group the allowlist by capability domain (V-452)

scope.domain.action, the shape Hexis has always spoken, derived from the row
rather than stored — a derivation is one place to argue with, a column is
whatever the last person to enable the tool typed. The name stays the primary
key and nothing about lookup or execution changes: this is a way to read the
allowlist, not a second allowlist.

MatchCapability widens one way, so house.lock covers every action on the
locks and nothing narrower can claim a wider pattern.

.gitignore

@@ -9,6 +9,7 @@
 /mavwaked
 /mavmaild
 /mavupdate
+/mavgpud
 
 # Certs (private keys, don't commit)
 certs/

CLAUDE.md

@@ -36,6 +36,13 @@ stays on homesrv permanently, because it backs that floor. Read `docs/offload.md
 touching a daemon seam or adding a model caller. Vikunja #483 is the umbrella, #484 to #487
 are the work.
 
+Both halves are wired as of 2026-08-03. Routing and replies prefer the workstation silently
+through `modelSeam`; nudge and reminder phrasing prefer it silently inside the phraser. A
+world question goes through `LLMPhraser.PhraseWorld` and names the gap when the card is not
+free — `worldGap` in `cmd/mavend/worldmodel.go`, which he hears instead of an invented
+answer. A box with no `workstation` block behaves exactly as it did before the seam: naming
+a gap requires a gap. The offload table in `docs/offload.md` says which caller is which.
+
 ## Build & test
 
 CGO daemons (`mavend`, `mavsttd`, `mavttsd`, `mavenclient`) need the vendored toolchain
@@ -143,7 +150,15 @@ seed additions, both of which now score inside the classifier baseline. Qwen3-1.
 not a doubling, and the trade is worth re-arguing rather than assuming. **The ≈2.7s figure
 that stood here until 2026-08-02 was contention, not the model.** See `docs/evals/2026-07-31-routing.md` line 61, which measures the LLM router at
 p50 825ms / p95 1.2s / max 3.0s and the full cascade at p50 0.80-1.04s. Do not plan latency
-work off the bakeoff table. `Confidence: 1.0` used to be hardcoded in `llmrouter.go`, so the LLM
+work off the bakeoff table.
+
+**The numbers above are the homesrv floor, not the ceiling.** With the workstation up, routing
+completes through `llm.Pair` against gemma-4-12b and scores **84.4% full / 93.5% intent-only at
+p50 329ms** — better than the resident model and about 2.5× faster (`docs/evals/2026-08-02-workstation-gemma4-12b.md`,
+Vikunja #485). The workstation is never assumed up, so both sets of numbers are live. Judge a
+routing change against the classifier and the resident model, since those are what always answer.
+
+`Confidence: 1.0` used to be hardcoded in `llmrouter.go`, so the LLM
 path could never ask for clarification (6/6 refusal cases missed on the fixture) — Vikunja
 #359. Fixed 31-07-2026 with structural signal (single-token utterance, keyless fact, act with
 no allowlisted fn) feeding the same stage-3 gate the classifier path already had — see

Makefile

@@ -16,11 +16,11 @@ PIPER_BIN     := $(shell pwd)/deps/piper/piper
 PIPER_MODEL   := $(shell pwd)/models/tts/ru_RU-irina-medium.onnx
 PIPER_ESPEAK  := $(shell pwd)/deps/piper/espeak-ng-data
 
-.PHONY: simulate stt-fixtures test-stt-golden all build build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav clean test fmt-check vet run-stt run-tts run-web download-embedder deps-go eval-router eval-recall eval-phrasing eval-models
+.PHONY: simulate stt-fixtures test-stt-golden all build build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav clean test fmt-check vet run-stt run-tts run-web download-embedder deps-go eval-router eval-recall eval-phrasing eval-models build-gpud
 
 all: build
 
-build: build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav build-mail build-update
+build: build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav build-mail build-update build-gpud
 
 build-stt:
 	CGO_CFLAGS="$(CGO_CFLAGS)" CGO_LDFLAGS="$(CGO_LDFLAGS)" LD_LIBRARY_PATH="$(shell pwd)/deps/lib" \
@@ -59,6 +59,12 @@ build-mail:
 build-update:
 	$(GO) build $(GOFLAGS) -o mavupdate ./cmd/mavupdate/
 
+# mavgpud runs on the workstation, not here. It is built with the rest so a
+# broken supervisor is caught by `make build` on homesrv rather than by the
+# workstation refusing to serve. Copy the binary over, do not `make deploy` it.
+build-gpud:
+	$(GO) build $(GOFLAGS) -o mavgpud ./cmd/mavgpud/
+
 run-web: build-web
 	./mavweb -addr :9200 -voice 127.0.0.1:9100
 

cmd/mavend/actions_act.go

@@ -51,6 +51,12 @@ func (h *reactiveHandler) actionAct(ctx context.Context, dec router.Decision) st
 			phrase := actPhrase(dec.Slots.Fn, dec.Slots.Args)
 			h.park(dec.Slots.Fn, dec.Slots.Args, phrase)
 			return "выполнить «" + phrase + "»? скажи «да» или «нет»."
+		case errors.Is(err, tool.ErrNeedsAuthedSurface):
+			// Irreversible (internal/tool/risk.go). A confirm turn would not
+			// help: everything that proposed this act — the STT, the router,
+			// the fuzzy allowlist match — is a guess, and a spoken "да" checks
+			// none of it. She names the gap instead.
+			return "это я из голоса не выполню — после него ничего не вернуть. запусти сам, если правда надо."
 		case errors.Is(err, tool.ErrNotEnabled):
 			return h.proposeGap(ctx, dec)
 		case errors.Is(err, tool.ErrNotConnected), errors.Is(err, mcp.ErrNotConnected), errors.Is(err, mcp.ErrNoServer):

cmd/mavend/actions_act_risk_test.go

@@ -0,0 +1,73 @@
+package main
+
+import (
+	"context"
+	"strings"
+	"testing"
+
+	"github.com/kami/maven/internal/router"
+)
+
+// The act path speaks each tier (Vikunja #449): a safe row runs, a destructive
+// one costs a confirm turn, an irreversible one is refused with the reason.
+func TestActPathSpeaksTheTiers(t *testing.T) {
+	h, st, _ := newClarifyHandler(t)
+	ctx := context.Background()
+	now := h.now()
+	for _, tc := range []struct {
+		name        string
+		cmd         []string
+		destructive bool
+	}{
+		{"status", []string{"true"}, false},
+		{"restart", []string{"true"}, true},
+		{"wipe", []string{"rm", "-rf"}, true},
+	} {
+		if _, err := st.ProposeTool(ctx, tc.name, "test", "homelab", now); err != nil {
+			t.Fatalf("propose %s: %v", tc.name, err)
+		}
+		if err := st.EnableTool(ctx, tc.name, tc.cmd, tc.destructive, "homelab", now); err != nil {
+			t.Fatalf("enable %s: %v", tc.name, err)
+		}
+	}
+
+	act := func(fn string) string {
+		return h.actionAct(ctx, router.Decision{
+			Intent:    router.IntentAct,
+			Utterance: fn,
+			Slots:     router.Slots{Fn: fn, HasFn: true},
+		})
+	}
+
+	if reply := act("status"); !strings.HasPrefix(reply, "готово") {
+		t.Errorf("safe act replied %q; want it to have run", reply)
+	}
+	if reply := act("restart"); !strings.Contains(reply, "скажи «да»") {
+		t.Errorf("destructive act replied %q; want a confirm turn", reply)
+	}
+	// Clear the confirm the destructive act parked, so what is pending after
+	// the irreversible one is only what the irreversible one parked.
+	h.mu.Lock()
+	h.pending = nil
+	h.mu.Unlock()
+
+	reply := act("wipe")
+	if strings.Contains(reply, "скажи «да»") {
+		t.Fatalf("irreversible act asked for a confirm: %q", reply)
+	}
+	if !strings.Contains(reply, "не вернуть") {
+		t.Errorf("irreversible act replied %q; want it to name the reason", reply)
+	}
+	// Nothing was parked, so a later "да" cannot pick it up.
+	h.mu.Lock()
+	pending := h.pending
+	h.mu.Unlock()
+	if pending != nil {
+		t.Errorf("an irreversible act parked %+v", pending)
+	}
+	// And it is still an enabled row — refusing to run it from voice is not
+	// the same as taking it off the allowlist.
+	if got, err := st.LookupTool(ctx, "wipe"); err != nil || got.Status != "enabled" {
+		t.Errorf("wipe is %+v, %v; want it still enabled", got, err)
+	}
+}

cmd/mavend/actions_fact.go

@@ -7,6 +7,7 @@ import (
 
 	"github.com/kami/maven/internal/ipc"
 	"github.com/kami/maven/internal/router"
+	"github.com/kami/maven/internal/store"
 )
 
 // actionFact handles router.IntentFact: persist a tapped self-fact, index
@@ -15,14 +16,41 @@ func (h *reactiveHandler) actionFact(ctx context.Context, dec router.Decision) s
 	if !dec.Slots.HasKey {
 		return "не разобрала, что записать — попробуй иначе."
 	}
+	// A question is never a fact about him (#470). "какая последняя версия
+	// языка Go?" used to land here, and the value stored was whatever the
+	// model invented for it, at confidence 1.00, indexed for recall under the
+	// question's own text. Two such rows then claimed seven unrelated world
+	// questions through recall and silently disabled world answering.
+	//
+	// The routing error itself is not fixed here — the answer is to answer.
+	// Sending the turn down the query chain is what he asked for anyway, and
+	// it costs a mis-routed capture nothing: an explicit "запиши ..." is not
+	// question-shaped, so it never takes this branch.
+	if router.IsQuestionShaped(dec.Utterance) {
+		log.Printf("voice: fact write refused, utterance is a question: %q (key %q) — answering as a query",
+			dec.Utterance, dec.Slots.Key)
+		q := dec
+		q.Intent = router.IntentQuery
+		// The key the model extracted is its guess at what to store, not a
+		// fact he has. Left in place, queryFactByKey would read it back and
+		// claim the turn before any real source ran.
+		q.Slots.Key, q.Slots.HasKey = "", false
+		q.Slots.Value = ""
+		return h.actionQuery(ctx, q)
+	}
 	now := h.now()
 	req := ipc.WriteFactReq{
-		Ts:         now,
-		Kind:       "self",
-		Key:        dec.Slots.Key,
-		Value:      dec.Slots.Value,
-		Source:     "tap:voice",
-		Confidence: 1.0,
+		Ts:     now,
+		Kind:   "self",
+		Key:    dec.Slots.Key,
+		Value:  dec.Slots.Value,
+		Source: "tap:voice",
+		// Not 1.00 unconditionally any more (#470). A value he said is
+		// evidence; a value the model supplied for words he never said is a
+		// guess, and writing a guess at full confidence is the same mistake
+		// the act path already refuses under "LLM output is not
+		// authorization".
+		Confidence: factConfidence(dec.Utterance, dec.Slots.Value),
 		// Subject: the key doubles as the entity-resolution candidate —
 		// a voice-tapped fact's key is usually the thing/person it's
 		// about ("espresso_machine", "kate"), so queueing it for Nexus
@@ -36,17 +64,25 @@ func (h *reactiveHandler) actionFact(ctx context.Context, dec router.Decision) s
 		log.Printf("voice: write fact: %v", err)
 		return "не получилось сохранить факт."
 	}
-	// Index the fact utterance in long-term memory (best-effort, must not
-	// fail the fact write). Facts aren't in the notes table, so this is the
-	// only recall path for them — "когда я пил воду?" reads back from here.
+	// Index the fact in long-term memory (best-effort, must not fail the fact
+	// write). Facts aren't in the notes table, so this is the only recall path
+	// for them — "когда я пил воду?" reads back from here.
+	//
+	// The indexed text is the fact, not the utterance (#493). queryMemory
+	// returns a fact's stored text verbatim, so what goes in here is what he
+	// hears; storing the utterance meant recall answered with his own sentence
+	// rather than the value. The utterance stays alongside as provenance —
+	// readable on /trace, never the answer and never embedded.
 	if h.memStore != nil {
-		if vec, err := router.EmbedPassage(ctx, h.embedder, dec.Utterance); err != nil {
+		text := store.FactRecallText(dec.Slots.Key, dec.Slots.Value)
+		if vec, err := router.EmbedPassage(ctx, h.embedder, text); err != nil {
 			log.Printf("voice: embed fact for memory: %v", err)
 		} else if err := h.memStore.Insert(ctx, "fact:"+dec.Slots.Key+":"+strconv.FormatInt(now.Unix(), 10), vec, map[string]string{
-			"source": "voice",
-			"type":   "fact",
-			"text":   dec.Utterance,
-			"ts":     strconv.FormatInt(now.Unix(), 10),
+			"source":    "voice",
+			"type":      "fact",
+			"text":      text,
+			"utterance": dec.Utterance,
+			"ts":        strconv.FormatInt(now.Unix(), 10),
 		}); err != nil {
 			log.Printf("voice: memory insert fact: %v", err)
 		}

cmd/mavend/actions_list.go

@@ -0,0 +1,143 @@
+package main
+
+import (
+	"context"
+	"log"
+	"strings"
+
+	"github.com/kami/maven/internal/router"
+	"github.com/kami/maven/internal/store"
+)
+
+// Standing lists on the voice path (Vikunja #453).
+//
+// Three halves, mirroring what task capture already does: an add that runs at
+// the top of actionNote, a read-back query source, and a crossing-off that runs
+// on the same note path because "всё купил" is note-shaped.
+//
+// These read h.dataStore rather than the CoreAPI. A list is local to the core
+// and nothing outside it writes one: the web UI has no list page, no reach
+// files groceries, and the digestion worker does not read the table. When
+// something outside mavend needs to add to a list, the ipc seam is what it
+// grows through — the intake rules that CaptureTaskReq documents are about
+// shared intake, and there is none here yet.
+//
+// Nothing here speaks unprompted. A list is answered when asked about.
+
+// captureListFromNote claims the turn when the utterance adds to, clears, or
+// crosses one item off a list. ("", false) hands the turn back to the note path.
+func (h *reactiveHandler) captureListFromNote(ctx context.Context, dec router.Decision) (string, bool) {
+	if h.dataStore == nil {
+		return "", false
+	}
+	// Clearing is read before removing on purpose: "всё купил" and "купил
+	// молоко" start with the same word, and only the second one names an item.
+	if list, ok := router.ParseListClear(dec.Utterance); ok {
+		n, err := h.dataStore.ClearList(ctx, list, h.now())
+		if err != nil {
+			log.Printf("voice: clear list: %v", err)
+			return "не получилось обновить список.", true
+		}
+		if n == 0 {
+			return "в списке и так ничего не было.", true
+		}
+		return "вычеркнула всё, список пустой.", true
+	}
+	if cap, ok := router.ParseListRemove(dec.Utterance); ok {
+		if reply, ok := h.removeListItem(ctx, cap); ok {
+			return reply, true
+		}
+		// Nothing on the list by that name. "купил новый ноутбук" is a note and
+		// must stay one, so the turn goes back rather than claiming a removal
+		// that removed nothing.
+		return "", false
+	}
+	cap, ok := router.ParseListCapture(dec.Utterance)
+	if !ok {
+		return "", false
+	}
+	res, err := h.dataStore.AddListItem(ctx, store.ListItem{
+		List:      cap.List,
+		Item:      cap.Item,
+		Source:    "tap:voice",
+		CreatedTs: h.now(),
+	})
+	if err != nil {
+		log.Printf("voice: add list item: %v", err)
+		return "не получилось добавить в список.", true
+	}
+	if !res.Created {
+		return cap.Item + " уже в списке.", true
+	}
+	return "добавила в список: " + cap.Item + ".", true
+}
+
+// removeListItem crosses one named item off. It reports false when the list
+// holds nothing by that name, which is what keeps the marker words from
+// swallowing ordinary notes.
+func (h *reactiveHandler) removeListItem(ctx context.Context, cap router.ListCapture) (string, bool) {
+	items, err := h.dataStore.ListItems(ctx, cap.List, "")
+	if err != nil {
+		log.Printf("voice: list items: %v", err)
+		return "", false
+	}
+	want := store.NormalizeTaskText(cap.Item)
+	for _, li := range items {
+		if store.NormalizeTaskText(li.Item) != want {
+			continue
+		}
+		if err := h.dataStore.SetListItemStatus(ctx, li.ID, store.ListItemDone, h.now()); err != nil {
+			log.Printf("voice: cross off list item: %v", err)
+			return "не получилось обновить список.", true
+		}
+		return "вычеркнула: " + li.Item + ".", true
+	}
+	return "", false
+}
+
+// queryList — "что в списке покупок?", "что мне купить?".
+//
+// A query source, so it sits in querySources and either claims the turn or
+// passes it on. It is before the recall sources for the reason every specific
+// source is: the notes pass would otherwise answer a list question with
+// whatever note is nearest.
+func (h *reactiveHandler) queryList(ctx context.Context, t *queryTurn) (string, bool) {
+	list, ok := router.ParseListQuery(t.dec.Utterance)
+	if !ok || h.dataStore == nil {
+		return "", false
+	}
+	items, err := h.dataStore.ListItems(ctx, list, "")
+	if err != nil {
+		log.Printf("voice: list items: %v", err)
+		return "не получилось посмотреть список.", true
+	}
+	return formatListRU(list, items), true
+}
+
+// formatListRU reads a list aloud. One sentence, comma-separated, because a
+// shopping list is heard in a shop and a numbered recital is unusable there.
+func formatListRU(list string, items []store.ListItem) string {
+	name := "списке " + listGenitive(list)
+	if len(items) == 0 {
+		return "в " + name + " пусто."
+	}
+	names := make([]string, 0, len(items))
+	for _, li := range items {
+		names = append(names, li.Item)
+	}
+	return "в " + name + ": " + strings.Join(names, ", ") + "."
+}
+
+// listGenitive puts a list tag into the case "список <…>" needs. Russian
+// declines the noun and she must not say "в списке покупки".
+func listGenitive(list string) string {
+	switch list {
+	case "покупки":
+		return "покупок"
+	case "аптека":
+		return "аптеки"
+	case "хозяйство":
+		return "хозяйства"
+	}
+	return list
+}

cmd/mavend/actions_list_test.go

@@ -0,0 +1,184 @@
+package main
+
+import (
+	"context"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/kami/maven/internal/router"
+	"github.com/kami/maven/internal/store"
+)
+
+func listNow() time.Time { return time.Date(2026, 8, 4, 9, 0, 0, 0, time.UTC) }
+
+func listHandler(t *testing.T) *reactiveHandler {
+	t.Helper()
+	return &reactiveHandler{dataStore: newTestStore(t), now: listNow}
+}
+
+func say(t *testing.T, h *reactiveHandler, utterance string) (string, bool) {
+	t.Helper()
+	return h.captureListFromNote(context.Background(), router.Decision{
+		Intent: router.IntentNote, Utterance: utterance,
+	})
+}
+
+func TestListCaptureAddsAndReadsBack(t *testing.T) {
+	h := listHandler(t)
+	for _, u := range []string{"добавь в список покупок молоко", "добавь в список хлеб"} {
+		if reply, ok := say(t, h, u); !ok {
+			t.Fatalf("%q was not claimed (reply %q)", u, reply)
+		}
+	}
+	if reply, ok := say(t, h, "добавь в список покупок молоко"); !ok || !strings.Contains(reply, "уже") {
+		t.Errorf("second молоко replied %q, %v; want an already-there answer", reply, ok)
+	}
+	answer, ok := h.queryList(context.Background(), &queryTurn{
+		dec: router.Decision{Intent: router.IntentQuery, Utterance: "что в списке покупок?"},
+	})
+	if !ok {
+		t.Fatal("the list question was not claimed")
+	}
+	if !strings.Contains(answer, "молоко") || !strings.Contains(answer, "хлеб") {
+		t.Errorf("answer %q; want both items", answer)
+	}
+	if strings.Contains(answer, "списке покупки") {
+		t.Errorf("answer %q declines the list name wrong", answer)
+	}
+}
+
+// An utterance with no list marker is a note and must stay one, whichever half
+// of the parser it brushes against.
+func TestListCapturePassesOrdinaryNotes(t *testing.T) {
+	h := listHandler(t)
+	for _, u := range []string{
+		"молоко закончилось",
+		"надо бы съездить в магазин",
+		"купил новый ноутбук",
+		"добавь в список покупок",
+	} {
+		if reply, ok := say(t, h, u); ok {
+			t.Errorf("%q was claimed as a list turn: %q", u, reply)
+		}
+	}
+}
+
+func TestListCrossOffOneItemAndThenAll(t *testing.T) {
+	h := listHandler(t)
+	for _, u := range []string{
+		"добавь в список покупок молоко",
+		"добавь в список покупок хлеб",
+		"добавь в список аптеки бинт",
+	} {
+		if _, ok := say(t, h, u); !ok {
+			t.Fatalf("%q was not claimed", u)
+		}
+	}
+	reply, ok := say(t, h, "вычеркни молоко")
+	if !ok || !strings.Contains(reply, "молоко") {
+		t.Fatalf("cross off replied %q, %v", reply, ok)
+	}
+	open, err := h.dataStore.ListItems(context.Background(), "покупки", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(open) != 1 || open[0].Item != "хлеб" {
+		t.Fatalf("open list %+v; want only хлеб", open)
+	}
+	if reply, ok := say(t, h, "всё купил"); !ok || !strings.Contains(reply, "пустой") {
+		t.Errorf("clear replied %q, %v", reply, ok)
+	}
+	open, err = h.dataStore.ListItems(context.Background(), "покупки", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(open) != 0 {
+		t.Errorf("%d items still open after всё купил", len(open))
+	}
+	// The other list is untouched, and it is read back on its own.
+	answer, ok := h.queryList(context.Background(), &queryTurn{
+		dec: router.Decision{Intent: router.IntentQuery, Utterance: "покажи список аптеки"},
+	})
+	if !ok || !strings.Contains(answer, "бинт") {
+		t.Errorf("аптека answer %q, %v; want бинт", answer, ok)
+	}
+}
+
+func TestQueryListSaysWhenItIsEmpty(t *testing.T) {
+	h := listHandler(t)
+	answer, ok := h.queryList(context.Background(), &queryTurn{
+		dec: router.Decision{Intent: router.IntentQuery, Utterance: "что мне купить?"},
+	})
+	if !ok {
+		t.Fatal("the list question was not claimed")
+	}
+	if !strings.Contains(answer, "пусто") {
+		t.Errorf("empty answer %q; want it to say so", answer)
+	}
+	if _, ok := h.queryList(context.Background(), &queryTurn{
+		dec: router.Decision{Intent: router.IntentQuery, Utterance: "какие у меня задачи?"},
+	}); ok {
+		t.Error("the list source claimed a task question")
+	}
+}
+
+// Stage 0 answers a list turn without the model: the grammars route it, and the
+// action handlers re-parse what the grammar matched.
+func TestListGrammarsRouteWithoutTheModel(t *testing.T) {
+	cases := []struct {
+		utterance string
+		want      router.Intent
+	}{
+		{"добавь в список покупок молоко", router.IntentNote},
+		{"что в списке покупок?", router.IntentQuery},
+		{"всё купил", router.IntentNote},
+	}
+	for _, c := range cases {
+		var got router.Intent
+		claimed := false
+		for _, g := range router.ListGrammars() {
+			m := g.Pattern.FindStringSubmatch(c.utterance)
+			if m == nil {
+				continue
+			}
+			if dec, ok := g.Build(m); ok {
+				got, claimed = dec.Intent, true
+				break
+			}
+		}
+		if !claimed {
+			t.Errorf("no list grammar claimed %q", c.utterance)
+			continue
+		}
+		if got != c.want {
+			t.Errorf("%q routed to %v; want %v", c.utterance, got, c.want)
+		}
+	}
+	for _, g := range router.ListGrammars() {
+		m := g.Pattern.FindStringSubmatch("напомни купить молоко завтра")
+		if m == nil {
+			continue
+		}
+		if _, ok := g.Build(m); ok {
+			t.Errorf("grammar %s claimed a reminder", g.Name)
+		}
+	}
+}
+
+func TestListStoreSourceIsVoice(t *testing.T) {
+	h := listHandler(t)
+	if _, ok := say(t, h, "добавь в список покупок молоко"); !ok {
+		t.Fatal("not claimed")
+	}
+	items, err := h.dataStore.ListItems(context.Background(), "покупки", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(items) != 1 || items[0].Source != "tap:voice" {
+		t.Errorf("stored %+v; want one row from tap:voice", items)
+	}
+	if items[0].Status != store.ListItemOpen {
+		t.Errorf("status %q; want open", items[0].Status)
+	}
+}

cmd/mavend/actions_note.go

@@ -17,6 +17,12 @@ func (h *reactiveHandler) actionNote(ctx context.Context, dec router.Decision) s
 	if reply, ok := h.captureTaskFromNote(ctx, dec); ok {
 		return reply
 	}
+	// A standing list is neither work nor recall (Vikunja #453). Checked here
+	// for the same reason and at the same cost: before the embedding is paid
+	// for, and it passes the turn straight back when no marker matches.
+	if reply, ok := h.captureListFromNote(ctx, dec); ok {
+		return reply
+	}
 	// embed the note text with the same model the classifier uses, persist
 	// via CoreAPI (source=tap:voice). Semantic recall lives in `notes`, not
 	// facts — no predicate reads it (spec's two-memory split).

cmd/mavend/actions_query.go

@@ -13,6 +13,7 @@ import (
 	"github.com/kami/maven/internal/ipc"
 	"github.com/kami/maven/internal/memory"
 	"github.com/kami/maven/internal/morning"
+	"github.com/kami/maven/internal/phraser"
 	"github.com/kami/maven/internal/router"
 	"github.com/kami/maven/internal/rss"
 	"github.com/kami/maven/internal/store"
@@ -84,6 +85,11 @@ var querySources = []querySource{
 	// the money facts the poller wrote, and the notes pass would otherwise
 	// answer it from whatever he once said about spending. Its matcher needs a
 	// money noun plus an actual ask, so "я потратил весь день" is untouched.
+	// Next to "tasks" and for the same reason: "что мне купить?" is a question
+	// about the shopping list, and the recall pass would otherwise answer it
+	// from an old note about the shop. Its matcher needs an explicit list
+	// marker, so "надо бы съездить в магазин" is untouched.
+	{name: "list", answer: (*reactiveHandler).queryList},
 	{name: "money", answer: (*reactiveHandler).queryMoney},
 	// Before the recall sources and before general knowledge: "что нового?" is
 	// a question about the feeds she reads, and general knowledge would answer
@@ -433,6 +439,14 @@ func (h *reactiveHandler) queryMemory(ctx context.Context, t *queryTurn) (string
 		return "", false
 	}
 	text := hit.Meta["text"]
+	// The score cleared the gate and the topic still has to match (#470). A
+	// note about his slow network scored high enough to answer "почему небо
+	// синее?", because the right-note and must-be-silent score ranges overlap
+	// and no threshold sits between them.
+	if !memory.RecallAllowed(t.dec.Utterance, text) {
+		log.Printf("voice: recall %q rejected for %q: a world question and no shared topic word", text, t.dec.Utterance)
+		return "", false
+	}
 	// A note is phrased in Maven's voice; a fact is read back as it was
 	// stored.
 	if hit.Meta["type"] == "note" {
@@ -468,6 +482,12 @@ func (h *reactiveHandler) queryNotes(ctx context.Context, t *queryTurn) (string,
 	if !memory.ConfidentScores(noteScores, h.queryMinScore, h.queryMinMargin) {
 		return "", false
 	}
+	// Same topic veto as queryMemory above: the best note must be about what
+	// he asked, not merely the nearest vector in the index.
+	if !memory.RecallAllowed(t.dec.Utterance, notes[0].Text) {
+		log.Printf("voice: note %q rejected for %q: a world question and no shared topic word", notes[0].Text, t.dec.Utterance)
+		return "", false
+	}
 	texts := make([]string, len(notes))
 	for i, n := range notes {
 		texts[i] = n.Text
@@ -523,10 +543,7 @@ func (h *reactiveHandler) queryWeb(ctx context.Context, t *queryTurn) (string, b
 	// the question he actually asked. She answers the question, she does not
 	// recite the page.
 	snippet := page.Title + "\n" + crawl.TrimRunes(page.Text, webPageContextRunes)
-	reply, perr := h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{snippet})
-	if perr != nil {
-		log.Printf("voice: web: phrase: %v", perr)
-	}
+	reply := h.phraseSource(ctx, "web", t.dec.Utterance, []string{snippet})
 	if reply == "" {
 		// No phraser (or it failed): read back the top of the page rather than
 		// pretend the fetch did not happen.
@@ -592,14 +609,7 @@ func (h *reactiveHandler) querySearch(ctx context.Context, t *queryTurn) (string
 	// question he asked, not something to recite. The trim is one budget over the
 	// joined block, so a long first snippet cannot crowd out the rest.
 	evidence := crawl.TrimRunes(strings.Join(resp.Snippets(), "\n"), h.search.runes)
-	var reply string
-	if h.phraser != nil {
-		var perr error
-		reply, perr = h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{evidence})
-		if perr != nil {
-			log.Printf("voice: search: phrase: %v", perr)
-		}
-	}
+	reply := h.phraseSource(ctx, "search", t.dec.Utterance, []string{evidence})
 	if reply == "" {
 		// No phraser, or it failed. Read back the best evidence rather than
 		// pretend the search did not happen.
@@ -680,14 +690,7 @@ func (h *reactiveHandler) queryKiwix(ctx context.Context, t *queryTurn) (string,
 	// Handed over the same way a note or a page is: context for the question he
 	// asked, not something to recite.
 	snippet := top.Title + "\n" + crawl.TrimRunes(page.Text, h.kiwix.runes)
-	var reply string
-	if h.phraser != nil {
-		var perr error
-		reply, perr = h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{snippet})
-		if perr != nil {
-			log.Printf("voice: kiwix: phrase: %v", perr)
-		}
-	}
+	reply := h.phraseSource(ctx, "kiwix", t.dec.Utterance, []string{snippet})
 	if reply == "" {
 		// No phraser, or it failed. Read back the best hit rather than pretend
 		// the search did not happen.
@@ -717,7 +720,7 @@ func (h *reactiveHandler) queryKiwix(ctx context.Context, t *queryTurn) (string,
 // not be sent to an upstream engine at all. The guard closes both holes with
 // the same test.
 func (h *reactiveHandler) queryPersonal(ctx context.Context, t *queryTurn) (string, bool) {
-	if !isPersonalQuery(t.dec.Utterance) {
+	if !h.isPersonalTurn(ctx, t) {
 		return "", false
 	}
 	log.Printf("voice: %q is about him and his own data did not answer it; not asking the world", t.dec.Utterance)
@@ -742,7 +745,9 @@ var personalMarkers = []*regexp.Regexp{
 	regexp.MustCompile(`(?i)\bdid\s+i\b`),
 }
 
-// isPersonalQuery reports whether the utterance asks about something of his.
+// isPersonalQuery — the offline floor under the boundary. Possession only, and
+// deliberately still narrow: it answers when there is no embedder to ask, and a
+// broad guess made blind is worse than a narrow one.
 func isPersonalQuery(utterance string) bool {
 	if utterance == "" {
 		return false
@@ -755,11 +760,45 @@ func isPersonalQuery(utterance string) bool {
 	return false
 }
 
-// queryGeneral — general knowledge from the phraser, the last source before
-// giving up. It always claims: either the model answers or Maven says she
-// doesn't know.
+// isPersonalTurn — the boundary test. The seeds decide when the embedder is
+// there, which is every deployed box; the possession markers are the floor
+// underneath, for a handler with no embedder or a turn whose vector never got
+// computed. Same shape as the cascade: the better test leads, the offline one
+// always answers.
+func (h *reactiveHandler) isPersonalTurn(ctx context.Context, t *queryTurn) bool {
+	h.boundary.load(ctx, h.embedder)
+	if personal, world, ok := h.boundary.score(t.vec); ok {
+		if personal > world {
+			log.Printf("voice: %q scores personal %.4f vs world %.4f", t.dec.Utterance, personal, world)
+			return true
+		}
+		return false
+	}
+	return isPersonalQuery(t.dec.Utterance)
+}
+
+// queryGeneral — general knowledge, the last source before giving up. It always
+// claims: either a model answers, or Maven names the gap, or she says she does
+// not know.
+//
+// This is the sharpest case for the naming half. Nothing has been fetched, so
+// there is no passage to fall back on and no floor under the answer except the
+// model's weights — and a 1.7B's weights are where the invented answers come
+// from. With a workstation configured and asleep he is told that, rather than
+// told something false in a confident voice. With no workstation configured at
+// all the resident model answers exactly as it does today: naming a gap requires
+// a gap, and on that box the 1.7B is the whole product.
 func (h *reactiveHandler) queryGeneral(ctx context.Context, t *queryTurn) (string, bool) {
-	reply, err := h.phraser.PhraseQuery(ctx, t.dec.Utterance, nil)
+	if h.phraser == nil {
+		// No model of any size. That is not the workstation being asleep, so it
+		// is not that gap: it is simply not knowing.
+		return "не знаю.", true
+	}
+	reply, err := h.phraseWorld(ctx, t.dec.Utterance, nil)
+	if errors.Is(err, phraser.ErrNoWorldModel) {
+		log.Printf("voice: %q needs the world model and it is not available", t.dec.Utterance)
+		return worldGap, true
+	}
 	if err != nil || reply == "" {
 		return "не знаю.", true
 	}

cmd/mavend/actions_query_personal_test.go

@@ -19,6 +19,10 @@ func TestIsPersonalQuery(t *testing.T) {
 		"when is my meeting",
 		"do i have anything today",
 		"did i take my vitamins",
+		// Speech, but only the forms possession already covers ("did i").
+		// The verb forms the floor cannot see are the seeds' job, scored in
+		// TestONNXPersonalBoundary.
+		"what did i say about backups",
 	} {
 		if !isPersonalQuery(s) {
 			t.Errorf("isPersonalQuery(%q) = false, want true", s)
@@ -33,6 +37,10 @@ func TestIsPersonalQuery(t *testing.T) {
 		"почему небо синее",
 		"столица франции",
 		"how do i boil an egg",
+		// The floor is possession-only by design: a speech verb it cannot see
+		// passes here and is caught by the seeds instead.
+		"что я говорил про бэкапы?",
+		"как я говорил, почему небо синее",
 		"",
 	} {
 		if isPersonalQuery(s) {

cmd/mavend/capture.go

@@ -100,7 +100,7 @@ func (l llmCompleter) Complete(ctx context.Context, system, user string) (string
 // grammar, or a llama-server too old to honour one, gets the plain text it used
 // to get rather than an empty meeting summary.
 func unwrapSummary(raw string) string {
-	s := stripThink(strings.TrimSpace(raw))
+	s := phraser.StripThink(strings.TrimSpace(raw))
 	start := strings.Index(s, "{")
 	end := strings.LastIndex(s, "}")
 	if start < 0 || end <= start {

cmd/mavend/clarify_test.go

@@ -465,3 +465,39 @@ func TestExpiryNoticeSurvivesAConfirmTurn(t *testing.T) {
 		t.Fatal("the expired question must be gone")
 	}
 }
+
+// The other half of the subject question: his answer must fill the empty slot,
+// not replace the request. Slots.Text used to be the whole raw utterance for
+// every intent, so the branch that fills a text slot could only ever overwrite
+// (Vikunja #383). Here the parked request holds the hour and the answer holds
+// what to say at it, and the reminder that lands has both.
+func TestClarifySubjectAnswerFillsRatherThanClobbers(t *testing.T) {
+	ctx := context.Background()
+	h, st, _ := newClarifyHandler(t)
+	at := h.now().Add(2 * time.Hour)
+
+	question, asked := h.askClarify(clarifyDec(router.IntentReminder,
+		router.Slots{Time: at, HasTime: true}, "напомни в 11"))
+	if !asked || question != "О чём напомнить?" {
+		t.Fatalf("expected the subject question, got %q asked=%v", question, asked)
+	}
+
+	reply, handled := h.resolveClarifyAnswer(ctx, "позвонить маме")
+	if !handled {
+		t.Fatal("the answer to an open question must be consumed as an answer")
+	}
+	if reply == clarifyGaveUp {
+		t.Fatalf("a good answer must not drop the request: %q", reply)
+	}
+
+	reminders, err := st.DueReminders(ctx, h.now().Add(48*time.Hour))
+	if err != nil || len(reminders) != 1 {
+		t.Fatalf("clarified reminder was not created: reminders=%v err=%v", reminders, err)
+	}
+	if !strings.Contains(reminders[0].Payload, "маме") {
+		t.Fatalf("the answer never reached the reminder: %q", reminders[0].Payload)
+	}
+	if !strings.Contains(reminders[0].Payload, "11") {
+		t.Fatalf("the answer clobbered the original request: %q", reminders[0].Payload)
+	}
+}

cmd/mavend/factgate.go

@@ -0,0 +1,82 @@
+package main
+
+import (
+	"log"
+	"strings"
+	"unicode"
+)
+
+// ungroundedConfidence — what a self fact is worth when its value appears
+// nowhere in what he said. Below `query_min_score` is not the point (recall
+// gates on vector distance, not on this number); the point is that
+// `/history` and every future reader can tell a value he said from a value
+// the model supplied.
+const ungroundedConfidence = 0.6
+
+// factConfidence scores a self fact by whether its value is grounded in the
+// utterance it came from. Grounded stays 1.00, which is what a tapped fact
+// has always been worth. Ungrounded drops, and says so in the log.
+//
+// An empty value is grounded by definition: the key alone carries the fact
+// ("поужинал"), and there is nothing for the model to have invented.
+func factConfidence(utterance, value string) float64 {
+	if strings.TrimSpace(value) == "" {
+		return 1.0
+	}
+	if valueGrounded(utterance, value) {
+		return 1.0
+	}
+	log.Printf("voice: fact value %q is not in %q — writing at confidence %.2f",
+		value, utterance, ungroundedConfidence)
+	return ungroundedConfidence
+}
+
+// valueGrounded reports whether every word of value traces back to a word he
+// actually said. The comparison is on a 4-rune prefix, so the model's
+// normalization survives ("пил воду" → "вода") while an invented value
+// ("1.20" for a question about Go) does not.
+func valueGrounded(utterance, value string) bool {
+	said := factTokens(utterance)
+	words := factTokens(value)
+	if len(words) == 0 {
+		return true
+	}
+	for _, w := range words {
+		if !anyTokenMatches(said, w) {
+			return false
+		}
+	}
+	return true
+}
+
+func anyTokenMatches(said []string, w string) bool {
+	for _, s := range said {
+		if s == w || sameStem(s, w) {
+			return true
+		}
+	}
+	return false
+}
+
+// sameStem is inflection tolerance and nothing more: it compares all but the
+// last rune of the shorter word, and never fewer than three. Russian marks
+// case on the ending, so "пил воду" and the stored "вода" are the same word he
+// said, while "1.20" and "версия" are not. A word of three runes or fewer must
+// match outright, where a shorter prefix would match half the language.
+func sameStem(a, b string) bool {
+	ar, br := []rune(a), []rune(b)
+	shorter := min(len(ar), len(br))
+	n := shorter - 1
+	if n < 3 || len(ar) < n || len(br) < n {
+		return false
+	}
+	return string(ar[:n]) == string(br[:n])
+}
+
+// factTokens lowercases and splits on everything that is not a letter or a
+// digit, the same shape planTokens uses in the router.
+func factTokens(s string) []string {
+	return strings.FieldsFunc(strings.ToLower(s), func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	})
+}

cmd/mavend/factgate_test.go

@@ -0,0 +1,125 @@
+package main
+
+import (
+	"context"
+	"testing"
+	"time"
+
+	"github.com/kami/maven/internal/ipc"
+	"github.com/kami/maven/internal/memory"
+	"github.com/kami/maven/internal/router"
+	"github.com/kami/maven/internal/tool"
+	"github.com/kami/maven/internal/voice"
+)
+
+func newFactGateHandler(t *testing.T, now time.Time) (*reactiveHandler, ipc.CoreAPI) {
+	t.Helper()
+	st := newTestStore(t)
+	api := ipc.NewStoreAPI(st)
+	emb := router.NewHashEmbedder(1024)
+	h := &reactiveHandler{
+		api:       api,
+		embedder:  emb,
+		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil),
+		replier:   voice.NewStubReplier(),
+		now:       func() time.Time { return now },
+		memStore:  memory.NewInMemoryStore(),
+		dataStore: st,
+	}
+	return h, api
+}
+
+// The write half of #470: a question routed to IntentFact must not become a
+// fact about him, and must not leave a vector behind for recall to serve.
+func TestActionFact_QuestionIsNotWritten(t *testing.T) {
+	ctx := context.Background()
+	h, api := newFactGateHandler(t, time.Now())
+
+	reply := h.actionFact(ctx, router.Decision{
+		Intent:    router.IntentFact,
+		Utterance: "какая последняя версия языка Go?",
+		Slots:     router.Slots{Key: "go_version", HasKey: true, Value: `"1.20"`},
+	})
+
+	if _, err := api.LatestFact(ctx, "go_version"); err == nil {
+		t.Fatal("a question was stored as a fact about him")
+	}
+	hits, err := h.memStore.Search(ctx, mustEmbedPassage(t, h, "какая последняя версия языка Go?"), 3)
+	if err != nil {
+		t.Fatalf("memory search: %v", err)
+	}
+	if len(hits) != 0 {
+		t.Fatalf("the question was indexed for recall: %+v", hits)
+	}
+	// It went down the query chain instead. Nothing is configured to answer a
+	// world question in this harness, so "не знаю." is the honest outcome —
+	// what matters is that the turn was answered, not stored.
+	if reply == "" {
+		t.Fatal("the turn was neither stored nor answered")
+	}
+}
+
+// The capture that must survive the gate: an explicit instruction to record,
+// even though it contains an interrogative.
+func TestActionFact_ExplicitCaptureStillWrites(t *testing.T) {
+	ctx := context.Background()
+	h, api := newFactGateHandler(t, time.Now())
+
+	h.actionFact(ctx, router.Decision{
+		Intent:    router.IntentFact,
+		Utterance: "запиши что я пил воду",
+		Slots:     router.Slots{Key: "water", HasKey: true, Value: `"вода"`},
+	})
+
+	f, err := api.LatestFact(ctx, "water")
+	if err != nil {
+		t.Fatalf("an explicit capture was refused: %v", err)
+	}
+	if f.Confidence != 1.0 {
+		t.Errorf("confidence = %v, want 1.0 for a value he said", f.Confidence)
+	}
+	// #493: what recall reads back is the fact, not the sentence he said.
+	// queryMemory returns a fact's text verbatim, so the utterance sitting here
+	// meant "запиши что я пил воду" was the answer to "когда я пил воду?".
+	hits, err := h.memStore.Search(ctx, mustEmbedPassage(t, h, "вода"), 3)
+	if err != nil {
+		t.Fatalf("memory search: %v", err)
+	}
+	if len(hits) != 1 {
+		t.Fatalf("the fact was not indexed once: %+v", hits)
+	}
+	if got := hits[0].Meta["text"]; got != "water — вода" {
+		t.Errorf("indexed text = %q, want the fact", got)
+	}
+	if got := hits[0].Meta["utterance"]; got != "запиши что я пил воду" {
+		t.Errorf("utterance provenance = %q, want it kept alongside", got)
+	}
+}
+
+func TestFactConfidence(t *testing.T) {
+	cases := []struct {
+		utterance, value string
+		want             float64
+	}{
+		{"запиши что я пил воду", `"вода"`, 1.0},
+		{"я выпил кофе", `"кофе"`, 1.0},
+		{"поужинал", "", 1.0},
+		{"отметь что я полил кактус", `"полил кактус"`, 1.0},
+		{"какая последняя версия языка Go", `"1.20"`, ungroundedConfidence},
+		{"кто премьер Японии", `"Тонио Озаки"`, ungroundedConfidence},
+	}
+	for _, c := range cases {
+		if got := factConfidence(c.utterance, c.value); got != c.want {
+			t.Errorf("factConfidence(%q, %q) = %v, want %v", c.utterance, c.value, got, c.want)
+		}
+	}
+}
+
+func mustEmbedPassage(t *testing.T, h *reactiveHandler, text string) []float32 {
+	t.Helper()
+	vec, err := router.EmbedQuery(context.Background(), h.embedder, text)
+	if err != nil {
+		t.Fatalf("embed %q: %v", text, err)
+	}
+	return vec
+}

cmd/mavend/main.go

@@ -251,6 +251,7 @@ func run(args []string) error {
 				Listen:       cfg.Phraser.Listen,
 				NGpuLayers:   cfg.Phraser.NGpuLayers,
 				NCtx:         cfg.Phraser.NCtx,
+				CacheRAMMiB:  cacheRAMMiB(cfg.Phraser.CacheRAMMiB),
 				Timeout:      time.Duration(cfg.Phraser.Timeout),
 				LLMNudges:    cfg.Phraser.LLMNudges,
 				ContextBlock: contextBlockFn(cfg, time.Now),
@@ -525,6 +526,7 @@ func run(args []string) error {
 					Listen:       cfg.Phraser.Listen,
 					NGpuLayers:   cfg.Phraser.NGpuLayers,
 					NCtx:         cfg.Phraser.NCtx,
+					CacheRAMMiB:  cacheRAMMiB(cfg.Phraser.CacheRAMMiB),
 					Timeout:      time.Duration(cfg.Phraser.Timeout),
 					LLMNudges:    cfg.Phraser.LLMNudges,
 					ContextBlock: contextBlockFn(cfg, time.Now),
@@ -788,6 +790,22 @@ func personaFacts(cfg *config.Config) persona.Facts {
 	return f
 }
 
+// cacheRAMMiB resolves phraser.cache_ram_mib into the phraser's field. Unset
+// means 512 MiB and not "whatever the server does", because the server's own
+// default is 8 GiB of prompt cache and that is what put 7.9 GB of RSS and half
+// a gigabyte of swap on homesrv for a 1.1 GB model. A negative value is the
+// deliberate opt-out: no flag is passed, the server's default applies, and the
+// operator owns the consequence.
+func cacheRAMMiB(configured int) int {
+	if configured == 0 {
+		return 512
+	}
+	if configured < 0 {
+		return 0
+	}
+	return configured
+}
+
 // contextBlockFn returns the per-turn renderer of the shared context block.
 // Per turn, not once at startup, because the block states the current time.
 func contextBlockFn(cfg *config.Config, now func() time.Time) func() string {

cmd/mavend/modelseam_test.go

@@ -0,0 +1,91 @@
+package main
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/kami/maven/internal/config"
+	"github.com/kami/maven/internal/llm"
+)
+
+// No `workstation` block is the shipping deploy. The seam must then be the
+// resident client itself, with nothing probing anything.
+func TestModelSeamUnconfiguredIsResidentOnly(t *testing.T) {
+	resident := llm.New("http://127.0.0.1:1", time.Second)
+	hot, pair := modelSeam(&config.Config{}, resident)
+	if pair != nil {
+		t.Error("built a pair with no workstation configured")
+	}
+	if hot == nil {
+		t.Fatal("no seam at all, so the cascade would route with the classifier")
+	}
+}
+
+// A workstation with no resident model behind it has no floor, and a Pair with
+// no floor is a configuration mistake rather than a degraded mode.
+func TestModelSeamWithoutResidentIsNil(t *testing.T) {
+	cfg := &config.Config{Workstation: &config.WorkstationConfig{URL: "http://127.0.0.1:1"}}
+	cfg.Workstation.Health = strings.TrimRight(cfg.Workstation.URL, "/") + "/health"
+	hot, pair := modelSeam(cfg, nil)
+	if hot != nil || pair != nil {
+		t.Errorf("built a seam with no floor: hot=%v pair=%v", hot, pair)
+	}
+}
+
+// The configured case: the seam is the pair, and the pair notices a workstation
+// that answers /health.
+func TestModelSeamPrefersAnAnsweringWorkstation(t *testing.T) {
+	up := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.WriteHeader(http.StatusOK)
+	}))
+	defer up.Close()
+
+	cfg := &config.Config{Workstation: &config.WorkstationConfig{
+		URL:   up.URL,
+		Probe: config.Duration(10 * time.Millisecond),
+	}}
+	cfg.Workstation.Health = strings.TrimRight(cfg.Workstation.URL, "/") + "/health"
+
+	hot, pair := modelSeam(cfg, llm.New("http://127.0.0.1:1", time.Second))
+	if pair == nil || hot == nil {
+		t.Fatal("no pair built for a configured workstation")
+	}
+	defer pair.Stop()
+
+	deadline := time.Now().Add(2 * time.Second)
+	for !pair.Available() && time.Now().Before(deadline) {
+		time.Sleep(5 * time.Millisecond)
+	}
+	if !pair.Available() {
+		t.Fatal("the pair never saw a workstation that answers /health")
+	}
+}
+
+// A card held by a CPT run answers 503, and that must read as unavailable
+// rather than as an error a turn has to handle.
+func TestModelSeamHeldCardIsUnavailable(t *testing.T) {
+	busy := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		http.Error(w, "model not loaded", http.StatusServiceUnavailable)
+	}))
+	defer busy.Close()
+
+	cfg := &config.Config{Workstation: &config.WorkstationConfig{
+		URL:   busy.URL,
+		Probe: config.Duration(10 * time.Millisecond),
+	}}
+	cfg.Workstation.Health = strings.TrimRight(cfg.Workstation.URL, "/") + "/health"
+
+	_, pair := modelSeam(cfg, llm.New("http://127.0.0.1:1", time.Second))
+	if pair == nil {
+		t.Fatal("no pair built for a configured workstation")
+	}
+	defer pair.Stop()
+
+	time.Sleep(50 * time.Millisecond)
+	if pair.Available() {
+		t.Error("a 503 from the supervisor read as available")
+	}
+}

cmd/mavend/personaguard.go

@@ -0,0 +1,134 @@
+package main
+
+import (
+	"context"
+	"log"
+	"regexp"
+	"strings"
+	"sync"
+
+	"github.com/kami/maven/internal/delivery"
+	"github.com/kami/maven/internal/loop"
+	"github.com/kami/maven/internal/phraser"
+	"github.com/kami/maven/internal/phraser/eval"
+)
+
+// The persona checks, run before she speaks (Vikunja #399).
+//
+// RunChecks and RunTalkChecks only ever ran from the eval package, so
+// everything the fixtures measured was offline knowledge: we could say "about
+// one reply in three is broken" and still ship every one of them. This runs the
+// cheap half of that on the live path, and replaces a failing message with the
+// deterministic floor.
+//
+// Which checks: the unambiguous string tests only — feminine self-reference,
+// how she addresses him, and a leaked-reasoning test. Not length, which is
+// path-specific, and not ontopic, which compares against fragments the fixture
+// supplies and runtime does not have. Not hisgender either — see guardSpoken.
+//
+// No retry. A retry doubles the latency on the exact turn that is already going
+// badly, and on the nudge path the moment has passed.
+//
+// The known cost, written down because it is real: a wrongly flagged good reply
+// is replaced by a flatter stub one. That is the right trade — a stub sentence
+// is dull, a leaked reasoning trace is broken — but it means these checks can
+// no longer be tuned for sensitivity alone.
+
+// checkLeak — the name reported when the model's scaffolding reaches the text.
+const checkLeak = "leak"
+
+// leakPatterns — reasoning and protocol that belongs to the model, not to him.
+// The resident model is a Thinking variant, so an unclosed reasoning block is
+// the failure mode, not a hypothetical (Vikunja #398).
+var leakPatterns = []*regexp.Regexp{
+	regexp.MustCompile(`(?i)<\s*/?\s*think`),
+	regexp.MustCompile(`(?i)thinking\s*(process|:)`),
+	regexp.MustCompile(`(?i)^\s*(assistant|user|system)\s*:`),
+	// Raw contract JSON: the parser already unwraps a good one, so a body that
+	// still carries the keys is one it could not read.
+	regexp.MustCompile(`"(response|mood|body|summary)"\s*:`),
+	// The persona block quoted back at him.
+	regexp.MustCompile(`(?i)(ты\s+—?\s*мэйвен|системный промпт|system prompt)`),
+}
+
+// checkPersonaLeak reports whether the model's own scaffolding is in the text.
+func checkPersonaLeak(body string) (string, bool) {
+	for _, re := range leakPatterns {
+		if m := re.FindString(body); m != "" {
+			return "leaked " + strings.TrimSpace(m), false
+		}
+	}
+	return "", true
+}
+
+// personaRejects counts what the guard caught, by check name, so the real
+// production rate is knowable rather than inferred from the fixture.
+var personaRejects = struct {
+	mu sync.Mutex
+	by map[string]int
+}{by: map[string]int{}}
+
+func personaRejectCounts() map[string]int {
+	personaRejects.mu.Lock()
+	defer personaRejects.mu.Unlock()
+	out := make(map[string]int, len(personaRejects.by))
+	for k, v := range personaRejects.by {
+		out[k] = v
+	}
+	return out
+}
+
+// guardSpoken checks a phrased message. It returns the failed check and false
+// when the message must not be said; path names the caller, for the log.
+//
+// An empty message passes: the caller already treats that as a failure and
+// falls back on its own, and reporting it as a persona breach would put a
+// misleading line in the count.
+func guardSpoken(path, body string) (string, bool) {
+	if strings.TrimSpace(body) == "" {
+		return "", true
+	}
+	if detail, ok := checkPersonaLeak(body); !ok {
+		return rejectSpoken(path, checkLeak, detail, body), false
+	}
+	// Feminine and address only. HisGender is not run here: it reads a
+	// sentence-initial feminine verb with no pronoun — "записала, что ты выпил
+	// воды" — as a woman being addressed, when it is her own correct
+	// self-reference. Offline that is a point of score; on this path it would
+	// replace a good reply with a stub one on every fact she confirms.
+	for _, r := range []eval.Result{eval.Feminine(body), eval.Address(body)} {
+		if !r.Pass {
+			return rejectSpoken(path, r.Name, r.Detail, body), false
+		}
+	}
+	return "", true
+}
+
+// rejectSpoken logs what she nearly said and counts it. The whole text, not a
+// prefix: the point of the log line is that the failure can be read back later
+// and argued with.
+func rejectSpoken(path, check, detail, body string) string {
+	personaRejects.mu.Lock()
+	personaRejects.by[check]++
+	personaRejects.mu.Unlock()
+	log.Printf("persona: %s rejected on %s (%s): %q", path, check, detail, body)
+	return check
+}
+
+// guardNudge checks a phrased nudge and falls back to the deterministic floor
+// when it fails. The nudge path, unlike the reply path, cannot ask again: the
+// tick has already decided she speaks, so the choice is the floor's wording or
+// a broken sentence.
+func guardNudge(pn delivery.PhrasedNudge, cand loop.Candidate) delivery.PhrasedNudge {
+	if _, ok := guardSpoken("nudge", pn.Body); ok {
+		return pn
+	}
+	stub, err := phraser.NewStub().PhraseNudge(context.Background(), cand)
+	if err != nil {
+		// The Stub is templates over the candidate and does not fail. If it
+		// somehow does, the model's text is still what the rule decided to
+		// say, and saying nothing is the worse outcome.
+		return pn
+	}
+	return stub
+}

cmd/mavend/personaguard_test.go

@@ -0,0 +1,75 @@
+package main
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/kami/maven/internal/delivery"
+	"github.com/kami/maven/internal/loop"
+)
+
+func TestGuardPassesWhatSheShouldSay(t *testing.T) {
+	good := []string{
+		"записала: купить хлеб.",
+		"поняла, напомню в 11:00.",
+		"ты не пил воду с утра.",
+		"я рада, что получилось.",
+		"",
+	}
+	for _, body := range good {
+		if check, ok := guardSpoken("test", body); !ok {
+			t.Errorf("guardSpoken(%q) rejected on %s", body, check)
+		}
+	}
+}
+
+func TestGuardStopsWhatSheShouldNot(t *testing.T) {
+	bad := []struct {
+		body string
+		want string
+	}{
+		{"<think>он просил воду</think> попей воды.", checkLeak},
+		{"Thinking Process: он давно не пил.", checkLeak},
+		{`{"response": "попей воды", "mood": "neutral"}`, checkLeak},
+		{"я напомнил тебе про воду.", "feminine"},
+		{"вы давно не пили воду.", "address"},
+	}
+	for _, c := range bad {
+		check, ok := guardSpoken("test", c.body)
+		if ok {
+			t.Errorf("guardSpoken(%q) let it through", c.body)
+			continue
+		}
+		if check != c.want {
+			t.Errorf("guardSpoken(%q) failed on %s; want %s", c.body, check, c.want)
+		}
+	}
+}
+
+func TestGuardCountsWhatItCaught(t *testing.T) {
+	before := personaRejectCounts()[checkLeak]
+	if _, ok := guardSpoken("test", "<think>…"); ok {
+		t.Fatal("a leaked reasoning block was let through")
+	}
+	if after := personaRejectCounts()[checkLeak]; after != before+1 {
+		t.Errorf("leak count %d; want %d", after, before+1)
+	}
+}
+
+// TestGuardNudgeFallsBackToTheFloor — a broken nudge is replaced by the
+// deterministic wording, not dropped and not retried.
+func TestGuardNudgeFallsBackToTheFloor(t *testing.T) {
+	cand := loop.Candidate{Rule: loop.Rule{Name: "water"}}
+	bad := delivery.PhrasedNudge{Candidate: cand, Body: "Thinking Process: он не пил.", Mood: "neutral"}
+	got := guardNudge(bad, cand)
+	if got.Body == bad.Body {
+		t.Fatal("the broken nudge was delivered unchanged")
+	}
+	if strings.TrimSpace(got.Body) == "" {
+		t.Fatal("the nudge was dropped rather than re-worded")
+	}
+	good := delivery.PhrasedNudge{Candidate: cand, Body: "попей воды.", Mood: "neutral"}
+	if guardNudge(good, cand).Body != good.Body {
+		t.Error("a good nudge was replaced")
+	}
+}

cmd/mavend/personalboundary.go

@@ -0,0 +1,154 @@
+package main
+
+import (
+	"context"
+	"log"
+	"math"
+	"sync"
+
+	"github.com/kami/maven/internal/router"
+)
+
+// The personal boundary decides one thing: is this question about him. It used
+// to decide it by matching possession words, and that was the whole defect
+// behind Vikunja #495. "что я говорил про бэкапы?" is his data by definition —
+// nothing outside the box has ever heard him say anything — and it carried no
+// possession word, so it walked past the boundary into SearXNG and came back
+// answered out of a Habr article about somebody else's backups.
+//
+// The first fix was one more marker class, `я говорил|сказал|писал|…`, plus a
+// carve-out so "как я говорил, почему небо синее" stayed a world question. Both
+// halves are a lexicon, and a lexicon is the wrong instrument here: Russian
+// gives every verb a dozen surface forms, the preamble list has no end, and
+// every utterance the list misses is one that reaches the world. It also drifts
+// silently — a missing verb looks exactly like no bug.
+//
+// So the boundary asks the embedder instead. Two frozen seed sets — questions
+// about him, questions about the world — are embedded once, and the turn's own
+// query vector, already computed by queryEmbed upstream, is scored against
+// both. Nearest side wins. Word order, verb form and unseen phrasing stop
+// mattering, which is exactly what a lexicon could not do.
+//
+// Measured 03-08-2026 against multilingual-e5-small on 19 held-out utterances,
+// none of them a seed: 19 right (TestONNXPersonalBoundary). A 20th, "as i said,
+// what is the population of india", missed by +0.008 during the first pass and
+// is a world seed now, which is why it is not in the held-out set. True
+// positives clear the world side by +0.014 to +0.089 and the nearest true
+// negative sits at -0.005, so the gate is the sign of the difference and
+// nothing tighter: the margins are too thin to justify a threshold, and the
+// asymmetry favours claiming anyway. A false claim costs one honest "не знаю";
+// a false pass sends his life to an upstream engine.
+//
+// The embedder is the one model CLAUDE.md pins to homesrv permanently, and it
+// is what makes this affordable: no llama-server call, no network, one cosine
+// per seed against a vector the turn already has.
+
+// personalSeeds — questions about him. Frozen: they are scoring data, so
+// editing one moves the boundary and must be re-measured, not eyeballed. Cover
+// both classes the boundary owns, possession and first-person speech, in both
+// languages.
+var personalSeeds = []string{
+	"что я говорил про это",
+	"я тебе рассказывал об этом?",
+	"что я записал про врача",
+	"я упоминал эту тему?",
+	"что у меня сегодня",
+	"когда моя встреча",
+	"what did i say about this",
+	"did i mention this to you",
+}
+
+// worldSeeds — questions the world can answer, including the two shapes that
+// look personal and are not: a first-person preamble on a world question ("как
+// я говорил, ..."), and first person without possession ("что я могу
+// посмотреть вечером"). Refusing those is the opposite mistake and the older
+// comment on personalMarkers already named it.
+var worldSeeds = []string{
+	"почему небо синее",
+	"какая столица франции",
+	"как сварить борщ",
+	"кто написал эту книгу",
+	"what is the capital of france",
+	"how do i boil an egg",
+	"как я говорил, почему небо синее",
+	"as i said, why is the sky blue",
+	"as i said, what is the population of india",
+	"что я могу посмотреть вечером",
+	"что мне почитать про историю",
+	"что я должен знать про питон",
+	"what can i watch tonight",
+}
+
+// personalBoundary holds the embedded seeds. Zero value is usable and means
+// "not loaded yet"; a handler built without an embedder never loads and the
+// boundary falls back to personalMarkers.
+type personalBoundary struct {
+	once     sync.Once
+	personal [][]float32
+	world    [][]float32
+	loaded   bool
+}
+
+// load embeds both seed sets, once per process. Seeds are embedded on the QUERY
+// side, like the utterance they are compared with — a question against a
+// question. Mixing sides would measure the e5 prefix, not the meaning.
+func (b *personalBoundary) load(ctx context.Context, emb router.Embedder) {
+	b.once.Do(func() {
+		if emb == nil {
+			return
+		}
+		embedAll := func(ss []string) [][]float32 {
+			out := make([][]float32, 0, len(ss))
+			for _, s := range ss {
+				v, err := router.EmbedQuery(ctx, emb, s)
+				if err != nil {
+					log.Printf("voice: personal boundary seeds unavailable (%v); falling back to possession markers", err)
+					return nil
+				}
+				out = append(out, v)
+			}
+			return out
+		}
+		p, w := embedAll(personalSeeds), embedAll(worldSeeds)
+		if p == nil || w == nil {
+			return
+		}
+		b.personal, b.world, b.loaded = p, w, true
+	})
+}
+
+// score returns the best similarity to each side. ok is false when the seeds
+// are not loaded, which is the caller's signal to use the markers instead.
+func (b *personalBoundary) score(vec []float32) (personal, world float64, ok bool) {
+	if !b.loaded || len(vec) == 0 {
+		return 0, 0, false
+	}
+	best := func(seeds [][]float32) float64 {
+		m := -1.0
+		for _, s := range seeds {
+			if c := cosine(vec, s); c > m {
+				m = c
+			}
+		}
+		return m
+	}
+	return best(b.personal), best(b.world), true
+}
+
+// cosine — same math as internal/router and internal/memory, small enough that
+// importing one of them for it would be the larger coupling.
+func cosine(a, b []float32) float64 {
+	if len(a) != len(b) {
+		return 0
+	}
+	var dot, na, nb float64
+	for i := range a {
+		dot += float64(a[i]) * float64(b[i])
+		na += float64(a[i]) * float64(a[i])
+		nb += float64(b[i]) * float64(b[i])
+	}
+	if na == 0 || nb == 0 {
+		return 0
+	}
+	return dot / (math.Sqrt(na) * math.Sqrt(nb))
+}

cmd/mavend/personalboundary_test.go

@@ -0,0 +1,94 @@
+package main
+
+import (
+	"context"
+	"os"
+	"path/filepath"
+	"testing"
+
+	"github.com/kami/maven/internal/router"
+)
+
+// A handler with no embedder never loads the seeds, so the boundary falls back
+// to the possession markers. That is the offline floor and it must keep working
+// — an embedder that fails to load must not open the boundary.
+func TestBoundaryFallsBackToMarkersWithNoEmbedder(t *testing.T) {
+	h := personalHandler()
+	if !h.isPersonalTurn(context.Background(), &queryTurn{
+		dec: router.Decision{Utterance: "во сколько у меня встреча"},
+	}) {
+		t.Error("no embedder: a possession question must still be personal")
+	}
+	if h.isPersonalTurn(context.Background(), &queryTurn{
+		dec: router.Decision{Utterance: "почему небо синее"},
+	}) {
+		t.Error("no embedder: a world question must still pass")
+	}
+}
+
+// TestONNXPersonalBoundary — the number that matters, scored against the
+// embedder homesrv actually runs. Opt-in via MAVEN_ONNX_LIB, exactly like
+// TestONNXRecall in internal/memory/recalleval.
+//
+// Every case here is held out: none of these strings is a seed. The #495
+// regression is the first row — "что я говорил про бэкапы?" reached SearXNG and
+// was answered from a Habr article, and no possession word appears in it.
+func TestONNXPersonalBoundary(t *testing.T) {
+	lib := os.Getenv("MAVEN_ONNX_LIB")
+	if lib == "" {
+		t.Skip("MAVEN_ONNX_LIB unset — see AGENTS.md § Embedder model for intent routing")
+	}
+	dir := filepath.Join("../..", "models/embedder/multilingual-e5-small")
+	emb, err := router.NewONNXEmbedder(filepath.Join(dir, "model_quantized.onnx"), filepath.Join(dir, "tokenizer.json"), lib)
+	if err != nil {
+		t.Skipf("onnx embedder unavailable: %v", err)
+	}
+	defer emb.Close()
+
+	cases := []struct {
+		utterance string
+		personal  bool
+	}{
+		{"что я говорил про бэкапы?", true},
+		{"что я сказал вчера про отпуск", true},
+		{"я писал что-нибудь про сервер", true},
+		{"я упоминал про конференцию?", true},
+		{"что я отмечал по поводу переезда", true},
+		{"я рассказывал тебе про новую работу?", true},
+		{"во сколько у меня встреча", true},
+		{"когда мой следующий отпуск", true},
+		{"what did i say about backups", true},
+		{"did i tell you about the doctor", true},
+		{"как я говорил, почему небо синее", false},
+		{"как уже я говорил, какая столица франции", false},
+		{"почему трава зелёная", false},
+		{"столица франции", false},
+		{"как мне сварить борщ", false},
+		{"что мне посмотреть вечером", false},
+		{"я хочу узнать про рим", false},
+		{"кто такой гагарин", false},
+		{"how do i boil an egg", false},
+	}
+
+	h := &reactiveHandler{embedder: emb}
+	ctx := context.Background()
+	wrong := 0
+	for _, c := range cases {
+		vec, err := router.EmbedQuery(ctx, emb, c.utterance)
+		if err != nil {
+			t.Fatalf("embed %q: %v", c.utterance, err)
+		}
+		turn := &queryTurn{dec: router.Decision{Utterance: c.utterance}, vec: vec}
+		got := h.isPersonalTurn(ctx, turn)
+		p, w, ok := h.boundary.score(vec)
+		if !ok {
+			t.Fatal("seeds did not load with a working embedder")
+		}
+		if got != c.personal {
+			wrong++
+			t.Errorf("%q: personal=%v want %v (personal %.4f world %.4f)", c.utterance, got, c.personal, p, w)
+		}
+		t.Logf("personal=%-5v personal %.4f world %.4f delta %+.4f  %s", got, p, w, p-w, c.utterance)
+	}
+	t.Logf("personal boundary: %d/%d held-out utterances correct", len(cases)-wrong, len(cases))
+}

cmd/mavend/replier_llm.go

@@ -2,122 +2,38 @@ package main
 
 import (
 	"context"
-	"encoding/json"
-	"strings"
-	"time"
 
-	"github.com/kami/maven/internal/llm"
-	"github.com/kami/maven/internal/persona"
 	"github.com/kami/maven/internal/phraser"
 	"github.com/kami/maven/internal/router"
 	"github.com/kami/maven/internal/voice"
 )
 
-// completer is the LLM seam for the replier (subset of router.Completer).
-// *llm.Client satisfies it.
-type completer interface {
-	Complete(ctx context.Context, r llm.Req) (string, error)
-}
-
-// llmReplier phrases reactive confirmations with the resident model
-// (Qwen3-1.7B). Stub is the
-// floor on any error (offline-safe). Maven speaks as "she", feminine RU.
+// llmReplier is the daemon-side wiring around phraser.Replier: it owns the
+// deterministic floor, and nothing else. The phrasing itself, the prompt and the
+// output parsing live in internal/phraser so the eval can score them (#396).
 type llmReplier struct {
-	c    completer
+	p    *phraser.Replier
 	stub *voice.StubReplier
-
-	// block renders the shared context block per turn (who he is, the time).
-	// nil ⇒ the prompt stands alone.
-	block func() string
 }
 
-func newLLMReplier(c completer, block func() string) *llmReplier {
-	return &llmReplier{c: c, stub: voice.NewStubReplier(), block: block}
+func newLLMReplier(c phraser.Completer, block func() string) *llmReplier {
+	return &llmReplier{p: phraser.NewReplier(c, block), stub: voice.NewStubReplier()}
 }
 
-const replySystem = `Ты — Maven, домашняя ассистентка (о себе — в женском роде). Владелец — мужчина, говоришь с ним на "ты", в единственном числе; никогда не "вы"/"ваш" и не "он"/"его". Подтверди действие РОВНО ОДНИМ коротким предложением (≤120 символов), по-русски, спокойно и без официальных формулировок. Не задавай вопросов, не повторяй слова, не добавляй ничего после точки. Отвечай ТОЛЬКО одним объектом JSON с полями "response" (текст) и "mood" (ровно одно из: neutral, happy, thinking, tired, confused).
-Пример: {"response": "Записала, что ты выпил стакан воды.", "mood": "neutral"}
-Никогда не пиши "..." в поле response.`
-
+// Reply never fails: a clarify, a model error and an unusable generation all
+// answer from the stub, which is what keeps a turn from breaking on the model.
 func (r *llmReplier) Reply(d router.Decision) string {
 	if d.Clarify {
 		return r.stub.Reply(d)
 	}
-	ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
-	defer cancel()
-	out, err := r.c.Complete(ctx, llm.Req{
-		System:        persona.Prepend(r.block, replySystem),
-		User:          replyContext(d),
-		Grammar:       phraser.ResponseGrammar,
-		MaxTokens:     512,
-		RepeatPenalty: 1.3,
-	})
-	if err != nil {
+	out, err := r.p.PhraseReply(context.Background(), d)
+	if err != nil || out == "" {
 		return r.stub.Reply(d)
 	}
-	out = stripThink(out)
-	if response, _ := parseResponseMood(out); response != "" {
-		return response
-	}
-	// fallback: try plain-text parsing
-	if out = firstSentence(out); out != "" {
-		return out
-	}
-	return r.stub.Reply(d)
-}
-
-// firstSentence trims the model's output to a single clean confirmation: first
-// line, first sentence, whitespace-normalized — the last-line defense against a
-// small model that rambles past the first period despite the prompt + stop.
-// stripThink removes the <think> block that Thinking-variant models emit.
-func stripThink(s string) string {
-	if i := strings.LastIndex(s, "</think>"); i >= 0 {
-		s = strings.TrimSpace(s[i+8:])
-	}
-	return s
-}
-
-func firstSentence(s string) string {
-	s = strings.TrimSpace(s)
-	if i := strings.IndexByte(s, '\n'); i >= 0 {
-		s = s[:i]
-	}
-	// keep up to and including the first sentence-ending punctuation.
-	if i := strings.IndexAny(s, ".!?"); i >= 0 {
-		s = s[:i+1]
-	}
-	return strings.TrimSpace(s)
-}
-
-// parseResponseMood extracts {"response","mood"} from LLM output, tolerant
-// of thinking tokens and extra text before/after the JSON block.
-func parseResponseMood(raw string) (response, mood string) {
-	cleaned := strings.TrimSpace(raw)
-	start := strings.Index(cleaned, "{")
-	end := strings.LastIndex(cleaned, "}")
-	if start < 0 || end < 0 || end <= start {
-		return "", ""
-	}
-	var parsed struct {
-		Response string `json:"response"`
-		Mood     string `json:"mood"`
-	}
-	if err := json.Unmarshal([]byte(cleaned[start:end+1]), &parsed); err != nil {
-		return "", ""
-	}
-	return parsed.Response, parsed.Mood
-}
-
-// replyContext renders the decision into a compact RU description for the model.
-func replyContext(d router.Decision) string {
-	switch d.Intent {
-	case router.IntentFact:
-		return "записала факт: " + d.Slots.Key + " " + d.Slots.Value
-	case router.IntentNote:
-		return "сохранила заметку: " + d.Slots.Text
-	case router.IntentReminder:
-		return "поставила напоминание: " + d.Slots.Text
-	default:
-		return string(d.Intent) + ": " + d.Slots.Text
+	// The persona checks, on the live path (personaguard.go). A reply that
+	// leaks reasoning or calls him "вы" is worse than a flat one.
+	if _, ok := guardSpoken("reply", out); !ok {
+		return r.stub.Reply(d)
 	}
+	return out
 }

cmd/mavend/replier_llm_test.go

@@ -5,28 +5,22 @@ import (
 	"testing"
 
 	"github.com/kami/maven/internal/llm"
-	"github.com/kami/maven/internal/phraser"
 	"github.com/kami/maven/internal/router"
 	"github.com/kami/maven/internal/voice"
 )
 
-type mockCompleter struct {
+// The phrasing itself is tested in internal/phraser. What is left here is the
+// only thing the daemon adds: the stub floor, on the three ways a reply can
+// fail to arrive.
+type stubCompleter struct {
 	out string
 	err error
 }
 
-func (m mockCompleter) Complete(_ context.Context, _ llm.Req) (string, error) { return m.out, m.err }
+func (s stubCompleter) Complete(_ context.Context, _ llm.Req) (string, error) { return s.out, s.err }
 
-func TestLLMReplierReturnsLLMReply(t *testing.T) {
-	r := newLLMReplier(mockCompleter{out: `{"response":"записала, кофе закончился","mood":"neutral"}`}, nil)
-	got := r.Reply(router.Decision{Intent: router.IntentNote, Slots: router.Slots{Text: "кофе закончился"}})
-	if got != "записала, кофе закончился" {
-		t.Errorf("got %q, want %q", got, "записала, кофе закончился")
-	}
-}
-
-func TestLLMReplierFallsBackToPlainText(t *testing.T) {
-	r := newLLMReplier(mockCompleter{out: "записала, кофе закончился"}, nil)
+func TestLLMReplierPassesTheModelReplyThrough(t *testing.T) {
+	r := newLLMReplier(stubCompleter{out: `{"response":"записала, кофе закончился","mood":"neutral"}`}, nil)
 	got := r.Reply(router.Decision{Intent: router.IntentNote, Slots: router.Slots{Text: "кофе закончился"}})
 	if got != "записала, кофе закончился" {
 		t.Errorf("got %q, want %q", got, "записала, кофе закончился")
@@ -34,54 +28,30 @@ func TestLLMReplierFallsBackToPlainText(t *testing.T) {
 }
 
 func TestLLMReplierFallsBackToStubOnError(t *testing.T) {
-	r := newLLMReplier(mockCompleter{err: errTestLLMDown}, nil)
-	noteDec := router.Decision{Intent: router.IntentNote}
-	got := r.Reply(noteDec)
-	want := voice.NewStubReplier().Reply(noteDec)
-	if got != want {
-		t.Errorf("on llm error: got %q, want stub %q", got, want)
-	}
+	r := newLLMReplier(stubCompleter{err: errReplierTest}, nil)
+	assertStub(t, r, router.Decision{Intent: router.IntentNote}, "llm error")
 }
 
 func TestLLMReplierFallsBackToStubOnEmpty(t *testing.T) {
-	r := newLLMReplier(mockCompleter{out: ""}, nil)
-	noteDec := router.Decision{Intent: router.IntentNote}
-	got := r.Reply(noteDec)
-	want := voice.NewStubReplier().Reply(noteDec)
-	if got != want {
-		t.Errorf("on empty llm: got %q, want stub %q", got, want)
-	}
+	r := newLLMReplier(stubCompleter{out: ""}, nil)
+	assertStub(t, r, router.Decision{Intent: router.IntentNote}, "empty llm")
 }
 
 func TestLLMReplierClarifyUsesStub(t *testing.T) {
-	r := newLLMReplier(mockCompleter{out: "я всё поняла"}, nil)
-	clarifyDec := router.Decision{Clarify: true}
-	got := r.Reply(clarifyDec)
-	want := voice.NewStubReplier().Reply(clarifyDec)
+	r := newLLMReplier(stubCompleter{out: "я всё поняла"}, nil)
+	assertStub(t, r, router.Decision{Clarify: true}, "clarify")
+}
+
+func assertStub(t *testing.T, r *llmReplier, d router.Decision, what string) {
+	t.Helper()
+	got, want := r.Reply(d), voice.NewStubReplier().Reply(d)
 	if got != want {
-		t.Errorf("on clarify: got %q, want stub %q", got, want)
+		t.Errorf("on %s: got %q, want stub %q", what, got, want)
 	}
 }
 
-var errTestLLMDown = errTest("llm down")
+var errReplierTest = errTest("llm down")
 
 type errTest string
 
 func (e errTest) Error() string { return string(e) }
-
-// grammarRecorder captures the request so the grammar can be asserted on.
-type grammarRecorder struct{ req llm.Req }
-
-func (g *grammarRecorder) Complete(_ context.Context, r llm.Req) (string, error) {
-	g.req = r
-	return `{"response":"записала","mood":"neutral"}`, nil
-}
-
-func TestLLMReplierCarriesTheResponseGrammar(t *testing.T) {
-	rec := &grammarRecorder{}
-	r := newLLMReplier(rec, nil)
-	r.Reply(router.Decision{Intent: router.IntentNote, Slots: router.Slots{Text: "кофе закончился"}})
-	if rec.req.Grammar != phraser.ResponseGrammar {
-		t.Errorf("grammar = %q, want phraser.ResponseGrammar", rec.req.Grammar)
-	}
-}

cmd/mavend/tick.go

@@ -176,6 +176,9 @@ func (t *tickLoop) tick(ctx context.Context, now time.Time) {
 			t.queueNudge(ctx, cand, state, now)
 		} else {
 			pn, err := t.phraser.PhraseNudge(ctx, *cand)
+			if err == nil {
+				pn = guardNudge(pn, *cand)
+			}
 			if err != nil {
 				log.Printf("tick: phrase nudge %s: %v", cand.Rule.Name, err)
 			} else {

cmd/mavend/voice.go

@@ -76,6 +76,10 @@ type reactiveHandler struct {
 	tts      tts.Synthesizer
 	router   *router.Router
 	embedder router.Embedder // reused for note write/query (same model as the classifier)
+	// boundary — the embedded seed sets behind the personal boundary
+	// (personalboundary.go). Zero value is usable and loads on first query;
+	// with no embedder it never loads and the boundary uses personalMarkers.
+	boundary personalBoundary
 	// api — the CoreAPI the handler reads and writes through. Wired with the
 	// bare store adapter and UPGRADED by main once the daemonAPI exists; see
 	// upgradeAPI.

cmd/mavend/voicewire.go

@@ -48,7 +48,11 @@ type voiceWiring struct {
 	// mcp — the MCP client, nil unless the `mcp` block configures an enabled
 	// server (Vikunja #251). Its tools land in the same allowlist as every
 	// other act, so nothing else here has to know about it.
-	mcp *mcpWiring
+	// pair — the workstation model with the resident one as the floor, nil
+	// unless a `workstation` block names an address. Held here only so the
+	// prober is stopped on shutdown; callers were handed it at build time.
+	pair *llm.Pair
+	mcp  *mcpWiring
 	// home — the Home Assistant client, nil unless the `smarthome` block is
 	// enabled (Vikunja #256). Its devices land in the same allowlist as every
 	// other act, so nothing else here has to know about it.
@@ -76,6 +80,9 @@ func (w *voiceWiring) close() {
 	if w.ttsClient != nil {
 		_ = w.ttsClient.Close()
 	}
+	if w.pair != nil {
+		w.pair.Stop()
+	}
 	w.mcp.close()
 }
 
@@ -139,6 +146,7 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 		emb = router.NewHashEmbedder(1024)
 	}
 	w.embedder = emb
+	repairFactVectors(dataStore, emb)
 	checkStoredEmbedder(dataStore, emb)
 
 	// ----- tool executor (the enabled act allowlist, store-backed) -----
@@ -188,6 +196,18 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 		// the new llama-server when the resident model is swapped (Vikunja #250).
 		llmClient = llmClientFor(lp, 60*time.Second)
 	}
+	// The workstation model sits above that one when it is configured and its
+	// card is free. hot is what the router and the replier complete through:
+	// either the pair, or the resident client alone, or nothing at all.
+	hot, pair := modelSeam(cfg, llmClient)
+	w.pair = pair
+	// The phraser gets the same pair, which is what carries the workstation model
+	// into the paths that do not go through `hot`: world questions (the naming
+	// half), and the digestion worker's nudge and reminder phrasing (the silent
+	// half). Wiring, so it happens once and before the voice server listens.
+	if lp, ok := phr.(*phraser.LLMPhraser); ok && pair != nil {
+		lp.UseRemote(pair)
+	}
 	// ----- router (the cascade; floor examples seed the classifier) -----
 	// The act matcher's allowlist is exactly the enabled tool names — the
 	// router only matches acts the executor can run (one source of truth).
@@ -199,7 +219,7 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 	// against the classifier's 50.0%, at about 1s a turn instead of 30ms (see
 	// config.VoiceConfig.LLMRouter). The classifier always stays wired as the
 	// fallback, so a model error never breaks a turn.
-	rtr := buildRouter(emb, matcher, threshold, pickLLMRouter(cfg.Voice.UseLLMRouter(), llmClient))
+	rtr := buildRouter(emb, matcher, threshold, pickLLMRouter(cfg.Voice.UseLLMRouter(), hot))
 
 	// ----- sessions registry (shared with voicesink) -----
 	sessions := voice.NewSessions()
@@ -233,8 +253,8 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 
 	// ----- replier (LLM-backed when the engine is on, Stub floor otherwise) -----
 	replier := voice.Replier(voice.NewStubReplier())
-	if llmClient != nil {
-		replier = newLLMReplier(llmClient, contextBlockFn(cfg, time.Now))
+	if hot != nil {
+		replier = newLLMReplier(hot, contextBlockFn(cfg, time.Now))
 	}
 
 	// ----- the handler (the reactive path; closes over stt / tts / router / coreAPI / memory) -----
@@ -291,7 +311,42 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 // pickLLMRouter returns the LLM router when the operator asked for it and there
 // is a llama-server to talk to, and nil otherwise. nil is safe: the cascade then
 // routes with the classifier, so an unusable setting costs accuracy, not turns.
-func pickLLMRouter(enabled bool, c *llm.Client) *router.LLMRouter {
+// modelSeam builds the completion seam the hot paths use: routing and replies.
+//
+// With no `workstation` block it is the resident client and nothing probes
+// anything, which is today's deploy exactly. With one, it is an llm.Pair that
+// prefers the workstation and falls back to the resident model silently — the
+// silent half of the degradation rule (docs/offload.md), because the big model
+// is only better here and the 1.7B is today's shipping quality. He is never
+// told which of the two phrased his reply.
+//
+// A nil resident client means the phraser is not an LLM phraser. There is then
+// no floor, and a Pair with no floor is a configuration mistake rather than a
+// degraded mode, so the seam is nil and the cascade routes with the classifier.
+func modelSeam(cfg *config.Config, resident *llm.Client) (router.Completer, *llm.Pair) {
+	if resident == nil {
+		if cfg.Workstation != nil {
+			log.Printf("voice: a workstation is configured but there is no resident model to floor it with — ignoring the block")
+		}
+		return nil, nil
+	}
+	if cfg.Workstation == nil {
+		return resident, nil
+	}
+	ws := cfg.Workstation
+	pair := llm.NewPair(
+		llm.New(ws.URL, time.Duration(ws.Timeout)),
+		resident,
+		ws.Health,
+		time.Duration(ws.Probe),
+	)
+	pair.Start(context.Background())
+	log.Printf("voice: workstation model at %s, probed every %s, resident model as the floor",
+		ws.URL, time.Duration(ws.Probe))
+	return pair, pair
+}
+
+func pickLLMRouter(enabled bool, c router.Completer) *router.LLMRouter {
 	if !enabled {
 		return nil
 	}
@@ -324,6 +379,7 @@ func buildRouter(emb router.Embedder, acts router.ActMatcher, threshold float64,
 	// question and must keep reaching replySystem, while "что у меня сегодня"
 	// is an agenda question and must not.
 	grammars = append(grammars, router.AgendaQueryGrammars()...)
+	grammars = append(grammars, router.ListGrammars()...)
 	grammars = append(grammars, router.ReminderGrammar())
 	return router.New(router.Config{
 		Grammars:   grammars,
@@ -419,6 +475,36 @@ func seedTools(api ipc.CoreAPI, tools []config.ToolConfig) {
 	log.Printf("voice: seeded %d act tools from config", n)
 }
 
+// repairFactVectors brings stored fact vectors in line with the facts they name
+// (#493), once per box, before the embedder marker is even looked at.
+//
+// Automatic and not a flag, unlike -reembed: only voice-tapped facts are in
+// this index, so the work is tens of embeddings rather than the thousands of
+// notes that made the backfill a deliberate act. And the box that needs it is
+// broken in a way nobody can see — recall answers with the wrong text and
+// nothing logs an error — so waiting for an operator to know to run it is how
+// the defect survived four restarts in the first place.
+func repairFactVectors(dataStore *store.Store, emb router.Embedder) {
+	if dataStore == nil {
+		return
+	}
+	res, err := dataStore.RepairFactVectors(context.Background(),
+		// EmbedPassage, the stored side, same as every other writer of these
+		// vectors.
+		func(ctx context.Context, text string) ([]float32, error) {
+			return router.EmbedPassage(ctx, emb, text)
+		})
+	if err != nil {
+		log.Printf("voice: fact vector repair failed, no marker written and nothing half-done — retried next start: %v", err)
+		return
+	}
+	if res.Skipped || res.Rewritten+res.Dropped == 0 {
+		return
+	}
+	log.Printf("voice: fact vector repair — %d re-embedded from the fact they name, %d dropped as voided or superseded, %d already right, took %s (#493)",
+		res.Rewritten, res.Dropped, res.Kept, res.Took.Round(time.Millisecond))
+}
+
 // reembedOnStart is the -reembed flag (set in run()). Opt-in on purpose: see
 // runReembed.
 var reembedOnStart bool

cmd/mavend/worldmodel.go

@@ -0,0 +1,60 @@
+package main
+
+import (
+	"context"
+	"errors"
+	"log"
+
+	"github.com/kami/maven/internal/phraser"
+)
+
+// worldPhraser — the naming half of the degradation rule (docs/offload.md), as
+// the query sources see it. Only *phraser.LLMPhraser implements it, so the
+// Stub and every test double stay exactly as they are.
+type worldPhraser interface {
+	PhraseWorld(ctx context.Context, utterance string, sources []string) (string, error)
+}
+
+// worldGap — what he hears when the question is about the world, the workstation
+// model is the one configured to answer it, and that machine is not answering.
+//
+// It says the true thing. The resident 1.7B is not a worse answer here, it is an
+// invented one: "Война и мир" came back with Левитан as its author, and a
+// question about his meeting came back as a swimming competition in Nottingham.
+// Naming the gap is the rule CLAUDE.md already applies to a sibling service
+// being down.
+const worldGap = "сейчас не могу ответить — большая модель недоступна, а придумывать не хочу."
+
+// phraseWorld asks the world model, or reports the gap.
+//
+// The three outcomes come straight from LLMPhraser.PhraseWorld: no workstation
+// configured means the resident model answers as it always has, a workstation
+// that is up answers, and a workstation that is down returns
+// phraser.ErrNoWorldModel. A phraser that has no world seam at all — the Stub,
+// and the doubles in the tests — is the first of those three.
+func (h *reactiveHandler) phraseWorld(ctx context.Context, utterance string, sources []string) (string, error) {
+	if h.phraser == nil {
+		return "", phraser.ErrNoWorldModel
+	}
+	if w, ok := h.phraser.(worldPhraser); ok {
+		return w.PhraseWorld(ctx, utterance, sources)
+	}
+	return h.phraser.PhraseQuery(ctx, utterance, sources)
+}
+
+// phraseSource asks the world model to answer from a passage someone already
+// fetched — a live search result, a ZIM article, a page he named. It returns ""
+// rather than the gap phrase, because these callers hold something better than a
+// gap: the passage itself, which their own floor reads back to him. Nothing is
+// invented either way, and a real quote beats "не могу сейчас".
+func (h *reactiveHandler) phraseSource(ctx context.Context, name, utterance string, sources []string) string {
+	reply, err := h.phraseWorld(ctx, utterance, sources)
+	switch {
+	case errors.Is(err, phraser.ErrNoWorldModel):
+		log.Printf("voice: %s: no world model, reading the source back instead", name)
+		return ""
+	case err != nil:
+		log.Printf("voice: %s: phrase: %v", name, err)
+	}
+	return reply
+}

cmd/mavend/worldmodel_test.go

@@ -0,0 +1,85 @@
+package main
+
+import (
+	"context"
+	"strings"
+	"testing"
+
+	"github.com/kami/maven/internal/phraser"
+	"github.com/kami/maven/internal/router"
+)
+
+// gapPhraser — a phraser whose world model is configured and asleep, which is
+// the state the naming half exists for.
+type gapPhraser struct {
+	*phraser.Stub
+	worldCalls int
+}
+
+func (g *gapPhraser) PhraseWorld(context.Context, string, []string) (string, error) {
+	g.worldCalls++
+	return "", phraser.ErrNoWorldModel
+}
+
+func worldTurn(utterance string) *queryTurn {
+	return &queryTurn{dec: router.Decision{Intent: router.IntentQuery, Utterance: utterance}}
+}
+
+// A world question with the workstation asleep says so. The resident model is
+// not asked, because what it produces here is an invention with no signal that
+// it is one.
+func TestQueryGeneralNamesTheGap(t *testing.T) {
+	g := &gapPhraser{Stub: phraser.NewStub()}
+	h := &reactiveHandler{phraser: g}
+	reply, ok := h.queryGeneral(context.Background(), worldTurn("почему небо голубое"))
+	if !ok {
+		t.Fatal("queryGeneral passed on the last source in the chain")
+	}
+	if reply != worldGap {
+		t.Fatalf("reply = %q, want the named gap", reply)
+	}
+	if g.worldCalls != 1 {
+		t.Fatalf("PhraseWorld called %d times, want 1", g.worldCalls)
+	}
+}
+
+// A phraser with no world seam at all — the Stub, and every box with no
+// `workstation` block — answers exactly as it did before this seam existed.
+func TestQueryGeneralWithoutAWorldModelIsUnchanged(t *testing.T) {
+	h := &reactiveHandler{phraser: phraser.NewStub()}
+	reply, ok := h.queryGeneral(context.Background(), worldTurn("почему небо голубое"))
+	if !ok {
+		t.Fatal("queryGeneral passed on the last source in the chain")
+	}
+	if reply != "не знаю." {
+		t.Fatalf("reply = %q, want the Stub's answer", reply)
+	}
+}
+
+// The gap is spoken aloud by a Russian voice, so it is Russian, feminine and
+// informal. "не хочу" and "не могу" are her own verbs; there is no "вы" and no
+// English in it.
+func TestWorldGapIsInPersona(t *testing.T) {
+	for _, bad := range []string{"вы", "ваш", "рад ", "дорогой", "милый"} {
+		if strings.Contains(worldGap, bad) {
+			t.Errorf("the gap phrase contains %q: %s", bad, worldGap)
+		}
+	}
+	if strings.ContainsAny(worldGap, "abcdefghijklmnopqrstuvwxyz") {
+		t.Errorf("the gap phrase has Latin letters in it: %s", worldGap)
+	}
+}
+
+// The sources that hold a passage read it back rather than name a gap. He gets a
+// real quote instead of "не могу сейчас", and nothing is invented either way.
+func TestASourceWithAPassageReadsItBackInsteadOfNamingTheGap(t *testing.T) {
+	g := &gapPhraser{Stub: phraser.NewStub()}
+	h := &reactiveHandler{phraser: g}
+	if got := h.phraseSource(context.Background(), "search", "почему небо голубое",
+		[]string{"Рэлеевское рассеяние."}); got != "" {
+		t.Fatalf("phraseSource = %q, want \"\" so the caller's own floor reads the passage back", got)
+	}
+	if g.worldCalls != 1 {
+		t.Fatalf("PhraseWorld called %d times, want 1", g.worldCalls)
+	}
+}

cmd/mavgpud/runner.go

@@ -24,7 +24,12 @@ type runner struct {
 	mu    sync.Mutex
 	cmd   *exec.Cmd
 	ready bool
-	http  *http.Client
+	// yielding — stop() has sent the signal and the exit that follows is ours.
+	// llama-server aborts on SIGTERM (its static teardown throws, upstream
+	// ggml-org/llama.cpp), so a routine yield and a real crash produce the same
+	// "signal: aborted" and used to log identically (Vikunja #491).
+	yielding bool
+	http     *http.Client
 }
 
 func newRunner(bin string, args []string, readyURL string) *runner {
@@ -70,13 +75,18 @@ func (r *runner) start() error {
 	if err := cmd.Start(); err != nil {
 		return err
 	}
-	r.cmd, r.ready = cmd, false
+	r.cmd, r.ready, r.yielding = cmd, false, false
 	log.Printf("mavgpud: started llama-server pid=%d", cmd.Process.Pid)
 	go func() {
 		err := cmd.Wait()
 		r.mu.Lock()
-		r.cmd, r.ready = nil, false
+		yielded := r.yielding
+		r.cmd, r.ready, r.yielding = nil, false, false
 		r.mu.Unlock()
+		if yielded {
+			log.Printf("mavgpud: llama-server stopped, card yielded (%v)", err)
+			return
+		}
 		log.Printf("mavgpud: llama-server exited: %v", err)
 	}()
 	return nil
@@ -90,6 +100,10 @@ func (r *runner) stop(grace time.Duration) {
 	r.mu.Lock()
 	cmd := r.cmd
 	r.ready = false
+	if cmd != nil && cmd.Process != nil {
+		// The exit that follows is ours, not a crash.
+		r.yielding = true
+	}
 	r.mu.Unlock()
 	if cmd == nil || cmd.Process == nil {
 		return

cmd/mavgpud/runner_test.go

@@ -0,0 +1,59 @@
+package main
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+	"time"
+)
+
+// fakeServer writes an executable standing in for llama-server: it ignores
+// SIGTERM the way the real one effectively does — by dying messily rather than
+// cleanly — and reports a non-zero status.
+func fakeServer(t *testing.T, body string) string {
+	t.Helper()
+	path := filepath.Join(t.TempDir(), "fake-llama-server")
+	if err := os.WriteFile(path, []byte("#!/bin/sh\n"+body+"\n"), 0o755); err != nil {
+		t.Fatal(err)
+	}
+	return path
+}
+
+// A deliberate stop is a yield, and the log has to say so.
+//
+// llama-server aborts inside its own static teardown on SIGTERM, so the exit
+// status of a routine yield is identical to that of a real crash. Reading the
+// mavgpud log, the two were indistinguishable (Vikunja #491).
+func TestStopMarksTheExitAsAYield(t *testing.T) {
+	r := newRunner(fakeServer(t, "while : ; do sleep 1 ; done"), nil, "")
+	if err := r.start(); err != nil {
+		t.Fatalf("start: %v", err)
+	}
+	r.mu.Lock()
+	if r.yielding {
+		t.Error("a freshly started server is already marked as yielding")
+	}
+	r.mu.Unlock()
+
+	r.stop(2 * time.Second)
+	deadline := time.Now().Add(2 * time.Second)
+	for time.Now().Before(deadline) {
+		if !r.running() {
+			return
+		}
+		time.Sleep(10 * time.Millisecond)
+	}
+	t.Fatal("the child outlived stop")
+}
+
+// Stopping when nothing is running must not arm the flag for the next child.
+// The next exit after that would be a real crash logged as a yield.
+func TestStopWithNoChildDoesNotArmTheFlag(t *testing.T) {
+	r := newRunner("/nonexistent", nil, "")
+	r.stop(10 * time.Millisecond)
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	if r.yielding {
+		t.Error("stop armed the yield flag with no child running")
+	}
+}

cmd/mavweb/main.go

@@ -27,6 +27,7 @@ import (
 	"github.com/kami/maven/internal/ipc"
 	"github.com/kami/maven/internal/pattern"
 	"github.com/kami/maven/internal/tasks"
+	"github.com/kami/maven/internal/tool"
 	"github.com/kami/maven/internal/voice"
 	"github.com/kami/maven/internal/webauthn"
 )
@@ -746,6 +747,8 @@ func handleDash(w http.ResponseWriter, r *http.Request, core ipc.CoreAPI) {
 var toolsTmpl = template.Must(template.New("tools").Funcs(func() template.FuncMap {
 	m := shellFuncs()
 	m["join"] = strings.Join
+	m["capability"] = func(t ipc.Tool) string { return tool.CapabilityOf(t).String() }
+	m["risk"] = func(t ipc.Tool) string { return string(tool.RiskOf(t)) }
 	return m
 }()).Parse(shellTopHTML + toolsHTML + shellBottomHTML))
 
@@ -756,9 +759,9 @@ const toolsHTML = `{{template "shellTop" "tools"}}
 <section class=card>
 <h2 class=card-title>proposed <span class=badge>{{len .Proposed}}</span></h2>
 {{if .Proposed}}<p class=hint>maven drafted these from acts she couldn't run. Fill the command (argv, space-separated) and enable. A row in an <code>mcp:</code> scope came from an MCP server and already knows what it calls — check the command, then enable.</p>
-<div class=scroll><table><tr><th>name</th><th>scope</th><th>from utterance</th><th>enable as</th></tr>
+<div class=scroll><table><tr><th>name</th><th>capability</th><th>scope</th><th>from utterance</th><th>enable as</th></tr>
 {{range .Proposed}}<tr>
-<td><code>{{.Name}}</code></td><td><span class=badge>{{.Scope}}</span></td><td>{{.Utterance}}</td>
+<td><code>{{.Name}}</code></td><td><code>{{capability .}}</code></td><td><span class=badge>{{.Scope}}</span></td><td>{{.Utterance}}</td>
 <td><form method=post action=/tools>
 <input type=hidden name=name value="{{.Name}}">
 <input type=hidden name=scope value="{{.Scope}}">
@@ -779,14 +782,16 @@ const toolsHTML = `{{template "shellTop" "tools"}}
 </section>
 <section class=card>
 <h2 class=card-title>enabled <span class=badge>{{len .Enabled}}</span></h2>
-{{if .Enabled}}<div class=scroll><table><tr><th>name</th><th>scope</th><th>command</th><th></th><th></th></tr>
-{{range .Enabled}}<tr><td><code>{{.Name}}</code></td><td><span class=badge>{{.Scope}}</span></td><td><code>{{join .Cmd " "}}</code></td>
-<td>{{if .Destructive}}<span class=red>destructive</span>{{end}}</td>
+{{if .Enabled}}<p class=hint>grouped by capability domain. The dotted id is <code>scope.domain.action</code> — the same shape Hexis speaks — and it is derived from the row, so it always describes what the command actually does.</p>
+{{range .Groups}}<h3 class=card-title><code>{{.Prefix}}</code> <span class=badge>{{len .Tools}}</span></h3>
+<div class=scroll><table><tr><th>capability</th><th>name</th><th>command</th><th>risk</th><th></th></tr>
+{{range .Tools}}<tr><td><code>{{capability .}}</code></td><td><code>{{.Name}}</code></td><td><code>{{join .Cmd " "}}</code></td>
+<td>{{$r := risk .}}{{if eq $r "irreversible"}}<span class=red>irreversible</span>{{else if eq $r "destructive"}}<span class=red>destructive</span>{{else}}<span class=badge>safe</span>{{end}}</td>
 <td><form method=post action=/tools class=inline-form>
 <input type=hidden name=name value="{{.Name}}">
 <input type=hidden name=scope value="{{.Scope}}">
 <input type=hidden name=action value=disable>
-<button class=btn>disable</button></form></td></tr>{{end}}</table></div>
+<button class=btn>disable</button></form></td></tr>{{end}}</table></div>{{end}}
 {{else}}<div class=empty>
   <svg class=icon width="20" height="20"><use href="/ethos-icons.svg#i-settings"/></svg>
   <div>no tools enabled</div>
@@ -1459,12 +1464,16 @@ func handleTools(w http.ResponseWriter, r *http.Request, core ipc.CoreAPI, sessi
 		servers = nil
 	}
 	w.Header().Set("Content-Type", "text/html; charset=utf-8")
+	// Enabled rows are shown grouped by capability domain (Vikunja #452). A
+	// flat list stops answering "what can she do to the house" somewhere
+	// around fifteen rows, and that is the question this page exists for.
 	if err := toolsTmpl.Execute(w, struct {
 		Msg      string
 		Proposed []ipc.Tool
 		Enabled  []ipc.Tool
+		Groups   []tool.CapabilityGroup
 		MCP      []ipc.MCPServerStatus
-	}{msg, proposed, enabled, servers}); err != nil {
+	}{msg, proposed, enabled, tool.GroupByDomain(enabled), servers}); err != nil {
 		log.Printf("tools render: %v", err)
 	}
 }

deploy/mavend.json

@@ -20,6 +20,7 @@
     "bin_path": "llama-server",
     "n_gpu_layers": 99,
     "n_ctx": 4096,
+    "cache_ram_mib": 512,
     "timeout": "60s",
     "llm_nudges": false
   },
@@ -39,6 +40,23 @@
     "proxy": "socks5://192.168.240.1:10808"
   },
 
+  "//workstation": [
+    "The big model on the desk PC (workpc, 7900 GRE 16GB), fronted by",
+    "mavgpud on port 8080. It runs gemma-4-12b and it is preferred over the",
+    "resident Qwen3-1.7B for routing and replies whenever the card is free.",
+    "The machine is never assumed up: it sleeps, and the card is often held by",
+    "a CPT run, in which case mavgpud answers 503 and Maven falls back to the",
+    "resident model without saying so. Deleting this block restores exactly",
+    "the behaviour homesrv had before it existed.",
+    "Addressed by LAN address, not container name: mavgpud runs on another",
+    "machine and there is no shared docker network to name it on."
+  ],
+  "workstation": {
+    "url": "http://192.168.1.105:8080",
+    "probe": "15s",
+    "timeout": "90s"
+  },
+
   "//search": [
     "The live web, searched after his own notes and before Kiwix. Only the",
     "query string leaves the box — never a note, a fact, the persona block or",

deploy/mavgpud.json

@@ -0,0 +1,32 @@
+{
+  "listen": ":8080",
+  "llama_addr": "127.0.0.1:10000",
+  "llama_bin": "llama-server",
+  "llama_args": [
+    "-m", "/mnt/D/AI/gemma4/gemma-4-12B-it-qat-UD-Q4_K_XL.gguf",
+    "-md", "/mnt/D/AI/gemma4/mtp-gemma-4-12B-it-BF16.gguf",
+    "-ngl", "99",
+    "-fa", "on",
+    "-np", "1",
+    "--host", "127.0.0.1",
+    "--port", "10000",
+    "--ctx-size", "32768",
+    "--threads", "6",
+    "--batch-size", "2048",
+    "--ubatch-size", "512",
+    "--jinja",
+    "--chat-template-kwargs", "{\"enable_thinking\":false}",
+    "--spec-type", "draft-mtp",
+    "--spec-draft-n-max", "2"
+  ],
+
+  "kfd_root": "/sys/class/kfd/kfd/proc",
+  "drm_device": "/sys/class/drm/card1/device",
+
+  "poll": "1s",
+  "idle_timeout": "15m",
+  "stop_grace": "20s",
+  "min_free_vram_bytes": 10737418240,
+  "evict_after_polls": 2,
+  "start_after_polls": 5
+}

deploy/mavgpud.service

@@ -0,0 +1,28 @@
+[Unit]
+# Runs on the workstation (bugmachine), not on homesrv. Install as a systemd
+# user unit and turn on lingering, so the card is supervised after a reboot
+# with nobody logged in:
+#
+#   scp mavgpud workpc:~/.local/bin/mavgpud
+#   scp deploy/mavgpud.json workpc:~/.config/mavgpud.json
+#   scp deploy/mavgpud.service workpc:~/.config/systemd/user/mavgpud.service
+#   ssh workpc 'systemctl --user daemon-reload && systemctl --user enable --now mavgpud'
+#   sudo loginctl enable-linger kami
+Description=Maven GPU supervisor (holds llama-server while the card is free)
+After=network.target
+
+[Service]
+ExecStart=%h/.local/bin/mavgpud -config %h/.config/mavgpud.json
+Restart=always
+RestartSec=5
+# The card must come back when the supervisor goes down. mavgpud stops
+# llama-server on SIGTERM, so give it longer than stop_grace to do that.
+KillSignal=SIGTERM
+TimeoutStopSec=60
+# llama-server aborts inside its own static teardown on SIGTERM, so every
+# routine yield used to write a multi-gigabyte core into systemd-coredump
+# (Vikunja #491). Yielding is meant to happen several times a day.
+LimitCORE=0
+
+[Install]
+WantedBy=default.target

docs/design.md

@@ -293,6 +293,57 @@ don't improvise.** Destructive ones still gate behind confirm.
 Misroute correction is append-only and grows the router's examples with use —
 same shape as `nudges.outcome` tuning cooldowns, no retrain.
 
+#### Risk tiers, not one boolean
+
+`Destructive` on a tool row is one bit set by whoever ticked the checkbox on
+`/tools`. It is a mechanism, and it never said which acts are destructive,
+whether a confirmed act stays confirmed, or what a new tool domain inherits.
+`internal/tool/risk.go` is the policy (Vikunja #449). The tier is DERIVED from
+the row, not stored, so it can be argued with in one place instead of being
+whatever the last person to enable the tool believed.
+
+| Tier | What it is | What it costs |
+|---|---|---|
+| `safe` | a read, or a change he can undo by saying the opposite | runs on first hearing |
+| `destructive` | it changes something real and undoing it takes work | one confirm turn, every time |
+| `irreversible` | the thing does not come back: a wipe, a format, a delete with no bin | voice may not authorise it at all |
+
+Three rules fall out, and they are the part that was missing:
+
+- **Which acts are destructive is not only the checkbox.** A house row always
+  is, because there is no read-only way to turn the heating off. A row whose
+  argv names one of the irreversible verbs always is, whatever the row says.
+- **A confirmed act never stays confirmed.** At any tier. A confirmation binds
+  one capability, one target and one argument list, and it dies with the parked
+  turn (90s). "The same act again" is a new act. A sticky confirm is a standing
+  grant and nothing on the voice path may hold one.
+- **A new domain inherits `destructive`, not `safe`.** A dispatch shape the
+  policy does not recognise gets the confirm turn. A domain argues its way down
+  to running freely; it never has to argue its way up to being gated.
+
+#### Capability ids
+
+A row is also read as a dotted capability id, `scope.domain.action` — the same
+shape Hexis has always spoken, which made the local surface the odd one out
+(Vikunja #452). `homelab.docker.restart`, `house.lock.unlock`,
+`mcp_vikunja.vikunja.delete_task`.
+
+Derived, not stored, for the reason the tier is: a derivation is one place to
+argue with. The name is still the primary key and nothing about lookup or
+execution changed — this is a way to READ the allowlist, not a second one.
+`/tools` groups the enabled rows by `scope.domain` and prints the id and the
+tier beside each, because a flat list stops answering "what can she do to the
+house" somewhere around fifteen rows.
+
+`MatchCapability` widens one way: `house` and `house.lock` both cover
+`house.lock.unlock`, and nothing lets a narrower id claim a wider pattern.
+
+The irreversible tier is refused rather than asked about, because a confirm
+turn would be theatre: everything that proposed the act — an STT guess, a
+router guess, a fuzzy allowlist match — is a guess, and a spoken "да" checks
+none of it. She names the gap and he runs it himself. The row stays enabled;
+refusing to run it from voice is not the same as taking it off the allowlist.
+
 ---
 
 ## Voice pipeline (STT / TTS)
@@ -630,6 +681,31 @@ add a new principle; it applied the existing one at smaller and smaller scope.
 
 ---
 
+## A list is the fourth shape
+
+Facts, notes and tasks were the three append-only shapes. `list_items` is the
+fourth (Vikunja #453): an item, a status, and a list tag.
+
+It is not a task. Milk is not work, nothing prioritises it, and the ranker must
+not start counting groceries as outstanding errands. It is not a fact either,
+because it claims nothing about the world. What it is, is a set that grows and
+shrinks.
+
+The property that makes the separate table worth it: no predicate reads a list.
+Nothing ranks it, nothing nudges about it, the digestion worker ignores it. So
+two people adding to the same list at once cost nothing — there is no order to
+disagree about and no lifecycle past crossed-off.
+
+The unique index is the tasks one, per list, and live rows only. Saying "молоко"
+twice before the shop is one line; saying it again next week, after the last one
+was crossed off, is a new line.
+
+Spoken, it is four turns: add, read back, cross one item off, cross the lot off.
+All four are matched deterministically in `internal/router/list.go` and all four
+run at stage 0, because an add and a read-back are cheap and should not depend on
+the resident model having a good turn. Crossing one item off claims the turn only
+when the list holds that item, which is what keeps "купил новый ноутбук" a note.
+
 ## Calendar
 
 Integration with **Radicale** (self-hosted CalDAV), not Nextcloud. Scope is

docs/evals/2026-08-02-workstation-gemma4-12b.md

@@ -0,0 +1,82 @@
+# gemma-4-12b on the workstation, against the resident Qwen3-1.7B
+
+Measured 2026-08-02 on the fixtures as they stand. Dated file: it is not edited
+after today, and a newer number is a new file.
+
+Vikunja #485's first assumption was that a 7-14B measurably beats Qwen3-1.7B on
+the 77-case RU routing fixture and the 27-case talk fixture. It does, on both,
+and it is also faster.
+
+## The setup
+
+`gemma-4-12B-it-qat-UD-Q4_K_XL` with the `mtp-gemma-4-12B-it-BF16` draft model,
+served by `llama-server` b10220 on bugmachine (AMD 7900 GRE, 16GB), fronted by
+`mavgpud` on `192.168.1.105:8080`. Thinking is off through
+`--chat-template-kwargs '{"enable_thinking":false}'`, speculative decoding is
+`--spec-type draft-mtp --spec-draft-n-max 2`, context 32768. The exact line is
+`deploy/mavgpud.json`.
+
+Every number below crossed the LAN from homesrv. Note the trap: homesrv's shell
+exports `HTTP_PROXY`, Go honours it, and the runs need
+`env -u HTTP_PROXY -u HTTPS_PROXY -u http_proxy -u https_proxy`.
+
+## Routing, 77-case RU fixture
+
+| | full | intent-only | p50 | p95 |
+|---|---|---|---|---|
+| classifier alone (02-08) | 68.8% | — | 16.6µs | — |
+| Qwen3-1.7B through the cascade (31-07, 02-08) | 72.7% | 77.9% | 0.80-1.04s | — |
+| **gemma-4-12b through the cascade** | **84.4%** | **93.5%** | **329ms** | 429ms |
+| gemma-4-12b alone, no cascade | 55.8% | 85.7% | 335ms | 436ms |
+
+The workstation buys 11.7 points of full accuracy over the resident model. It
+buys 15.6 points of intent-only, at a third of the latency. The router's p50 was
+never the model's fault, which the 02-08 contention finding already said. A 12B
+on a free 16GB card answers a routing turn in a third of a second.
+
+Two things the table hides.
+
+The alone-versus-cascade gap is slots, not intents. gemma reads the intent right
+85.7% of the time on its own. It loses full accuracy on seven fact keys
+(`вода` instead of `water`, `ужин` instead of `meal`) and on six reminder times
+with no time slot. Stage 0 and the daemon's own extractor repair
+both, which is why the cascade is 28 points higher. The lesson is that the
+cascade earns its keep even under a much better model, not that it is scaffolding
+to remove.
+
+`errors: 6` in the alone row are declines on single-token and ambiguous
+utterances, all of which the cascade caught. The remaining defects through the
+cascade are three `query→fact` confusions, one `chat→query`, and one false
+clarify.
+
+## Talk, 27-case conversational fixture
+
+| | pass | notes |
+|---|---|---|
+| Qwen3-1.7B (31-07) | 20/27 | 11-17/27 for the 0.8B before it |
+| **gemma-4-12b** | **25/27 (92.6%)** | chat 8/9, knowledge 9/9, query 8/9 |
+
+Knowledge is the interesting column: 9/9, in Russian, with real answers about
+Rayleigh scattering, SSD versus HDD and thunder delay. That is the case the
+1.7B cannot do at all and the reason the naming half of the degradation rule
+exists.
+
+Two failures, and one of them is the persona defect the CPT (#122) targets:
+`query-notes-do-not-answer` wrote `заплатил` where Maven needs the feminine
+form. The other is `chat-joke`, where the model told a joke without using any of
+the words the check looks for. Run-to-run variance is about one case: a second
+run scored 24/27 with `chat-followup-server` also off-topic.
+
+## Nudge phrasing, 15-case fixture
+
+15/15, every check, no errors. `mood`, `lang`, `length`, `feminine`,
+`hisgender`, `address`, `cringe` and `ontopic` all clean.
+
+## What this settles and what it does not
+
+Settled: the size question. A 12B on the workstation beats the resident model on
+every fixture we have, and it is faster. The offload argument holds.
+
+Not settled: how often the card is free. That is #485's second assumption and
+only the `mavgpud` log answers it, after a week of the owner's normal work. A
+model that is better whenever it is up is worth little if it is never up.

docs/evals/2026-08-03-llama-prompt-cache.md

@@ -0,0 +1,84 @@
+# Where the resident model's 7.9GB of RSS goes (2026-08-03, homesrv)
+
+Measured for Vikunja #499. The deployed llama-server held 7.9GB RSS for a 1.1GB
+model file. Half a gigabyte of it was in swap, on a box that also runs
+whisper.cpp, piper and the embedder.
+
+## Method
+
+`maven-mavend-1` was stopped for the measurement, with the owner's approval.
+Its own binary then ran on the host with the exact deployed command line. That
+binary is `/opt/maven/bin/llama-server`, version `1 (4c65955)`, a Vulkan build.
+
+```sh
+llama-server -m /mnt/hdd1/llms/qwen3/Qwen3-1.7B-UD-Q4_K_XL.gguf \
+  --host 127.0.0.1 --port 18099 -c 4096 -ngl 99 --no-webui
+```
+
+RSS was read from `/proc/<pid>/status` after load and after each of 8 distinct
+1521-token prompts. `smaps` of the deployed process was read first, from inside
+the container, since the host user cannot read another user's maps.
+
+## The cause: the prompt cache, not the weights and not the offload
+
+The startup log says it outright:
+
+```text
+srv load_model: prompt cache is enabled, size limit: 8192 MiB
+srv llama_server: n_parallel is set to auto, using n_parallel = 4 and kv_unified = true
+```
+
+The server saves the full KV state of every idle slot it evicts. It keeps up to
+8GiB of those states in host RAM (llama.cpp PR 16391). One saved prompt of 1521
+tokens costs 166.377 MiB. That is 112 kiB per token, exactly Qwen3-1.7B's KV
+footprint (28 layers x 2 x 1024 dims x 2 bytes).
+
+RSS at rest, and per distinct prompt:
+
+| Prompts served | RSS, default | RSS, `--cache-ram 512` |
+|---|---|---|
+| 0 (just loaded) | 443 MB | 411 MB |
+| 1 | 445 MB | 411 MB |
+| 4 | 958 MB | 929 MB |
+| 8 | 1641 MB | 932 MB |
+
+Uncapped, RSS climbs about 170MB per distinct prompt and does not stop until
+the 8GiB limit. Capped at 512 MiB it plateaus at 932MB from the fourth prompt
+on, with the cache holding steady at `3 prompts, 499.132 MiB` and evicting.
+
+The 7.9GB on the running daemon was that climb, weeks of it. Its `smaps` showed
+one 6.03GB anonymous mapping at 5.32GB resident plus a 1.45GB mapping at 1.27GB
+resident, and only 30MB of file-backed RSS.
+
+## The task's leading guess was wrong
+
+`-ngl 99` on the Vega iGPU costs almost no process RSS. A freshly loaded server
+has 95MB of anonymous RSS in total. RADV allocates device memory through the
+kernel, outside the process, and the log sees 8202 MiB free on `Vulkan0`. The
+weights are mmapped and file-backed, so they are evictable and do not pin RSS. The logit buffer is not visible in the numbers above at all.
+
+## Decision
+
+`--cache-ram 512` is now the default, wired as `phraser.cache_ram_mib` and set
+in `deploy/mavend.json`. 512 MiB caps total RSS near 1GB, an eighth of what the
+box carried. It still holds three of the 1521-token probes above. Maven's real
+routing and phrasing prompts are much shorter, so it holds more of those than
+the table suggests. `-c 4096` is untouched, as #499
+required. A negative `cache_ram_mib` passes no flag, for a llama-server too old
+to know it.
+
+Not changed: `n_parallel = 4`. With `kv_unified = true` the four slots share one
+4096-token KV cache, so they do not multiply it.
+
+The other half of #499 was that none of these lines were reachable. mavend
+scraped llama-server's stderr for the listen line and discarded it, and never
+piped stdout at all. Both streams now go to mavend's log with a `llama:` prefix.
+The last 12 startup lines go into the error when the server dies before it
+listens.
+
+## Deployed
+
+The `mavenai:latest` image was rebuilt and `maven-mavend-1` recreated the same
+day. The daemon's own log now carries the child's startup, it reads
+`prompt cache is enabled, size limit: 512 MiB`, and the resident server sat at
+439MB RSS after load and 613MB after one served turn.

docs/evals/2026-08-03-personal-boundary.md

@@ -0,0 +1,46 @@
+# Personal boundary, seed scoring vs possession markers, 2026-08-03
+
+Vikunja #495. `что я говорил про бэкапы?` walked past the personal boundary into
+SearXNG and came back answered from a Habr article. The boundary matched
+possession words only, so a first-person speech verb was not a personal
+question.
+
+## What changed
+
+The boundary now scores the turn's query vector against two frozen seed sets.
+It claims the turn when the personal side is nearer than the world side. Seeds
+and code are in `cmd/mavend/personalboundary.go`. The possession markers stay as
+the offline floor for a handler with no embedder.
+
+A regex speech class was written first and dropped. Russian gives every verb a
+dozen surface forms, and the "как я говорил, ..." preamble list has no end. Each
+form the lexicon missed was one more question reaching the world.
+
+## Measurement
+
+Embedder: multilingual-e5-small int8, the one homesrv runs. Both sides are
+embedded on the query side. Cases are held out, none of them a seed. `make test`
+runs the offline part. The scored part is opt-in through `MAVEN_ONNX_LIB`, like
+`TestONNXRecall`.
+
+    19/19 held-out utterances correct   (TestONNXPersonalBoundary)
+
+    true positive margins   +0.014 to +0.089
+    nearest true negative   -0.005  ("кто такой гагарин")
+
+One case missed during the first pass and is not held out any more: `as i said,
+what is the population of india`, +0.008 to the personal side. It is a world seed
+now.
+
+The gate is the sign of the difference and nothing tighter. The margins are too
+thin for a threshold. The asymmetry favours claiming: a false claim costs one
+honest "не знаю", a false pass sends his life to an upstream engine.
+
+`make eval-recall` unchanged, 18/27 answered at gate 0.55. Recall does not touch
+this path.
+
+## Not verified
+
+The live probe on the deployed box. The daemon was not rebuilt in this session.
+The reply to `что я говорил про бэкапы?` with no matching note is still untested
+against a real SearXNG.

docs/evals/2026-08-03-recall-topic-veto.md

@@ -0,0 +1,65 @@
+# Recall topic veto, what it costs and what it buys, 2026-08-03
+
+Vikunja #496. The task asked for a cross-language fix. Skip the topic veto in
+`memory.RecallAllowed` when the question and the hit are in different scripts.
+An English question would then stop losing a Russian note.
+
+No such case exists. No fixture case puts the question and its wanted note in
+different scripts. The case the task named is not one either.
+
+    en-hard-024
+      query  "what fixed the screen problem"
+      note   "the flicker went away once i swapped the display cable"
+
+Both are English. It is a paraphrase failure, not a language failure. A script
+test would not have changed a single case, and neither would a bilingual stem
+map.
+
+## What the veto is worth today
+
+Measured with the real embedder, multilingual-e5-small int8, gate 0.55, margin
+0.008. The first row is the veto as it ships. The second is `RecallAllowed`
+forced to true.
+
+| | cases passing | answered | false recall | silenced by gate |
+|---|---|---|---|---|
+| veto on | 22/32 | 17/27 | 0/5 | 2 |
+| veto off | 22/32 | 18/27 | 1/5 | 1 |
+
+The pass count does not move. The veto trades one true recall for one false one.
+It costs `en-hard-024` and it buys `ru-silent-029`:
+
+    ru-silent-029
+      query  "во сколько отходит поезд"
+      note   "погулял вдоль реки"        0.835, margin 0.019
+
+The second case counted as silenced by the gate is `ru-home-026` at margin
+0.001, which the margin gate stops. The veto has nothing to do with it.
+
+## Why no lexical rule separates the two
+
+`en-hard-024` and `ru-silent-029` are in the same lexical class. Both questions
+share zero content words with their hit, and neither carries a first-person
+marker. The scores sit on top of each other, 0.826 against 0.835, and so do the
+margins, 0.023 against 0.019. Only one thing separates them. A screen problem
+and a swapped display cable are the same event. A train and a river walk are
+not. The embedder scores that difference at nine thousandths.
+
+So the signal is semantic and the gate is lexical. Any rule cheap enough to sit
+in `RecallAllowed` and strong enough to recover `en-hard-024` also re-admits
+`ru-silent-029`, which puts false recall back to 1/5.
+
+One near-miss rule was tried on paper and rejected: let the veto pass when the
+hit itself is first person. It works on these two, because the English note says
+"i swapped" and the Russian note says only "погулял". It is backwards as a
+principle. A first-person note is exactly the personal note the veto keeps away
+from a world question. The rule would weaken the veto where it was designed to
+bite. It survives here only because Russian drops the pronoun.
+
+## Decision
+
+Accept the loss. `en-hard-024` stays silenced and false recall stays 0/5.
+
+The way out is a reranker, not a longer word list. Recall@3 is 85.2% against
+recall@1 at 70.4%, so the right note is usually in the returned set and ranked
+wrong. That is where the remaining points are, and it is not this task.

docs/offload.md

@@ -1,6 +1,6 @@
 # Offloading model work to the workstation
 
-*Last verified: 2026-08-02 @ 5c05163. Living doc: correct it in place, do not append.*
+*Last verified: 2026-08-03 @ 12530c8. Living doc: correct it in place, do not append.*
 
 Owner's call, 2026-08-02. Vikunja #483 is the umbrella. Tasks #484 to #487 are the
 work, and this file holds the shape and the rules all four must obey.
@@ -47,6 +47,24 @@ service being down.
 
 Nothing in between. A turn never breaks on the workstation being asleep.
 
+Both halves are wired, 03-08-2026. `LLMPhraser.PhraseWorld`
+(`internal/phraser/world.go`) is the naming half and has three outcomes, not two:
+
+| State | What he hears |
+|---|---|
+| no `workstation` block | the resident model answers, exactly as before the seam existed |
+| configured, card free | the workstation answers |
+| configured, asleep or busy | the gap, `worldGap` in `cmd/mavend/worldmodel.go` |
+
+The first row is the one worth stating. Naming a gap requires a gap. On a box with
+no second model the 1.7B is the whole product. Refusing every world question there
+would remove a capability the owner has today.
+
+A source holding a passage is on the naming half too: a live search, a ZIM
+article, a page he named. None of them says "не могу сейчас". They read the
+passage back, which is what `phraseSource` returning `""` selects. A real quote
+beats a gap, and neither path invents.
+
 ## Admission control, not a scheduler
 
 There is no GPU arbiter. That is a service with its own failure modes, and nothing
@@ -71,6 +89,22 @@ It arbitrates nothing between callers. It reports whether it can take work and
 manages one process to back that answer. Maven never asks it to start anything
 and never learns that it did.
 
+Contention is decided by presence under `/sys/class/kfd/kfd/proc`, not by a VRAM
+threshold. A ROCm process registers there when it initialises HIP, before it
+allocates anything. So the supervisor sees a contender during that job's startup,
+and yields before the job loses the memory it asked for. A
+threshold reads the card too late. By the time free VRAM has dropped, the other
+job has already lost the allocation race. Free VRAM is still read, but only as a
+precondition for loading, never as the eviction signal. One blind spot is known.
+A job can take the card without registering on the KFD, as a Vulkan or a
+video-decode job would. `describe()` logs every contender's comm, and that log is
+how we find out whether the blind spot is real.
+
+`mavgpud` runs from a systemd unit on the workstation with
+`deploy/mavgpud.json` as its config, and `llama_args` is passed to llama-server
+untouched. The model, the context size, the layer count and the MTP flags are the
+owner's business and not this daemon's schema.
+
 ## What stays on homesrv, permanently
 
 The **embedder** (multilingual-e5-small, ONNX, CPU). It backs the classifier, which
@@ -88,17 +122,29 @@ it buys nothing. Four callers:
 
 ## Inventory: what runs a model on homesrv today
 
-The **resident model** is one llama-server with seven callers:
+The **resident model** is one llama-server with seven callers, and 03-08-2026 is
+the date each of them stopped or did not stop being resident-only:
 
-| Caller | What for |
-|---|---|
-| `cmd/mavend/voicewire.go` | routing |
-| `cmd/mavend/replier_llm.go` | replies |
-| `cmd/mavend/tick.go` | digestion worker: `PhraseNudge`, `PhraseReminder` |
-| `cmd/mavend/capture.go` | capture summarisation (unreachable, see #480) |
-| `cmd/mavend/mail.go` | mail extraction (off, no IMAP) |
-| `cmd/mavend/kiwixwire.go` | answering from a Kiwix, search or crawl passage |
-| `memoryeval.go`, `modelswap.go` | admin and evals |
+| Caller | What for | Offloaded |
+|---|---|---|
+| `cmd/mavend/voicewire.go` | routing | silently, through `hot` |
+| `cmd/mavend/replier_llm.go` | replies | silently, through `hot` |
+| `cmd/mavend/tick.go` | digestion worker: `PhraseNudge`, `PhraseReminder` | silently, inside the phraser |
+| `cmd/mavend/actions_query.go` | world questions, and any fetched passage | names the gap |
+| `cmd/mavend/capture.go` | capture summarisation (unreachable, see #480) | no, holds its own client |
+| `cmd/mavend/mail.go` | mail extraction (off, no IMAP) | no, holds its own client |
+| `memoryeval.go`, `modelswap.go` | admin and evals | no, and deliberately |
+
+The last three rows are resident-only on purpose. `memoryeval.go` and
+`modelswap.go` measure and swap the resident model, so sending their work
+elsewhere would measure the wrong thing. `capture.go` and `mail.go` are
+background jobs that hold a gated background client (`llmBackgroundClientFor`),
+and that priority has no equivalent on the remote yet. Both are also unreachable
+on this deploy, so wiring them would ship an untestable path.
+
+The `tick.go` row needs one caveat. `phraser.llm_nudges` is `false` in deploy, so
+nudges come from templates and the seam under them changes nothing until that
+flips. It is wired anyway: `PhraseReminder` is on the same transport and is on.
 
 Then the embedder above, **whisper.cpp** in `mavsttd`, and **piper** in `mavttsd`.
 `mavwaked` uses no model at all: an energy-threshold VAD over 30ms frames.
@@ -109,7 +155,14 @@ Then the embedder above, **whisper.cpp** in `mavsttd`, and **piper** in `mavttsd
    `internal/netaddr` landed in PR #92. A seam address now carries its own scheme,
    and a scheme-less one is still unix. A tcp seam requires a shared token, because
    the filesystem permission that authenticated the unix socket is gone.
-2. **The resident model** (#485). Biggest quality delta. A 16GB card runs a 7-14B,
+2. **The resident model** (#485, #490). Wired. A `workstation` block builds an
+   `llm.Pair` in `modelSeam` (`cmd/mavend/voicewire.go`), routing and replies
+   complete through it, and the phraser holds the same pair (`UseRemote`). Both
+   halves of the rule are live: see the table above for which caller gets which.
+   Measured, `docs/evals/2026-08-02-workstation-gemma4-12b.md`: gemma-4-12b
+   through the cascade scores 84.4% full accuracy at p50 329ms. The resident
+   model scores 72.7% at p50 0.80-1.04s. On the talk fixture it is 25/27
+   against 20/27. Biggest quality delta. A 16GB card runs a 7-14B,
    which fixes what the 1.7B gets wrong: world knowledge, and the persona the CPT
    targets. The degradation path is already written and measured, since the
    classifier scores 68.8% full accuracy at p50 16.6µs on its own.

docs/qa.md

@@ -498,12 +498,21 @@ rejects `https://api.openai.com`, and forget really deletes
 (`internal/store/memory.go:145` is a real `DELETE`, not a tombstone). Vision is
 19/19, speaker 22/22, media 16/16.
 
-**470 got worse.** Both poisoned facts show `voided` on `/history`, and the
-defect survives. Re-measured at 15:42, after four restarts: `почему небо синее?`
-still answers `какая последняя версия языка Go?` with no `search:` line. What
-comes back is the question he typed, not the value the fact held. So the poison
-is a vector in the memory index, and `revert` does not remove it. There is
-currently no documented way to repair a poisoned box.
+**470 got worse, then closed.** Both poisoned facts showed `voided` on
+`/history` and the defect survived. Re-measured at 15:42, after four restarts:
+`почему небо синее?` still answered `какая последняя версия языка Go?` with no
+`search:` line. What came back was the question he typed, not the value the fact
+held. So the poison was a vector in the memory index, and `revert` did not
+remove it.
+
+Repaired in two parts. 470 stopped the writes: a question is never a fact, and a
+void drops the key's vectors. 493 fixed what the index holds. A fact is indexed
+as the fact and not as the utterance, and a correction drops its superseded
+vector too.
+
+A poisoned box now repairs itself on the next start. `RepairFactVectors`
+re-embeds every fact vector from the fact it names, and deletes the voided and
+superseded ones. It runs once, guarded by a marker, and logs what it did.
 
 ---
 

internal/calendar/calendar.go

@@ -18,6 +18,7 @@ import (
 	"sort"
 	"strings"
 	"time"
+	"unicode"
 )
 
 // Fact sources. A calendar event reaches the store as a
@@ -153,14 +154,20 @@ func Overlapping(events []Event, from, to time.Time) []Event {
 	return out
 }
 
-// safeKey makes a summary safe to use inside a fact key (ASCII alphanumerics
-// and dashes). Non-Latin summaries collapse to their punctuation, which is why
-// the day prefix carries the identity and this only disambiguates within a day.
+// safeKey makes a summary safe to use inside a fact key: letters and digits in
+// any script, plus dashes, with space and underscore folded to a dash.
+//
+// It kept ASCII only until 04-08-2026, and dropped everything else. His
+// calendar is Russian, so "Встреча с Аней" and "Обед с мамой" both reduced to
+// "--" and produced the same key on the same day — the second event of the day
+// silently overwrote the first (Vikunja #443). Letting the letters through is
+// what makes the key identify the event. Migration #18 drops the keys written
+// under the old rule; they are re-derived on the next poll.
 func safeKey(s string) string {
 	var b strings.Builder
 	for _, r := range s {
 		switch {
-		case (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') || (r >= '0' && r <= '9') || r == '-':
+		case unicode.IsLetter(r) || unicode.IsDigit(r) || r == '-':
 			b.WriteRune(r)
 		case r == ' ' || r == '_':
 			b.WriteRune('-')

internal/calendar/calendar_test.go

@@ -139,6 +139,9 @@ func TestSafeKey(t *testing.T) {
 		{"Hello_World", "Hello-World"},
 		{"special@#$chars!!", "specialchars"},
 		{"ALL_CAPS_123", "ALL-CAPS-123"},
+		// His calendar is Russian. These reduced to "--" and "--" (Vikunja #443).
+		{"Встреча с Аней", "Встреча-с-Аней"},
+		{"Обед с мамой", "Обед-с-мамой"},
 	}
 	for _, tt := range tests {
 		if got := safeKey(tt.in); got != tt.want {
@@ -263,3 +266,19 @@ func TestSourceTrust(t *testing.T) {
 		t.Errorf("Sources() = %v", Sources())
 	}
 }
+
+// Two Russian events on one day must not share a key. They did: safeKey kept
+// ASCII only, so both summaries collapsed to their spaces and the second event
+// overwrote the first in the store (Vikunja #443).
+func TestFactKeyDistinguishesRussianEventsOnOneDay(t *testing.T) {
+	day := time.Date(2026, 8, 4, 0, 0, 0, 0, time.UTC)
+	a := Event{Summary: "Встреча с Аней", Start: day.Add(10 * time.Hour), End: day.Add(11 * time.Hour)}
+	b := Event{Summary: "Обед с мамой", Start: day.Add(13 * time.Hour), End: day.Add(14 * time.Hour)}
+	if FactKeyIn(a, time.UTC) == FactKeyIn(b, time.UTC) {
+		t.Fatalf("both events keyed as %q", FactKeyIn(a, time.UTC))
+	}
+	// The day prefix still has to survive, because the store range-scans on it.
+	if !strings.HasPrefix(FactKeyIn(a, time.UTC), KeyPrefixForDay(day)) {
+		t.Fatalf("key %q lost the day prefix %q", FactKeyIn(a, time.UTC), KeyPrefixForDay(day))
+	}
+}

internal/config/config.go

@@ -224,6 +224,11 @@ type Config struct {
 	// See SearchConfig.
 	Search *SearchConfig `json:"search,omitempty"`
 
+	// Workstation — the big model on the owner's desktop, preferred over the
+	// resident one when its GPU is free. nil / absent / url empty ⇒ homesrv
+	// behaves exactly as it does today. See WorkstationConfig.
+	Workstation *WorkstationConfig `json:"workstation,omitempty"`
+
 	// Praxis — the ecosystem attention-state service. When configured, maven
 	// calls the Praxis HTTP tools API for attention listing and item lifecycle.
 	// Maven never touches Praxis's database directly (ecosystem invariant: no
@@ -1105,6 +1110,44 @@ const (
 	DefaultKiwixSnippetRunes = 1500
 )
 
+// WorkstationConfig — the big model on the owner's desktop (workpc, a
+// 7900 GRE with 16GB), fronted by mavgpud.
+//
+// homesrv cannot grow a GPU, so the resident Qwen3-1.7B is the floor and this
+// is the preferred model above it (owner's call, 2026-08-02, docs/offload.md).
+// The workstation is never assumed up: its card is often held by a CPT run and
+// the machine sleeps. No block, or an empty URL, and homesrv behaves exactly as
+// it does today.
+//
+// Only the prompt crosses the LAN, and the workstation is not "the box". The
+// rules in CLAUDE.md about what may leave still apply.
+type WorkstationConfig struct {
+	// URL — where mavgpud listens, e.g. "http://192.168.1.105:8080". Empty ⇒
+	// the whole block is normalised to nil and nothing probes anything.
+	URL string `json:"url,omitempty"`
+
+	// Health — the admission endpoint. Empty ⇒ URL + "/health", which is what
+	// mavgpud serves. It answers 503 while the card is held, and that is the
+	// signal, so it must be the supervisor's endpoint and not llama-server's.
+	Health string `json:"health,omitempty"`
+
+	// Probe — how often admission is re-checked. 0 ⇒ DefaultWorkstationProbe.
+	// Nothing on the hot path waits for it: the answer is cached and read
+	// atomically, so this only sets how late Maven notices the card came back.
+	Probe Duration `json:"probe,omitempty"`
+
+	// Timeout — the per-request budget for a completion on the workstation.
+	// 0 ⇒ DefaultWorkstationTimeout. A big model on a LAN host is slower than
+	// the resident one, and a request that overruns falls back to the floor.
+	Timeout Duration `json:"timeout,omitempty"`
+}
+
+// Workstation defaults, applied in Normalise.
+const (
+	DefaultWorkstationProbe   = 15 * time.Second
+	DefaultWorkstationTimeout = 90 * time.Second
+)
+
 // SearchConfig — the self-hosted SearXNG instance she searches with.
 //
 // External search is allowed and off unless configured (CLAUDE.md). Configuring
@@ -1233,6 +1276,12 @@ type PhraserConfig struct {
 	NCtx       int      `json:"n_ctx,omitempty"`
 	Timeout    Duration `json:"timeout,omitempty"`
 
+	// CacheRAMMiB bounds llama-server's prompt cache. Omitted ⇒ 512 MiB, which
+	// is what keeps the resident model near 1 GB of RSS instead of the 7.9 GB
+	// measured on 2026-08-03. Set it to -1 to pass no flag at all and let the
+	// server apply its own 8 GiB default. See phraser.Config.CacheRAMMiB.
+	CacheRAMMiB int `json:"cache_ram_mib,omitempty"`
+
 	// LLMNudges — let the model word nudges again. Off by default: nudges are
 	// worded from hand-written Russian templates now (the model broke the
 	// persona and invented units). Chat, query and reminder phrasing always go
@@ -1500,6 +1549,24 @@ func (c *Config) applyDefaults() {
 		}
 	}
 
+	// No address, no preferred model. An unconfigured workstation is the
+	// default deploy and must be indistinguishable from today.
+	if c.Workstation != nil && strings.TrimSpace(c.Workstation.URL) == "" {
+		c.Workstation = nil
+	}
+	if c.Workstation != nil {
+		w := c.Workstation
+		if strings.TrimSpace(w.Health) == "" {
+			w.Health = strings.TrimRight(w.URL, "/") + "/health"
+		}
+		if w.Probe <= 0 {
+			w.Probe = Duration(DefaultWorkstationProbe)
+		}
+		if w.Timeout <= 0 {
+			w.Timeout = Duration(DefaultWorkstationTimeout)
+		}
+	}
+
 	if c.Voice != nil {
 		if c.Voice.RouterThreshold <= 0 {
 			c.Voice.RouterThreshold = DefaultRouterThreshold

internal/config/config_test.go

@@ -413,3 +413,56 @@ func TestNormaliseFillsKiwixDefaults(t *testing.T) {
 		t.Error("rewrite: false was not honoured")
 	}
 }
+
+// A workstation with no address is not a workstation. The unconfigured deploy
+// must be indistinguishable from today, so the block is dropped rather than
+// left to fail one probe at a time.
+func TestNormaliseDropsAddresslessWorkstation(t *testing.T) {
+	for _, tc := range []struct {
+		name string
+		in   *WorkstationConfig
+	}{
+		{"no url", &WorkstationConfig{Probe: Duration(time.Second)}},
+		{"blank url", &WorkstationConfig{URL: "   "}},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			c := &Config{Workstation: tc.in}
+			c.applyDefaults()
+			if c.Workstation != nil {
+				t.Errorf("kept an unusable workstation block: %+v", c.Workstation)
+			}
+		})
+	}
+}
+
+// The health endpoint defaults to the supervisor's, not llama-server's: mavgpud
+// answers 503 while the card is held, and that refusal is the whole signal.
+func TestNormaliseFillsWorkstationDefaults(t *testing.T) {
+	c := &Config{Workstation: &WorkstationConfig{URL: "http://192.168.1.105:8080/"}}
+	c.applyDefaults()
+	if c.Workstation == nil {
+		t.Fatal("dropped a usable workstation block")
+	}
+	if got, want := c.Workstation.Health, "http://192.168.1.105:8080/health"; got != want {
+		t.Errorf("Health = %q, want %q", got, want)
+	}
+	if time.Duration(c.Workstation.Probe) != DefaultWorkstationProbe {
+		t.Errorf("Probe = %s, want %s", time.Duration(c.Workstation.Probe), DefaultWorkstationProbe)
+	}
+	if time.Duration(c.Workstation.Timeout) != DefaultWorkstationTimeout {
+		t.Errorf("Timeout = %s, want %s", time.Duration(c.Workstation.Timeout), DefaultWorkstationTimeout)
+	}
+}
+
+// An explicit health URL is left alone: the supervisor may sit behind something
+// that does not put /health at the root.
+func TestNormaliseKeepsExplicitWorkstationHealth(t *testing.T) {
+	c := &Config{Workstation: &WorkstationConfig{
+		URL:    "http://192.168.1.105:8080",
+		Health: "http://192.168.1.105:9000/ready",
+	}}
+	c.applyDefaults()
+	if got, want := c.Workstation.Health, "http://192.168.1.105:9000/ready"; got != want {
+		t.Errorf("Health = %q, want %q", got, want)
+	}
+}

internal/llm/client.go

@@ -129,6 +129,11 @@ type Req struct {
 	RepeatPenalty float64
 	// Stop — sequences that end generation early (e.g. newline for a one-liner).
 	Stop []string
+	// Temperature — 0 (the zero value) is greedy decoding, and greedy is what
+	// every caller here wanted before this field existed. It is set only by the
+	// phraser, whose own transport has always sampled at 0.7: routing a phrasing
+	// call through this client must not quietly change how it decodes.
+	Temperature float64
 }
 
 type msg struct {
@@ -176,7 +181,7 @@ func (c *Client) Complete(ctx context.Context, r Req) (string, error) {
 		Messages:      []msg{{Role: "system", Content: r.System}, {Role: "user", Content: r.User}},
 		MaxTokens:     r.MaxTokens,
 		Grammar:       r.Grammar,
-		Temp:          0,
+		Temp:          r.Temperature,
 		RepeatPenalty: r.RepeatPenalty,
 		Stop:          r.Stop,
 	})

internal/memory/lexical.go

@@ -0,0 +1,134 @@
+package memory
+
+import (
+	"strings"
+	"unicode"
+)
+
+// stopwords — words that carry no topic. A question and a note that share only
+// these share nothing: "почему небо синее" and "сеть какая-то медленная" both
+// contain "какая"-shaped filler and are about different worlds.
+var stopwords = map[string]bool{
+	// interrogatives and demonstratives
+	"что": true, "чего": true, "какой": true, "какая": true, "какое": true,
+	"какие": true, "каких": true, "кто": true, "кого": true, "кому": true,
+	"почему": true, "зачем": true, "где": true, "куда": true, "откуда": true,
+	"когда": true, "сколько": true, "как": true, "то": true, "это": true,
+	"этот": true, "тот": true, "там": true, "тут": true, "такой": true,
+	// pronouns — every sentence he says is about him, so "я" is not a topic
+	"я": true, "меня": true, "мне": true, "мой": true, "моя": true, "мои": true,
+	"ты": true, "тебя": true, "тебе": true, "твой": true, "он": true, "она": true,
+	"они": true, "мы": true, "себя": true, "свой": true,
+	// prepositions, conjunctions, particles, copulas
+	"в": true, "во": true, "на": true, "с": true, "со": true, "у": true,
+	"о": true, "об": true, "про": true, "за": true, "из": true, "по": true,
+	"до": true, "от": true, "для": true, "над": true, "под": true, "при": true,
+	"и": true, "а": true, "но": true, "или": true, "же": true, "ли": true,
+	"не": true, "ни": true, "бы": true, "был": true, "была": true, "было": true,
+	"быть": true, "есть": true, "был-ли": true, "уже": true, "ещё": true,
+	"еще": true, "так": true, "вот": true, "там-же": true,
+	// English filler, for the mixed utterances he does say
+	"the": true, "a": true, "an": true, "is": true, "are": true, "was": true,
+	"were": true, "be": true, "of": true, "in": true, "on": true, "at": true,
+	"to": true, "for": true, "about": true, "and": true, "or": true, "not": true,
+	"what": true, "who": true, "why": true, "when": true, "where": true,
+	"which": true, "how": true, "i": true, "my": true, "me": true, "it": true,
+	"this": true, "that": true,
+}
+
+// firstPerson — the words that make an utterance a question about his own
+// life. Not possession only: "как я восстановил конфиги" owns nothing and is
+// still about him.
+var firstPerson = map[string]bool{
+	"я": true, "меня": true, "мне": true, "мной": true, "мой": true,
+	"моя": true, "моё": true, "мое": true, "мои": true, "моего": true,
+	"моей": true, "моих": true, "моим": true, "себя": true, "свой": true,
+	"своя": true, "свои": true, "своего": true, "мною": true,
+	"i": true, "me": true, "my": true, "mine": true, "myself": true,
+}
+
+// RecallAllowed is the second half of the recall gate (#470). A hit that
+// cleared the score and margin gate may still be about something else
+// entirely: the held-out fixture puts the right note at 0.791-0.890 and the
+// must-be-silent cases at 0.795-0.835, so no threshold sits between them, and
+// a note about his slow network answered "почему небо синее?".
+//
+// The veto applies only to a question that mentions nothing of his. That
+// restriction is what keeps the fix from costing more than it saves: recall
+// exists to find the note whose words he no longer remembers, and demanding a
+// shared word of every recall silenced four true recalls on the fixture to
+// kill one false one. A question about his own life keeps the embedder alone
+// as its judge. A question about the world has to name something the memory
+// actually mentions.
+//
+// The veto's price was re-measured on 2026-08-03 (#496,
+// docs/evals/2026-08-03-recall-topic-veto.md). It costs one true recall and
+// buys one false one, and the fixture pass count is the same either way. The
+// lost case is an English paraphrase, not the cross-language loss it was
+// reported as, and the fixture has no cross-language case at all. Do not add a
+// script test or a bilingual stem map for it — both are no-ops here. The
+// separating signal is semantic and belongs in a reranker, not in this file.
+func RecallAllowed(query, text string) bool {
+	if mentionsHim(query) {
+		return true
+	}
+	return SharesContentWord(query, text)
+}
+
+func mentionsHim(query string) bool {
+	for _, w := range strings.FieldsFunc(strings.ToLower(query), func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	}) {
+		if firstPerson[w] {
+			return true
+		}
+	}
+	return false
+}
+
+// SharesContentWord reports whether query and text have at least one topic
+// word in common, after dropping the words that carry no topic. Stems are
+// compared, so the note and the question do not have to inflect alike.
+func SharesContentWord(query, text string) bool {
+	q := contentWords(query)
+	if len(q) == 0 {
+		// Nothing to compare — a question made entirely of filler. The score
+		// gate is then the only judge it can have.
+		return true
+	}
+	t := contentWords(text)
+	for _, a := range q {
+		for _, b := range t {
+			if a == b || sameStem(a, b) {
+				return true
+			}
+		}
+	}
+	return false
+}
+
+func contentWords(s string) []string {
+	var out []string
+	for _, w := range strings.FieldsFunc(strings.ToLower(s), func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	}) {
+		if !stopwords[w] {
+			out = append(out, w)
+		}
+	}
+	return out
+}
+
+// sameStem is inflection and derivation tolerance: Russian marks case and
+// tense on the ending, and the note and the question rarely use the same form.
+// "воду" and "вода" are the same water, "кормить" and "корм" the same feeding.
+// All but the last rune of the shorter word must match, and never fewer than
+// three, which is what keeps "сеть" clear of "сеанс".
+func sameStem(a, b string) bool {
+	ar, br := []rune(a), []rune(b)
+	n := min(len(ar), len(br)) - 1
+	if n < 3 {
+		return false
+	}
+	return string(ar[:n]) == string(br[:n])
+}

internal/memory/lexical_test.go

@@ -0,0 +1,56 @@
+package memory
+
+import "testing"
+
+func TestRecallAllowed(t *testing.T) {
+	cases := []struct {
+		name        string
+		query, text string
+		want        bool
+	}{
+		// The #470 shape: a world question and a note about his box.
+		{"world question, unrelated note", "почему небо синее", "сеть какая-то медленная", false},
+		{"world question, unrelated fact", "какая столица Франции", "какая последняя версия языка Go", false},
+		{"silent fixture case", "во сколько отходит поезд", "бэкап запускается в три ночи", false},
+
+		// A world question that does name the topic keeps its answer.
+		{"world question, same topic", "какой поезд идёт в Минск", "поезда в Минск ходят утром", true},
+
+		// A question about his own life is judged by the embedder alone,
+		// because recall exists for words he no longer remembers.
+		{"about him, no shared word", "во сколько я обычно засыпаю", "ложусь около одиннадцати", true},
+		{"about him, english", "which colour scheme do i like", "тёмная тема везде", true},
+
+		// Inflection must not break a match.
+		{"inflected", "чем кормить кота", "корм для кота в шкафу", true},
+	}
+	for _, c := range cases {
+		if got := RecallAllowed(c.query, c.text); got != c.want {
+			t.Errorf("%s: RecallAllowed(%q, %q) = %v, want %v", c.name, c.query, c.text, got, c.want)
+		}
+	}
+}
+
+// The known cost of the veto and the thing that pays for it, both measured on
+// the held-out fixture with the real embedder (#496,
+// docs/evals/2026-08-03-recall-topic-veto.md). The two are one lexical class:
+// zero shared content words, no first-person marker, scores 0.826 against 0.835
+// and margins 0.023 against 0.019. Recovering the first re-admits the second,
+// which puts false recall back to 1/5. Anyone loosening the veto has to move
+// the first line without moving the second.
+func TestRecallVetoTradeIsPinned(t *testing.T) {
+	if RecallAllowed("what fixed the screen problem", "the flicker went away once i swapped the display cable") {
+		t.Error("en-hard-024 is expected to stay vetoed — if this passes now, re-measure false recall before celebrating")
+	}
+	if RecallAllowed("во сколько отходит поезд", "погулял вдоль реки") {
+		t.Error("ru-silent-029 must stay vetoed — this is the false recall the veto exists to stop")
+	}
+}
+
+// A question made only of filler has no topic word to match on, and the score
+// gate is then the only judge it can have.
+func TestRecallAllowedFallsBackWhenNothingToCompare(t *testing.T) {
+	if !RecallAllowed("что это", "сеть какая-то медленная") {
+		t.Error("a question with no content word must not be vetoed")
+	}
+}

internal/memory/recalleval/recalleval.go

@@ -378,7 +378,7 @@ func scoreCase(ctx context.Context, emb router.Embedder, newStore NewStore, minS
 		if len(hits) > 1 {
 			o.Margin = hits[0].Score - hits[1].Score
 		}
-		o.Recalled = bestRecall(hits, minScore, minMargin)
+		o.Recalled = bestRecall(c.Query, hits, minScore, minMargin)
 	}
 	for i, h := range hits {
 		if h.ID != c.Want {
@@ -424,11 +424,19 @@ func rankNote(inTop3 bool) string {
 // is not importable; recalleval_test.go asserts the two agree in behaviour.
 // The daemon returns the whole hit (a note and a fact are said differently);
 // the harness only scores what came back, so it keeps returning the text.
-func bestRecall(results []memory.Result, minScore, minMargin float64) string {
+// bestRecall mirrors the daemon's gate in cmd/mavend/recall.go, including the
+// topic veto added for #470: a score that clears the gate still has to be
+// about what he asked. Keep the two in step — a fixture that measures a
+// weaker gate than the daemon runs flatters it.
+func bestRecall(query string, results []memory.Result, minScore, minMargin float64) string {
 	if !memory.Confident(results, minScore, minMargin) {
 		return ""
 	}
-	return results[0].Meta["text"]
+	text := results[0].Meta["text"]
+	if !memory.RecallAllowed(query, text) {
+		return ""
+	}
+	return text
 }
 
 func bump(m map[string]TagStat, key string, pass bool) {

internal/memory/recalleval/recalleval_test.go

@@ -140,21 +140,22 @@ func words(s string) []string {
 
 // TestBestRecallMatchesDaemon — the harness duplicates bestRecall from
 // cmd/mavend/recall.go (package main is not importable). This pins the copy to
-// the original's three rules: no hits, below the gate, or no text ⇒ silence.
+// the original's rules: no hits, below the gate, no text, or no shared topic
+// word ⇒ silence.
 func TestBestRecallMatchesDaemon(t *testing.T) {
-	if got := bestRecall(nil, 0.55, 0); got != "" {
+	if got := bestRecall("чай", nil, 0.55, 0); got != "" {
 		t.Errorf("no hits: got %q, want silence", got)
 	}
 	low := []memory.Result{{ID: "a", Score: 0.4, Meta: map[string]string{"text": "чай"}}}
-	if got := bestRecall(low, 0.55, 0); got != "" {
+	if got := bestRecall("чай", low, 0.55, 0); got != "" {
 		t.Errorf("below gate: got %q, want silence", got)
 	}
 	noText := []memory.Result{{ID: "a", Score: 0.9, Meta: map[string]string{}}}
-	if got := bestRecall(noText, 0.55, 0); got != "" {
+	if got := bestRecall("чай", noText, 0.55, 0); got != "" {
 		t.Errorf("no text: got %q, want silence", got)
 	}
 	ok := []memory.Result{{ID: "a", Score: 0.9, Meta: map[string]string{"text": "чай"}}}
-	if got := bestRecall(ok, 0.55, 0); got != "чай" {
+	if got := bestRecall("чай", ok, 0.55, 0); got != "чай" {
 		t.Errorf("above gate: got %q, want %q", got, "чай")
 	}
 	// Margin: a close runner-up means the embedder cannot tell the two apart,
@@ -163,17 +164,23 @@ func TestBestRecallMatchesDaemon(t *testing.T) {
 		{ID: "a", Score: 0.86, Meta: map[string]string{"text": "чай"}},
 		{ID: "b", Score: 0.85, Meta: map[string]string{"text": "кофе"}},
 	}
-	if got := bestRecall(close, 0.55, 0.03); got != "" {
+	if got := bestRecall("чай", close, 0.55, 0.03); got != "" {
 		t.Errorf("thin margin: got %q, want silence", got)
 	}
-	if got := bestRecall(close, 0.55, 0); got != "чай" {
+	if got := bestRecall("чай", close, 0.55, 0); got != "чай" {
 		t.Errorf("margin off: got %q, want %q", got, "чай")
 	}
+	// The topic veto (#470): the score is fine and the note is about
+	// something else.
+	offTopic := []memory.Result{{ID: "a", Score: 0.9, Meta: map[string]string{"text": "сеть какая-то медленная"}}}
+	if got := bestRecall("почему небо синее", offTopic, 0.55, 0); got != "" {
+		t.Errorf("off topic: got %q, want silence", got)
+	}
 	clear := []memory.Result{
 		{ID: "a", Score: 0.86, Meta: map[string]string{"text": "чай"}},
 		{ID: "b", Score: 0.70, Meta: map[string]string{"text": "кофе"}},
 	}
-	if got := bestRecall(clear, 0.55, 0.03); got != "чай" {
+	if got := bestRecall("чай", clear, 0.55, 0.03); got != "чай" {
 		t.Errorf("wide margin: got %q, want %q", got, "чай")
 	}
 }

internal/phraser/eval/checks.go

@@ -67,6 +67,19 @@ func RunChecks(c Case, body, mood string) []Result {
 	}
 }
 
+// Feminine, HisGender and Address expose three checks one at a time, so the
+// daemon can run them on a phrased message before he hears it (Vikunja #399).
+// Only these three: they are unambiguous string tests with nothing to compare
+// against, while length is path-specific and ontopic needs the fixture's
+// expected fragments, which do not exist at runtime.
+func Feminine(body string) Result { return checkFeminine(body) }
+
+// HisGender — see checkHisGender.
+func HisGender(body string) Result { return checkHisGender(body) }
+
+// Address — see checkAddress.
+func Address(body string) Result { return checkAddress(body) }
+
 func checkMood(mood string) Result {
 	if Moods[mood] {
 		return Result{CheckMood, true, ""}

internal/phraser/eval/talk.go

@@ -26,20 +26,22 @@ import (
 	"time"
 
 	"github.com/kami/maven/internal/dialogue"
+	"github.com/kami/maven/internal/router"
 )
 
 //go:embed talk_v1.json
 var talkFixtureJSON []byte
 
-// The three phrasing paths under test. Values match the fixture's "path" field.
+// The phrasing paths under test. Values match the fixture's "path" field.
 const (
 	PathChat      = "chat"      // PhraseChat
 	PathQuery     = "query"     // PhraseQuery with notes
 	PathKnowledge = "knowledge" // PhraseQuery with no notes
+	PathReply     = "reply"     // PhraseReply, the reactive confirmation
 )
 
 // TalkPaths — report order.
-var TalkPaths = []string{PathChat, PathQuery, PathKnowledge}
+var TalkPaths = []string{PathChat, PathQuery, PathKnowledge, PathReply}
 
 // TalkCheckNames — the checks that apply to a free-form reply, in report order.
 // Deliberately a subset of CheckNames: length, mood and "no questions" are nudge
@@ -58,12 +60,19 @@ var TalkCheckNames = []string{
 // WantAny is the on-topic contract: at least one lowercased fragment must appear
 // in the reply. Fragments are stems ("пароль" → "парол") so declension does not
 // defeat them.
+//
+// Intent, Key and Value carry the reply path's decision: that path is phrased
+// from what the router already resolved, not from the raw utterance. Utterance
+// stays filled anyway, because it is what a human reads in the report.
 type TalkCase struct {
 	ID        string   `json:"id"`
 	Path      string   `json:"path"`
 	Utterance string   `json:"utterance"`
 	History   []string `json:"history,omitempty"`
 	Notes     []string `json:"notes,omitempty"`
+	Intent    string   `json:"intent,omitempty"`
+	Key       string   `json:"key,omitempty"`
+	Value     string   `json:"value,omitempty"`
 	WantAny   []string `json:"want_any"`
 	Tags      []string `json:"tags,omitempty"`
 	Note      string   `json:"note,omitempty"`
@@ -92,13 +101,27 @@ func LoadTalk() (TalkFixture, error) {
 	return f, nil
 }
 
-// Talker — the two methods a conversational path must have to be scorable.
-// *phraser.LLMPhraser satisfies it; same trick as Nudger.
+// Talker — the methods a conversational path must have to be scorable.
+// *phraser.LLMPhraser satisfies the first two; *phraser.Replier satisfies the
+// third, so a run that scores all four paths passes a Pair.
 type Talker interface {
 	PhraseChat(ctx context.Context, utterance string, history []dialogue.Turn) (string, error)
 	PhraseQuery(ctx context.Context, utterance string, notes []string) (string, error)
 }
 
+// Confirmer — the reply path. *phraser.Replier satisfies it.
+type Confirmer interface {
+	PhraseReply(ctx context.Context, d router.Decision) (string, error)
+}
+
+// Pair joins the two objects the daemon wires separately — the phraser and the
+// replier — so one ScoreTalk call covers every path Maven speaks through. A bare
+// Talker still works; its reply cases score as errors, which is honest.
+type Pair struct {
+	Talker
+	Confirmer
+}
+
 // TalkOutcome — one scored case.
 type TalkOutcome struct {
 	Case    TalkCase
@@ -194,10 +217,30 @@ func (c TalkCase) run(ctx context.Context, t Talker) (string, error) {
 		return t.PhraseQuery(ctx, c.Utterance, c.Notes)
 	case PathKnowledge:
 		return t.PhraseQuery(ctx, c.Utterance, nil)
+	case PathReply:
+		conf, ok := t.(Confirmer)
+		if !ok {
+			return "", fmt.Errorf("target cannot phrase replies — pass a Pair")
+		}
+		return conf.PhraseReply(ctx, c.decision())
 	}
 	return "", fmt.Errorf("unknown path %q", c.Path)
 }
 
+// decision rebuilds what the router would have handed the replier. Text is the
+// utterance for a note or a reminder, which is what the router puts there.
+func (c TalkCase) decision() router.Decision {
+	return router.Decision{
+		Intent: router.Intent(c.Intent),
+		Slots: router.Slots{
+			Key:    c.Key,
+			Value:  c.Value,
+			Text:   c.Utterance,
+			HasKey: c.Key != "",
+		},
+	}
+}
+
 func (c TalkCase) turns() []dialogue.Turn {
 	turns := make([]dialogue.Turn, 0, len(c.History))
 	for _, h := range c.History {

internal/phraser/eval/talk_test.go

@@ -11,6 +11,7 @@ import (
 	"github.com/kami/maven/internal/llm"
 	"github.com/kami/maven/internal/persona"
 	"github.com/kami/maven/internal/phraser"
+	"github.com/kami/maven/internal/router"
 )
 
 // perPathMinimum — the resolution floor. A per-path score built on a handful of
@@ -36,6 +37,10 @@ func TestTalkFixture(t *testing.T) {
 
 		switch c.Path {
 		case PathChat, PathQuery, PathKnowledge:
+		case PathReply:
+			if c.Intent == "" {
+				t.Errorf("%s: reply case has no intent — the replier is phrased from the decision", c.ID)
+			}
 		default:
 			t.Errorf("%s: unknown path %q", c.ID, c.Path)
 		}
@@ -69,10 +74,15 @@ type fakeTalker struct{ reply string }
 func (f fakeTalker) PhraseChat(context.Context, string, []dialogue.Turn) (string, error) {
 	return f.reply, nil
 }
+
 func (f fakeTalker) PhraseQuery(context.Context, string, []string) (string, error) {
 	return f.reply, nil
 }
 
+func (f fakeTalker) PhraseReply(context.Context, router.Decision) (string, error) {
+	return f.reply, nil
+}
+
 // TestScoreTalkCounts — a reply that fails on purpose must be counted on every
 // path, so a real run cannot report a hidden zero.
 func TestScoreTalkCounts(t *testing.T) {
@@ -104,7 +114,7 @@ func TestScoreTalkCounts(t *testing.T) {
 	}
 }
 
-// TestLLMTalkBaseline — the resident model on the three conversational paths.
+// TestLLMTalkBaseline — the resident model on all four phrasing paths.
 // Opt-in exactly like TestLLMPhrasingBaseline: CI has no model and a run costs
 // minutes on the CPU target.
 //
@@ -148,7 +158,12 @@ func TestLLMTalkBaseline(t *testing.T) {
 	}
 	t.Logf("scoring model %s at %s", model, base)
 
-	rep, err := ScoreTalk(ctx, "llm ("+model+", built-in persona)", p, f)
+	// The reply path is a separate object in the daemon too: the phraser owns its
+	// own llama-server, the replier is handed an llm.Client. Pair scores both.
+	block := func() string { return persona.Facts{}.Block(time.Now()) }
+	target := Pair{Talker: p, Confirmer: phraser.NewReplier(llm.New(base, cfg.Timeout), block)}
+
+	rep, err := ScoreTalk(ctx, "llm ("+model+", built-in persona)", target, f)
 	if err != nil {
 		t.Fatalf("ScoreTalk: %v", err)
 	}

internal/phraser/eval/talk_v1.json

@@ -222,6 +222,90 @@
       "utterance": "почему гром слышно позже молнии?",
       "want_any": ["звук", "све", "быстр", "гром", "молни"],
       "tags": ["general"]
+    },
+    {
+      "id": "reply-fact-coffee",
+      "path": "reply",
+      "intent": "fact",
+      "key": "кофе",
+      "value": "закончился",
+      "utterance": "кофе закончился",
+      "want_any": ["коф"],
+      "tags": ["fact"],
+      "note": "The plainest confirmation there is, and the sentence he hears most often."
+    },
+    {
+      "id": "reply-fact-weight",
+      "path": "reply",
+      "intent": "fact",
+      "key": "вес",
+      "value": "82",
+      "utterance": "мой вес 82",
+      "want_any": ["вес", "82"],
+      "tags": ["fact", "number"],
+      "note": "A number must survive into the confirmation; a paraphrase that drops it is useless."
+    },
+    {
+      "id": "reply-fact-pill",
+      "path": "reply",
+      "intent": "fact",
+      "key": "таблетки",
+      "value": "выпил",
+      "utterance": "таблетки выпил",
+      "want_any": ["таблетк"],
+      "tags": ["fact", "feminine"],
+      "note": "He says 'выпил', masculine and about himself. She must not copy the form onto herself."
+    },
+    {
+      "id": "reply-note-router",
+      "path": "reply",
+      "intent": "note",
+      "utterance": "роутер перезагружается сам по ночам",
+      "want_any": ["роутер"],
+      "tags": ["note"]
+    },
+    {
+      "id": "reply-note-long",
+      "path": "reply",
+      "intent": "note",
+      "utterance": "если диск снова отвалится, посмотреть кабель, а не контроллер, в прошлый раз был кабель",
+      "want_any": ["диск", "кабел"],
+      "tags": ["note", "length"],
+      "note": "A long note baits a long confirmation. One sentence is the contract."
+    },
+    {
+      "id": "reply-reminder-evening",
+      "path": "reply",
+      "intent": "reminder",
+      "utterance": "напомни вечером полить цветы",
+      "want_any": ["цвет", "полит", "вечер"],
+      "tags": ["reminder"]
+    },
+    {
+      "id": "reply-reminder-tomorrow",
+      "path": "reply",
+      "intent": "reminder",
+      "utterance": "напомни завтра позвонить в поликлинику",
+      "want_any": ["поликлиник", "позвон", "звон"],
+      "tags": ["reminder"]
+    },
+    {
+      "id": "reply-formality-bait",
+      "path": "reply",
+      "intent": "note",
+      "utterance": "запишите пожалуйста что счётчики я сдал",
+      "want_any": ["счётчик", "счетчик"],
+      "tags": ["note", "persona-bait", "address"],
+      "note": "Polite plural in the input. The confirmation must still be на ты."
+    },
+    {
+      "id": "reply-question-bait",
+      "path": "reply",
+      "intent": "note",
+      "utterance": "надо купить фильтр для воды, не помню какой",
+      "want_any": ["фильтр"],
+      "tags": ["note", "no-question"],
+      "note": "An unresolved note invites her to ask which filter. A confirmation does not ask."
     }
   ]
 }

internal/phraser/llmphraser.go

@@ -1,6 +1,7 @@
 package phraser
 
 import (
+	"bufio"
 	"bytes"
 	"context"
 	"encoding/json"
@@ -8,6 +9,7 @@ import (
 	"io"
 	"log"
 	"net/http"
+	"os"
 	"os/exec"
 	"regexp"
 	"strings"
@@ -46,6 +48,12 @@ type LLMPhraser struct {
 	launch func(ctx context.Context, cfg Config) (backend, error)
 	probe  func(ctx context.Context, base string) (string, error)
 
+	// remote — the workstation model, when one is configured. Set once at wiring
+	// time by UseRemote and read on every phrasing call. nil ⇒ every call goes to
+	// the resident llama-server this phraser owns, which is the whole deploy
+	// before a `workstation` block exists. See world.go.
+	remote Remote
+
 	// swapMu — single-flight around Swap. Held for the whole swap, including the
 	// model load, so two concurrent swap requests can never both be loading.
 	swapMu sync.Mutex
@@ -77,6 +85,19 @@ type Config struct {
 	NCtx       int
 	Timeout    time.Duration
 
+	// CacheRAMMiB bounds llama-server's prompt cache, which is what actually ate
+	// this box. Measured on homesrv 2026-08-03: the server's own default limit is
+	// 8192 MiB, it stores the full KV state of every idle slot it evicts (112 kiB
+	// per token, so 166 MiB for one 1521-token prompt), and RSS climbed by that
+	// much per distinct prompt until it hit 7.9 GB and half a gigabyte went to
+	// swap. Weights are only 1.1 GB and mmapped, and -ngl 99 costs almost no RSS
+	// because RADV keeps device memory outside the process.
+	//
+	// 0 ⇒ the flag is not passed and the server's own 8 GiB default applies. That
+	// is the escape hatch for a llama-server too old to know --cache-ram, not a
+	// recommendation. See docs/evals/2026-08-03-llama-prompt-cache.md.
+	CacheRAMMiB int
+
 	// ContextBlock renders the shared context block (who he is, how to
 	// address him, the time) fresh for each turn. See internal/persona.
 	// nil ⇒ no block, the prompts stand alone.
@@ -110,7 +131,9 @@ func DefaultConfig(modelPath string) Config {
 		Listen:     "127.0.0.1:0",
 		NGpuLayers: -1,
 		NCtx:       2048,
-		Timeout:    30 * time.Second,
+		// 512 MiB caps total RSS near 1 GB and still holds several recent prompts.
+		CacheRAMMiB: 512,
+		Timeout:     30 * time.Second,
 	}
 }
 
@@ -219,8 +242,10 @@ func spawnLlamaServer(ctx context.Context, cfg Config) (backend, error) {
 	return p, nil
 }
 
-func startLlamaProc(ctx context.Context, cfg Config) (*llamaProc, error) {
-	p := &llamaProc{}
+// llamaArgs is the command line for one resident server. It is a function and
+// not an inline literal because kill-maven.sh's orphan sweep matches against
+// this exact line, and a test pins the two together.
+func llamaArgs(cfg Config) []string {
 	args := []string{
 		"-m", cfg.ModelPath,
 		"--host", "127.0.0.1",
@@ -229,7 +254,15 @@ func startLlamaProc(ctx context.Context, cfg Config) (*llamaProc, error) {
 		"-ngl", fmt.Sprintf("%d", cfg.NGpuLayers),
 		"--no-webui",
 	}
-	cmd := exec.CommandContext(ctx, cfg.BinPath, args...)
+	if cfg.CacheRAMMiB > 0 {
+		args = append(args, "--cache-ram", fmt.Sprintf("%d", cfg.CacheRAMMiB))
+	}
+	return args
+}
+
+func startLlamaProc(ctx context.Context, cfg Config) (*llamaProc, error) {
+	p := &llamaProc{}
+	cmd := exec.CommandContext(ctx, cfg.BinPath, llamaArgs(cfg)...)
 	// Pdeathsig: the kernel SIGKILLs llama-server the moment mavend dies — by
 	// ANY means, including SIGKILL/OOM/panic where our Close() never runs. Without
 	// it a hard-killed mavend orphans its llama-server (reparented to init, keeps
@@ -240,63 +273,104 @@ func startLlamaProc(ctx context.Context, cfg Config) (*llamaProc, error) {
 	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true, Pdeathsig: syscall.SIGKILL}
 	p.cmd = cmd
 
-	stderr, err := cmd.StderrPipe()
+	// One pipe for both streams. llama.cpp writes its buffer sizes, KV-cache
+	// layout and offload lines to stderr and its request log to stdout, and
+	// stdout used to go nowhere at all — so nothing about the model's memory was
+	// diagnosable from a running box. Both ends land in mavend's log now.
+	pr, pw, err := os.Pipe()
 	if err != nil {
-		return nil, fmt.Errorf("llm: stderr pipe: %w", err)
+		return nil, fmt.Errorf("llm: output pipe: %w", err)
 	}
+	cmd.Stdout = pw
+	cmd.Stderr = pw
 
 	if err := cmd.Start(); err != nil {
-		stderr.Close()
+		pr.Close()
+		pw.Close()
 		return nil, fmt.Errorf("llm: start: %w", err)
 	}
+	// The child holds the only other reference to the write end. Dropping ours
+	// is what makes the reader see EOF when the child dies.
+	pw.Close()
 
 	portCh := make(chan string, 1)
 	errCh := make(chan error, 1)
+	tail := &lineTail{}
 	p.wg.Add(1)
 	go func() {
 		defer p.wg.Done()
-		buf := make([]byte, 4096)
-		var leftover []byte
-		for {
-			n, err := stderr.Read(buf)
-			if n > 0 {
-				data := append(leftover, buf[:n]...)
-				lines := bytes.Split(data, []byte("\n"))
-				for _, line := range lines[:len(lines)-1] {
-					if m := listenRE.FindSubmatch(line); len(m) > 1 {
-						addr := string(m[1])
-						portCh <- addr
-						close(portCh)
-					}
+		defer pr.Close()
+		sc := bufio.NewScanner(pr)
+		// llama.cpp prints one prompt per line and a prompt can be long.
+		sc.Buffer(make([]byte, 0, 64*1024), 1024*1024)
+		listening := false
+		for sc.Scan() {
+			line := sc.Bytes()
+			log.Printf("llama: %s", line)
+			if !listening {
+				tail.add(string(line))
+				if m := listenRE.FindSubmatch(line); len(m) > 1 {
+					listening = true
+					portCh <- string(m[1])
+					close(portCh)
 				}
-				leftover = lines[len(lines)-1]
-			}
-			if err != nil {
-				errCh <- err
-				return
 			}
 		}
+		err := sc.Err()
+		if err == nil {
+			err = io.EOF
+		}
+		errCh <- err
 	}()
 
+	fail := func(err error) (*llamaProc, error) {
+		_ = cmd.Process.Kill()
+		_ = cmd.Wait()
+		return nil, err
+	}
 	select {
 	case addr := <-portCh:
 		p.base = addr
 		return p, nil
 	case err := <-errCh:
-		_ = cmd.Process.Kill()
-		_ = cmd.Wait()
-		return nil, fmt.Errorf("llm: server output: %w", err)
+		// The tail is the whole diagnosis when the server dies during load: bare
+		// "EOF" never said which layer or which allocation it choked on.
+		return fail(fmt.Errorf("llm: server output: %w; last output: %s", err, tail.String()))
 	case <-ctx.Done():
-		_ = cmd.Process.Kill()
-		_ = cmd.Wait()
-		return nil, ctx.Err()
+		return fail(ctx.Err())
 	case <-time.After(60 * time.Second):
-		_ = cmd.Process.Kill()
-		_ = cmd.Wait()
-		return nil, fmt.Errorf("llm: server did not start within 60s")
+		return fail(fmt.Errorf("llm: server did not start within 60s; last output: %s", tail.String()))
 	}
 }
 
+// lineTail keeps the last few startup lines so a server that dies before it
+// listens can say why in the error, not just "EOF". Written by the reader
+// goroutine and read by whoever gives up on startup, so it takes a lock.
+type lineTail struct {
+	mu    sync.Mutex
+	lines []string
+}
+
+const lineTailMax = 12
+
+func (t *lineTail) add(line string) {
+	t.mu.Lock()
+	defer t.mu.Unlock()
+	t.lines = append(t.lines, line)
+	if len(t.lines) > lineTailMax {
+		t.lines = t.lines[len(t.lines)-lineTailMax:]
+	}
+}
+
+func (t *lineTail) String() string {
+	t.mu.Lock()
+	defer t.mu.Unlock()
+	if len(t.lines) == 0 {
+		return "(no output)"
+	}
+	return strings.Join(t.lines, " | ")
+}
+
 // BaseURL is the llama-server this phraser talks to right now. It changes when
 // the model is swapped, so callers that cache it must register an observer
 // (OnSwap) rather than keeping the string forever.
@@ -363,10 +437,7 @@ func (p *LLMPhraser) PhraseQuery(ctx context.Context, utterance string, notes []
 	// prompt guaranteed to make a small model fill the gap from memory.
 	notes = nonEmpty(notes)
 	if len(notes) == 0 {
-		// General knowledge — no notes to ground the answer. The system
-		// prompt is the single tested source in router.KnowledgePrompt.
-		sys := persona.Prepend(p.cfg.ContextBlock, router.KnowledgePrompt())
-		prompt := fmt.Sprintf("Пользователь спрашивает: \"%s\".", utterance)
+		sys, prompt := p.knowledgePrompt(utterance)
 		resp, err := p.chatWithSystem(ctx, sys, prompt, 768)
 		if err != nil || resp == "" {
 			return "не знаю.", nil
@@ -381,11 +452,7 @@ func (p *LLMPhraser) PhraseQuery(ctx context.Context, utterance string, notes []
 		}
 		return resp, nil
 	}
-	sys := p.querySystemPrompt()
-	prompt := fmt.Sprintf(
-		"Он спрашивает: \"%s\"\n\nИсточники:\n%s\nОтветь ему коротко и своими словами, опираясь только на эти источники. Если ответа в них нет — так и скажи.",
-		utterance, evidenceBlock(notes),
-	)
+	sys, prompt := p.evidencePrompt(utterance, notes)
 	resp, err := p.chatWithSystem(ctx, sys, prompt, 768)
 	text, _, perr := parseResponseMood(resp)
 	if err != nil || perr != nil {
@@ -481,6 +548,17 @@ func chatSystemPrompt(block func() string) string {
 // the LLM completion endpoint. Like chatWithSystem but for an arbitrary message
 // slice — the caller owns the system prompt placement.
 func (p *LLMPhraser) chatWithMessages(ctx context.Context, msgs []chatMsg, maxTokens int) (string, error) {
+	// Same silent preference as chatWithSystem, when the array is the shape
+	// llm.Req can carry: one system turn and one user turn. PhraseChat already
+	// folds the history into a single user message (some chat templates reject
+	// consecutive user turns), so today that is every call. A longer array goes
+	// to the resident model rather than get flattened here, because flattening a
+	// conversation is a decision its owner should make.
+	if len(msgs) == 2 && msgs[0].Role == "system" && msgs[1].Role == "user" {
+		if out, ok := p.remoteChat(ctx, msgs[0].Content, msgs[1].Content, maxTokens); ok {
+			return out, nil
+		}
+	}
 	base, release, err := p.acquire()
 	if err != nil {
 		return "", err
@@ -640,6 +718,12 @@ func (p *LLMPhraser) chat(ctx context.Context, userPrompt string) (string, error
 }
 
 func (p *LLMPhraser) chatWithSystem(ctx context.Context, system, user string, maxTokens int) (string, error) {
+	// The workstation model first when it will take work, and silently: every
+	// caller of this helper is on the silent half of the degradation rule. It
+	// answering is not news, and it being asleep is not news either.
+	if out, ok := p.remoteChat(ctx, system, user, maxTokens); ok {
+		return out, nil
+	}
 	base, release, err := p.acquire()
 	if err != nil {
 		return "", err
@@ -733,6 +817,27 @@ func (p *LLMPhraser) systemPrompt() string {
 	return persona.Prepend(p.cfg.ContextBlock, nudgeSystem)
 }
 
+// knowledgePrompt — the no-sources branch: a world question, answered from
+// weights alone. The system prompt is the single tested source in
+// router.KnowledgePrompt.
+//
+// Split out of PhraseQuery so PhraseWorld sends the workstation model the same
+// bytes the resident model gets. Prompt parity across two models is a stated
+// constraint (CLAUDE.md), and two copies of a prompt is how it stops holding.
+func (p *LLMPhraser) knowledgePrompt(utterance string) (sys, user string) {
+	return persona.Prepend(p.cfg.ContextBlock, router.KnowledgePrompt()),
+		fmt.Sprintf("Пользователь спрашивает: \"%s\".", utterance)
+}
+
+// evidencePrompt — the sources branch: read these, add nothing. Shared with
+// PhraseWorld for the same reason as knowledgePrompt.
+func (p *LLMPhraser) evidencePrompt(utterance string, notes []string) (sys, user string) {
+	return p.querySystemPrompt(), fmt.Sprintf(
+		"Он спрашивает: \"%s\"\n\nИсточники:\n%s\nОтветь ему коротко и своими словами, опираясь только на эти источники. Если ответа в них нет — так и скажи.",
+		utterance, evidenceBlock(notes),
+	)
+}
+
 // querySystemPrompt returns the system prompt for the evidence branch of
 // PhraseQuery. Prepends the configured persona when set.
 //

internal/phraser/replier.go

@@ -0,0 +1,112 @@
+// phraser/replier.go — reactive reply phrasing, the confirmation he hears
+// after every fact, note and reminder.
+//
+// It lived in cmd/mavend as package main until Vikunja #396, which meant the
+// most frequently heard sentence Maven says was the one path the phrasing eval
+// could not import, let alone score. Nothing here talks to the daemon: the
+// caller supplies the completer and the context block, and cmd/mavend keeps the
+// stub fallback so a model error still answers.
+package phraser
+
+import (
+	"context"
+	"strings"
+	"time"
+
+	"github.com/kami/maven/internal/llm"
+	"github.com/kami/maven/internal/persona"
+	"github.com/kami/maven/internal/router"
+)
+
+// Completer is the model seam for the replier, a subset of router.Completer.
+// *llm.Client satisfies it.
+type Completer interface {
+	Complete(ctx context.Context, r llm.Req) (string, error)
+}
+
+// replyTimeout bounds one reply. Generous because the resident model on the CPU
+// floor is slow and the caller has a deterministic fallback anyway.
+const replyTimeout = 60 * time.Second
+
+// ReplySystemPrompt — the reactive confirmation contract: one short Russian
+// sentence, feminine self-reference, informal address, no question.
+const ReplySystemPrompt = `Ты — Maven, домашняя ассистентка (о себе — в женском роде). Владелец — мужчина, говоришь с ним на "ты", в единственном числе; никогда не "вы"/"ваш" и не "он"/"его". Подтверди действие РОВНО ОДНИМ коротким предложением (≤120 символов), по-русски, спокойно и без официальных формулировок. Не задавай вопросов, не повторяй слова, не добавляй ничего после точки. Отвечай ТОЛЬКО одним объектом JSON с полями "response" (текст) и "mood" (ровно одно из: neutral, happy, thinking, tired, confused).
+Пример: {"response": "Записала, что ты выпил стакан воды.", "mood": "neutral"}
+Никогда не пиши "..." в поле response.`
+
+// Replier phrases reactive confirmations with the resident model. It has no
+// fallback of its own: an error is returned, and the daemon answers from the
+// deterministic stub. That is also what makes it scorable — a dead server shows
+// up as an error rather than as bad phrasing.
+type Replier struct {
+	c Completer
+
+	// block renders the shared context block per turn (who he is, the time).
+	// nil ⇒ the prompt stands alone.
+	block func() string
+}
+
+// NewReplier builds a replier over c. block may be nil.
+func NewReplier(c Completer, block func() string) *Replier {
+	return &Replier{c: c, block: block}
+}
+
+// PhraseReply returns the confirmation for one decision. An empty string with a
+// nil error means the model produced nothing usable, which the caller must
+// treat exactly like an error.
+func (r *Replier) PhraseReply(ctx context.Context, d router.Decision) (string, error) {
+	ctx, cancel := context.WithTimeout(ctx, replyTimeout)
+	defer cancel()
+	out, err := r.c.Complete(ctx, llm.Req{
+		System:        persona.Prepend(r.block, ReplySystemPrompt),
+		User:          replyContext(d),
+		Grammar:       ResponseGrammar,
+		MaxTokens:     512,
+		RepeatPenalty: 1.3,
+	})
+	if err != nil {
+		return "", err
+	}
+	out = stripThink(out)
+	if response, _, perr := parseResponseMood(out); perr != nil {
+		return "", perr
+	} else if response != "" {
+		return response, nil
+	}
+	// fallback: the model answered in bare prose, which is fine here.
+	return firstSentence(out), nil
+}
+
+// firstSentence trims the model's output to a single clean confirmation: first
+// line, first sentence, whitespace-normalized — the last-line defense against a
+// small model that rambles past the first period despite the prompt + stop.
+func firstSentence(s string) string {
+	s = strings.TrimSpace(s)
+	if i := strings.IndexByte(s, '\n'); i >= 0 {
+		s = s[:i]
+	}
+	// keep up to and including the first sentence-ending punctuation.
+	if i := strings.IndexAny(s, ".!?"); i >= 0 {
+		s = s[:i+1]
+	}
+	return strings.TrimSpace(s)
+}
+
+// replyContext renders the decision into a compact RU description for the model.
+func replyContext(d router.Decision) string {
+	switch d.Intent {
+	case router.IntentFact:
+		return "записала факт: " + d.Slots.Key + " " + d.Slots.Value
+	case router.IntentNote:
+		return "сохранила заметку: " + d.Slots.Text
+	case router.IntentReminder:
+		return "поставила напоминание: " + d.Slots.Text
+	default:
+		return string(d.Intent) + ": " + d.Slots.Text
+	}
+}
+
+// StripThink removes the <think> block a Thinking-variant model emits before its
+// answer. Exported for the daemon's own model callers, which parse output that
+// never passes through a phraser method.
+func StripThink(s string) string { return stripThink(s) }

internal/phraser/replier_test.go

@@ -0,0 +1,90 @@
+package phraser
+
+import (
+	"context"
+	"testing"
+
+	"github.com/kami/maven/internal/llm"
+	"github.com/kami/maven/internal/router"
+)
+
+type mockCompleter struct {
+	out string
+	err error
+}
+
+func (m mockCompleter) Complete(_ context.Context, _ llm.Req) (string, error) { return m.out, m.err }
+
+func TestReplierReturnsLLMReply(t *testing.T) {
+	r := NewReplier(mockCompleter{out: `{"response":"записала, кофе закончился","mood":"neutral"}`}, nil)
+	got, err := r.PhraseReply(context.Background(), noteDecision())
+	if err != nil || got != "записала, кофе закончился" {
+		t.Errorf("got %q, %v, want %q, nil", got, err, "записала, кофе закончился")
+	}
+}
+
+func TestReplierFallsBackToPlainText(t *testing.T) {
+	r := NewReplier(mockCompleter{out: "записала, кофе закончился"}, nil)
+	got, err := r.PhraseReply(context.Background(), noteDecision())
+	if err != nil || got != "записала, кофе закончился" {
+		t.Errorf("got %q, %v, want %q, nil", got, err, "записала, кофе закончился")
+	}
+}
+
+func TestReplierReportsTheModelError(t *testing.T) {
+	r := NewReplier(mockCompleter{err: errTestLLMDown}, nil)
+	got, err := r.PhraseReply(context.Background(), noteDecision())
+	if err == nil {
+		t.Errorf("got %q, nil error — a dead model must be reported, not phrased around", got)
+	}
+}
+
+// A fragment the grammar left half-open is a failed generation. It must come
+// back as an error so the daemon reaches its stub, not as a reply.
+func TestReplierRejectsBrokenJSON(t *testing.T) {
+	r := NewReplier(mockCompleter{out: `{"response":"запис`}, nil)
+	got, err := r.PhraseReply(context.Background(), noteDecision())
+	if err == nil || got != "" {
+		t.Errorf("got %q, %v, want empty and an error", got, err)
+	}
+}
+
+func TestReplierEmptyOutputIsEmpty(t *testing.T) {
+	r := NewReplier(mockCompleter{out: ""}, nil)
+	got, err := r.PhraseReply(context.Background(), noteDecision())
+	if err != nil || got != "" {
+		t.Errorf("got %q, %v, want empty and no error", got, err)
+	}
+}
+
+// grammarRecorder captures the request so the grammar can be asserted on.
+type grammarRecorder struct{ req llm.Req }
+
+func (g *grammarRecorder) Complete(_ context.Context, r llm.Req) (string, error) {
+	g.req = r
+	return `{"response":"записала","mood":"neutral"}`, nil
+}
+
+func TestReplierCarriesTheResponseGrammar(t *testing.T) {
+	rec := &grammarRecorder{}
+	r := NewReplier(rec, nil)
+	if _, err := r.PhraseReply(context.Background(), noteDecision()); err != nil {
+		t.Fatalf("PhraseReply: %v", err)
+	}
+	if rec.req.Grammar != ResponseGrammar {
+		t.Errorf("grammar = %q, want ResponseGrammar", rec.req.Grammar)
+	}
+	if rec.req.System != ReplySystemPrompt {
+		t.Errorf("system prompt = %q, want ReplySystemPrompt", rec.req.System)
+	}
+}
+
+func noteDecision() router.Decision {
+	return router.Decision{Intent: router.IntentNote, Slots: router.Slots{Text: "кофе закончился"}}
+}
+
+var errTestLLMDown = errTest("llm down")
+
+type errTest string
+
+func (e errTest) Error() string { return string(e) }

internal/phraser/spawn_test.go

@@ -1,9 +1,11 @@
 package phraser
 
 import (
+	"bytes"
 	"context"
 	"errors"
 	"fmt"
+	"log"
 	"os"
 	"os/exec"
 	"path/filepath"
@@ -59,6 +61,19 @@ func TestExtractPort(t *testing.T) {
 	}
 }
 
+// The prompt cache is what ate 6.8GB of the deployed server's RSS, so the cap
+// has to reach the command line, and the opt-out has to leave it off.
+func TestLlamaArgsCapsPromptCache(t *testing.T) {
+	cfg := DefaultConfig("/m.gguf")
+	if got := strings.Join(llamaArgs(cfg), " "); !strings.Contains(got, "--cache-ram 512") {
+		t.Errorf("default args = %q, want --cache-ram 512", got)
+	}
+	cfg.CacheRAMMiB = 0
+	if got := strings.Join(llamaArgs(cfg), " "); strings.Contains(got, "--cache-ram") {
+		t.Errorf("args with the cap off = %q, want no --cache-ram flag", got)
+	}
+}
+
 func TestStartLlamaProcScrapesPortAndReaps(t *testing.T) {
 	bin := fakeLlama(t, listensThenSleeps)
 	ctx, cancel := context.WithCancel(context.Background())
@@ -86,6 +101,48 @@ func TestStartLlamaProcScrapesPortAndReaps(t *testing.T) {
 	}
 }
 
+// captureLog redirects the standard logger for the duration of a test and
+// returns what was written to it.
+func captureLog(t *testing.T) *bytes.Buffer {
+	t.Helper()
+	var buf bytes.Buffer
+	old := log.Writer()
+	flags := log.Flags()
+	log.SetOutput(&buf)
+	log.SetFlags(0)
+	t.Cleanup(func() { log.SetOutput(old); log.SetFlags(flags) })
+	return &buf
+}
+
+// The child's buffer-size, KV-cache and offload lines are the only way to
+// account for its memory on a running box, and they used to be dropped: stderr
+// was scraped for the listen line and thrown away, stdout was never piped.
+func TestStartLlamaProcForwardsChildOutput(t *testing.T) {
+	buf := captureLog(t)
+	bin := fakeLlama(t, `echo "load_tensors: Vulkan0 model buffer size = 1053.34 MiB" >&2
+echo "llama_context: KV self size = 448.00 MiB"
+`+listensThenSleeps)
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	p, err := startLlamaProc(ctx, testCfg(bin))
+	if err != nil {
+		t.Fatalf("startLlamaProc: %v", err)
+	}
+	p.cancel = cancel
+	defer p.Close()
+
+	got := buf.String()
+	for _, want := range []string{
+		"llama: load_tensors: Vulkan0 model buffer size = 1053.34 MiB", // stderr
+		"llama: llama_context: KV self size = 448.00 MiB",              // stdout, previously discarded
+	} {
+		if !strings.Contains(got, want) {
+			t.Errorf("log missing %q\nlog was:\n%s", want, got)
+		}
+	}
+}
+
 func TestStartLlamaProcFailureArms(t *testing.T) {
 	t.Run("binary missing", func(t *testing.T) {
 		cfg := testCfg(filepath.Join(t.TempDir(), "does-not-exist"))
@@ -96,13 +153,18 @@ func TestStartLlamaProcFailureArms(t *testing.T) {
 	})
 
 	t.Run("server exits without listening", func(t *testing.T) {
-		// stderr closes, so the reader goroutine reports EOF on errCh.
+		// stderr closes, so the reader goroutine reports EOF on errCh. The error
+		// must carry the child's last words: bare "EOF" named no cause.
+		captureLog(t)
 		bin := fakeLlama(t, `echo "ggml_vulkan: no device" >&2
 exit 1`)
 		_, err := startLlamaProc(context.Background(), testCfg(bin))
 		if err == nil || !strings.Contains(err.Error(), "llm: server output") {
 			t.Fatalf("err = %v, want the server-output arm", err)
 		}
+		if !strings.Contains(err.Error(), "ggml_vulkan: no device") {
+			t.Errorf("err = %v, want the child's last output in it", err)
+		}
 	})
 
 	t.Run("context cancelled during startup", func(t *testing.T) {
@@ -241,15 +303,7 @@ func TestKillMavenScriptMatchesRealCommandLine(t *testing.T) {
 	// startLlamaProc that breaks the sweep fails here instead of on the box.
 	cfg := DefaultConfig("/opt/maven/models/llm/Qwen3-1.7B-UD-Q4_K_XL.gguf")
 	cfg.NCtx, cfg.NGpuLayers = 4096, 99
-	cmdline := strings.Join([]string{
-		cfg.BinPath,
-		"-m", cfg.ModelPath,
-		"--host", "127.0.0.1",
-		"--port", extractPort(cfg.Listen),
-		"-c", fmt.Sprintf("%d", cfg.NCtx),
-		"-ngl", fmt.Sprintf("%d", cfg.NGpuLayers),
-		"--no-webui",
-	}, " ")
+	cmdline := cfg.BinPath + " " + strings.Join(llamaArgs(cfg), " ")
 	if !pat.MatchString(cmdline) {
 		t.Fatalf("kill-maven.sh pattern %q does not match %q — orphans would leak", m[1], cmdline)
 	}

internal/phraser/world.go

@@ -0,0 +1,137 @@
+package phraser
+
+import (
+	"context"
+	"errors"
+	"log"
+
+	"github.com/kami/maven/internal/llm"
+)
+
+// Remote — the workstation model, seen from the phraser. `*llm.Pair` satisfies
+// it, and a test fake satisfies it in three lines.
+//
+// Only the refusing half of Pair is here on purpose. Pair.Complete falls back to
+// its own floor client, and the phraser already owns a floor: the llama-server it
+// spawned. Two floors under one call is one too many, so the phraser asks whether
+// the remote will take work, uses it when it will, and otherwise does exactly
+// what it did before this file existed.
+type Remote interface {
+	// Available is an atomic read of a cached probe, so it is free to call per
+	// turn. See llm.Pair.
+	Available() bool
+	// CompleteRemote runs on the workstation or returns ErrRemoteUnavailable. It
+	// never falls back.
+	CompleteRemote(ctx context.Context, r llm.Req) (string, error)
+}
+
+// ErrNoWorldModel — a world question was asked, a workstation model is
+// configured to answer it, and that machine is not answering. The caller turns
+// this into a gap he is told about ("не могу сейчас"), never into an answer from
+// the resident model.
+//
+// This is the naming half of the degradation rule in docs/offload.md. The
+// resident Qwen3-1.7B does not answer a world question worse than the 12B, it
+// invents: measured, the workstation model scores knowledge 9/9 on the talk
+// fixture against the resident model's confabulations
+// (docs/evals/2026-08-02-workstation-gemma4-12b.md).
+var ErrNoWorldModel = errors.New("phraser: no world model available")
+
+// chatTemperature — what the phraser's own transport has always sampled at.
+// Named so the remote path cannot drift from it silently. Whether 0.7 is right
+// at all is Vikunja #402, and answering that here would hide a phrasing change
+// inside a routing change.
+const chatTemperature = 0.7
+
+// UseRemote points the phraser at the workstation model. Wiring time only, once,
+// before anything phrases: the field is read without a lock on every call
+// because a per-turn lock to answer a question that changes at deploy time is
+// not worth paying for.
+//
+// A nil remote is the normal state of a box with no `workstation` block, and it
+// must behave exactly as the box behaved before this seam existed.
+func (p *LLMPhraser) UseRemote(r Remote) {
+	p.remote = r
+}
+
+// PhraseWorld answers a question about the world — either from the model's own
+// knowledge (no sources) or from a passage someone fetched (a live search, a ZIM
+// article, a page he named). Three outcomes, and the middle one is the point:
+//
+//   - No workstation configured. The resident model answers, exactly as it does
+//     today. Naming a gap needs a gap: on a box that never had a second model,
+//     refusing every world question would remove a capability he has now.
+//   - Workstation configured and taking work. It answers.
+//   - Workstation configured and down. ErrNoWorldModel, and the caller says so.
+//
+// The prompts are the ones PhraseQuery uses, built by the same two functions, so
+// the two models are asked the same question in the same words.
+func (p *LLMPhraser) PhraseWorld(ctx context.Context, utterance string, sources []string) (string, error) {
+	sources = nonEmpty(sources)
+	if p.remote == nil {
+		return p.PhraseQuery(ctx, utterance, sources)
+	}
+	var sys, user string
+	if len(sources) == 0 {
+		sys, user = p.knowledgePrompt(utterance)
+	} else {
+		sys, user = p.evidencePrompt(utterance, sources)
+	}
+	if !p.remote.Available() {
+		return "", ErrNoWorldModel
+	}
+	resp, err := p.remote.CompleteRemote(ctx, llm.Req{
+		System:      sys,
+		User:        user,
+		Grammar:     p.grammar(),
+		MaxTokens:   768,
+		Temperature: chatTemperature,
+	})
+	if err != nil {
+		// The cached probe was one interval stale, or the card went away
+		// mid-request. Either way this is the gap, not an error to log and
+		// paper over with the smaller model.
+		log.Printf("phraser: world model: %v", err)
+		return "", errors.Join(ErrNoWorldModel, err)
+	}
+	resp = stripThink(resp)
+	text, _, perr := parseResponseMood(resp)
+	if perr != nil {
+		log.Printf("phraser: PhraseWorld: %v", perr)
+		return "", errors.Join(ErrNoWorldModel, perr)
+	}
+	if text != "" {
+		return text, nil
+	}
+	if resp == "" {
+		return "", ErrNoWorldModel
+	}
+	return resp, nil
+}
+
+// remoteChat is the silent half, for the phrasing paths where the workstation
+// model is only better: a nudge, a reminder, a reply, a question answered from
+// his own notes. It reports whether it answered; it never reports why not,
+// because the caller's next move is the resident model either way.
+//
+// He is not told which of the two models phrased his reply. That is the rule.
+func (p *LLMPhraser) remoteChat(ctx context.Context, system, user string, maxTokens int) (string, bool) {
+	if p.remote == nil || !p.remote.Available() {
+		return "", false
+	}
+	out, err := p.remote.CompleteRemote(ctx, llm.Req{
+		System:      system,
+		User:        user,
+		Grammar:     p.grammar(),
+		MaxTokens:   maxTokens,
+		Temperature: chatTemperature,
+	})
+	if err != nil {
+		log.Printf("phraser: workstation model declined, phrasing here instead: %v", err)
+		return "", false
+	}
+	if out = stripThink(out); out == "" {
+		return "", false
+	}
+	return out, true
+}

internal/phraser/world_test.go

@@ -0,0 +1,164 @@
+package phraser
+
+import (
+	"context"
+	"errors"
+	"strings"
+	"testing"
+
+	"github.com/kami/maven/internal/llm"
+	"github.com/kami/maven/internal/loop"
+)
+
+// fakeRemote — a workstation model that is up or down on command, and records
+// what it was asked.
+type fakeRemote struct {
+	up    bool
+	reply string
+	err   error
+	got   []llm.Req
+}
+
+func (f *fakeRemote) Available() bool { return f.up }
+
+func (f *fakeRemote) CompleteRemote(_ context.Context, r llm.Req) (string, error) {
+	f.got = append(f.got, r)
+	if f.err != nil {
+		return "", f.err
+	}
+	return f.reply, nil
+}
+
+// The three outcomes of the naming half, in one place. The middle one is the
+// whole task: a gap he is told about, not an answer from the smaller model.
+func TestPhraseWorldNamesTheGapOnlyWhenThereIsOne(t *testing.T) {
+	answer := `{"response": "Небо голубое из-за рэлеевского рассеяния.", "mood": "neutral"}`
+
+	t.Run("no workstation configured: the resident model answers as today", func(t *testing.T) {
+		spy := newPromptSpy(t)
+		p := NewLLMPhraserAt(spy.srv.URL, Config{})
+		got, err := p.PhraseWorld(context.Background(), "почему небо голубое", nil)
+		if err != nil {
+			t.Fatalf("PhraseWorld: %v", err)
+		}
+		if got == "" {
+			t.Fatal("no reply from the resident model")
+		}
+		if len(spy.user) != 1 {
+			t.Fatalf("resident model saw %d requests, want 1", len(spy.user))
+		}
+	})
+
+	t.Run("workstation up: it answers and the resident model is not asked", func(t *testing.T) {
+		spy := newPromptSpy(t)
+		p := NewLLMPhraserAt(spy.srv.URL, Config{})
+		remote := &fakeRemote{up: true, reply: answer}
+		p.UseRemote(remote)
+		got, err := p.PhraseWorld(context.Background(), "почему небо голубое", nil)
+		if err != nil {
+			t.Fatalf("PhraseWorld: %v", err)
+		}
+		if !strings.Contains(got, "рассеяния") {
+			t.Errorf("reply is not the workstation's: %q", got)
+		}
+		if len(spy.user) != 0 {
+			t.Errorf("the resident model was asked %d times, want 0", len(spy.user))
+		}
+	})
+
+	t.Run("workstation down: the gap, and nothing invented", func(t *testing.T) {
+		spy := newPromptSpy(t)
+		p := NewLLMPhraserAt(spy.srv.URL, Config{})
+		p.UseRemote(&fakeRemote{up: false})
+		got, err := p.PhraseWorld(context.Background(), "почему небо голубое", nil)
+		if !errors.Is(err, ErrNoWorldModel) {
+			t.Fatalf("err = %v, want ErrNoWorldModel", err)
+		}
+		if got != "" {
+			t.Errorf("got a reply %q with no world model", got)
+		}
+		if len(spy.user) != 0 {
+			t.Errorf("the resident model answered a world question %d times, want 0", len(spy.user))
+		}
+	})
+
+	t.Run("workstation errors mid-request: still the gap", func(t *testing.T) {
+		spy := newPromptSpy(t)
+		p := NewLLMPhraserAt(spy.srv.URL, Config{})
+		p.UseRemote(&fakeRemote{up: true, err: errors.New("connection refused")})
+		if _, err := p.PhraseWorld(context.Background(), "почему небо голубое", nil); !errors.Is(err, ErrNoWorldModel) {
+			t.Fatalf("err = %v, want ErrNoWorldModel", err)
+		}
+		if len(spy.user) != 0 {
+			t.Errorf("the resident model answered a world question %d times, want 0", len(spy.user))
+		}
+	})
+}
+
+// Prompt parity: the workstation model is asked the same question in the same
+// words, or the fixtures measure one thing and the daemon ships another.
+func TestPhraseWorldSendsTheSamePromptsAsPhraseQuery(t *testing.T) {
+	spy := newPromptSpy(t)
+	resident := NewLLMPhraserAt(spy.srv.URL, Config{})
+	if _, err := resident.PhraseQuery(context.Background(), "кто написал войну и мир", []string{"Лев Толстой"}); err != nil {
+		t.Fatal(err)
+	}
+
+	remote := &fakeRemote{up: true, reply: `{"response": "Толстой.", "mood": "neutral"}`}
+	offloaded := NewLLMPhraserAt(spy.srv.URL, Config{})
+	offloaded.UseRemote(remote)
+	if _, err := offloaded.PhraseWorld(context.Background(), "кто написал войну и мир", []string{"Лев Толстой"}); err != nil {
+		t.Fatal(err)
+	}
+
+	if len(remote.got) != 1 {
+		t.Fatalf("the workstation saw %d requests, want 1", len(remote.got))
+	}
+	if remote.got[0].System != spy.system[0] {
+		t.Errorf("system prompts differ:\nremote:   %q\nresident: %q", remote.got[0].System, spy.system[0])
+	}
+	if remote.got[0].User != spy.user[0] {
+		t.Errorf("user prompts differ:\nremote:   %q\nresident: %q", remote.got[0].User, spy.user[0])
+	}
+}
+
+// The silent half. A nudge phrased on the workstation is not news, and one
+// phrased here because the card is busy is not news either — but it must be
+// sampled the same way, or the workstation quietly changes how she sounds.
+func TestNudgePhrasingPrefersTheWorkstationSilently(t *testing.T) {
+	spy := newPromptSpy(t)
+	p := NewLLMPhraserAt(spy.srv.URL, Config{LLMNudges: true})
+	remote := &fakeRemote{up: true, reply: `{"response": "Выпей воды.", "mood": "neutral"}`}
+	p.UseRemote(remote)
+
+	pn, err := p.PhraseNudge(context.Background(), loop.Candidate{Rule: loop.WaterRule(), Severity: loop.Sev1})
+	if err != nil {
+		t.Fatalf("PhraseNudge: %v", err)
+	}
+	if pn.Body != "Выпей воды." {
+		t.Errorf("body = %q, want the workstation's wording", pn.Body)
+	}
+	if len(remote.got) != 1 {
+		t.Fatalf("the workstation saw %d requests, want 1", len(remote.got))
+	}
+	if remote.got[0].Temperature != chatTemperature {
+		t.Errorf("temperature = %v, want %v (what the resident transport samples at)",
+			remote.got[0].Temperature, chatTemperature)
+	}
+	if len(spy.user) != 0 {
+		t.Errorf("the resident model phrased %d nudges, want 0", len(spy.user))
+	}
+}
+
+func TestNudgePhrasingFallsBackWhenTheCardIsBusy(t *testing.T) {
+	spy := newPromptSpy(t)
+	p := NewLLMPhraserAt(spy.srv.URL, Config{LLMNudges: true})
+	p.UseRemote(&fakeRemote{up: false})
+
+	if _, err := p.PhraseNudge(context.Background(), loop.Candidate{Rule: loop.WaterRule(), Severity: loop.Sev1}); err != nil {
+		t.Fatalf("PhraseNudge: %v", err)
+	}
+	if len(spy.user) != 1 {
+		t.Fatalf("the resident model phrased %d nudges, want 1", len(spy.user))
+	}
+}

internal/router/agenda_test.go

@@ -75,3 +75,47 @@ func TestAgendaGrammarSparesStatements(t *testing.T) {
 		}
 	}
 }
+
+// The tomorrow form and the bare event noun. Both were measured answering
+// "пока не умею" on the deployed daemon, 02-08-2026, while the same question
+// about today worked — the first rule set needed "у меня" or a calendar noun
+// and these phrasings carry neither (Vikunja #471).
+func TestAgendaCoversOtherDaysAndNamedEvents(t *testing.T) {
+	r := agendaRouter(t)
+	for _, u := range []string{
+		"какие планы на завтра?",
+		"какие планы на послезавтра",
+		"что по делам в среду",
+		"какие планы на выходные",
+		"когда планёрка?",
+		"во сколько созвон",
+		"когда будет совещание",
+	} {
+		d, err := r.Route(context.Background(), u, refNow())
+		if err != nil {
+			t.Fatalf("route(%q): %v", u, err)
+		}
+		if d.Intent != IntentQuery {
+			t.Errorf("route(%q) = %s, want query", u, d.Intent)
+		}
+	}
+}
+
+// The two new rules are narrow on purpose. A world question that opens with
+// "когда" is not an agenda question, and telling her about a plan is not
+// asking about one.
+func TestAgendaGrammarsLeaveTheWorldAlone(t *testing.T) {
+	r := agendaRouter(t)
+	for _, u := range []string{
+		"когда была битва при ватерлоо",
+		"когда изобрели телефон",
+	} {
+		d, err := r.Route(context.Background(), u, refNow())
+		if err != nil {
+			t.Fatalf("route(%q): %v", u, err)
+		}
+		if d.Stage == 0 {
+			t.Errorf("route(%q) was claimed at stage 0 as %s", u, d.Intent)
+		}
+	}
+}

internal/router/eval/ru_routing_v1.json

@@ -23,6 +23,8 @@
     { "id": "ru-query-012", "utterance": "какие заметки я оставил про полив", "lang": "ru", "intent": "query", "tags": ["recall"] },
     { "id": "ru-query-013", "utterance": "во сколько у меня встреча", "lang": "ru", "intent": "query", "tags": ["calendar"] },
     { "id": "ru-query-019", "utterance": "что у меня стоит в календаре на послезавтра", "lang": "ru", "intent": "query", "tags": ["calendar", "hard"], "note": "agenda, not the clock: the daemon answers this from CalendarEvents inside the query branch, so the clock/date system rule must not swallow it" },
+    { "id": "ru-query-022", "utterance": "какие планы на завтра?", "lang": "ru", "intent": "query", "tags": ["calendar"], "note": "the same agenda question as ru-query-019 aimed at another day; it answered \u043f\u043e\u043a\u0430 \u043d\u0435 \u0443\u043c\u0435\u044e on the deployed daemon while the today form worked (Vikunja #471)" },
+    { "id": "ru-query-023", "utterance": "\u043a\u043e\u0433\u0434\u0430 \u043f\u043b\u0430\u043d\u0451\u0440\u043a\u0430?", "lang": "ru", "intent": "query", "tags": ["calendar", "hard"], "note": "a named event with no calendar word — the noun is the only signal that this is a question about his day" },
     { "id": "ru-query-014", "utterance": "я успеваю до дедлайна", "lang": "ru", "intent": "query", "tags": ["hard", "no-question-word"] },
     { "id": "ru-query-015", "utterance": "сколько я прошёл шагов", "lang": "ru", "intent": "query", "tags": ["aggregate"] },
     { "id": "ru-query-016", "utterance": "покажи давление за неделю", "lang": "ru", "intent": "query", "tags": ["hard", "imperative"], "note": "imperative form but a read — must not route to act" },

internal/router/list.go

@@ -0,0 +1,247 @@
+package router
+
+import (
+	"regexp"
+	"strings"
+)
+
+// Standing lists, matched deterministically (Vikunja #453).
+//
+// Same posture as task capture in task.go and for the same reason: the intent
+// enum is a contract shared with the relabelling prompt, so a list is not an
+// eighth intent. It is a note-shaped or query-shaped utterance carrying an
+// explicit marker, and the marker is a lookup.
+//
+// The markers are deliberately explicit. "молоко закончилось" is an
+// observation about the world and belongs in a note; only an instruction to
+// put something on a list puts it there.
+
+// listStems — the lists he can name, by the stem every case form shares.
+// Russian declines the tag ("список покупок", "в покупки", "в покупках"), so
+// matching a stem is what makes those the same list.
+var listStems = []struct{ stem, list string }{
+	{"покуп", "покупки"},
+	{"продукт", "покупки"},
+	{"магазин", "покупки"},
+	{"аптек", "аптека"},
+	{"хозяйств", "хозяйство"},
+	{"shopping", "покупки"},
+	{"groceries", "покупки"},
+	{"pharmacy", "аптека"},
+}
+
+// listCapturePrefixes — an instruction to add to a list. Longest match wins.
+var listCapturePrefixes = []string{
+	"добавь в список",
+	"добавь в покупки",
+	"добавь к покупкам",
+	"запиши в список",
+	"внеси в список",
+	"положи в список",
+	"в список покупок",
+	"add to the list",
+	"add to my list",
+	"add to the shopping list",
+	"put on the list",
+}
+
+// listQueryPrefixes — an ask to read a list back.
+var listQueryPrefixes = []string{
+	"что в списке",
+	"что в покупках",
+	"что мне купить",
+	"что нужно купить",
+	"что надо купить",
+	"покажи список",
+	"прочитай список",
+	"список покупок",
+	"мой список",
+	"what is on the list",
+	"what's on the list",
+	"read me the list",
+	"show me the list",
+	"shopping list",
+}
+
+// listClearPhrases — the whole list is got. One sentence, one turn.
+var listClearPhrases = []string{
+	"всё купил",
+	"все купил",
+	"всё взял",
+	"все взял",
+	"очисти список",
+	"очисти покупки",
+	"список пустой",
+	"got everything",
+	"clear the list",
+}
+
+// listRemovePrefixes — one item off the list.
+var listRemovePrefixes = []string{
+	"вычеркни",
+	"убери из списка",
+	"убери со списка",
+	"купил",
+	"взял",
+	"cross off",
+	"remove from the list",
+}
+
+// listTrimCut — punctuation and connectives to strip off a parsed remainder.
+const listTrimCut = " .,;:!?—-"
+
+// ListCapture — a parsed list instruction: which list, and the item.
+type ListCapture struct {
+	List string
+	Item string
+}
+
+// ParseListCapture reports whether an utterance puts something on a list, and
+// returns the list tag and the item. A marker with nothing usable after it is
+// not a capture: there is no item in "добавь в список покупок".
+func ParseListCapture(text string) (ListCapture, bool) {
+	rest, ok := afterLongestPrefix(text, listCapturePrefixes)
+	if !ok {
+		return ListCapture{}, false
+	}
+	list, rest := takeListTag(rest)
+	rest = strings.Trim(rest, listTrimCut)
+	if rest == "" {
+		return ListCapture{}, false
+	}
+	return ListCapture{List: list, Item: rest}, true
+}
+
+// ParseListQuery reports whether an utterance asks for a list, and which one.
+func ParseListQuery(text string) (string, bool) {
+	rest, ok := afterLongestPrefix(text, listQueryPrefixes)
+	if !ok {
+		return "", false
+	}
+	list, _ := takeListTag(rest)
+	return list, true
+}
+
+// ParseListClear reports whether an utterance crosses off a whole list.
+func ParseListClear(text string) (string, bool) {
+	lower := strings.ToLower(strings.Trim(strings.TrimSpace(text), listTrimCut))
+	for _, p := range listClearPhrases {
+		if lower == p || strings.HasPrefix(lower, p+" ") {
+			list, _ := takeListTag(strings.TrimSpace(lower[len(p):]))
+			return list, true
+		}
+	}
+	return "", false
+}
+
+// ParseListRemove reports whether an utterance takes one named item off a
+// list, and returns the list and the item.
+//
+// The item is required. "купил" on its own is him reporting he shopped, which
+// ParseListClear reads first, and it must not fall through to here and remove
+// nothing while sounding like it did.
+func ParseListRemove(text string) (ListCapture, bool) {
+	rest, ok := afterLongestPrefix(text, listRemovePrefixes)
+	if !ok {
+		return ListCapture{}, false
+	}
+	list, rest := takeListTag(rest)
+	rest = strings.Trim(rest, listTrimCut)
+	for _, lead := range []string{"из списка ", "со списка ", "из ", "from the list "} {
+		rest = strings.TrimPrefix(rest, lead)
+	}
+	rest = strings.Trim(rest, listTrimCut)
+	if rest == "" {
+		return ListCapture{}, false
+	}
+	return ListCapture{List: list, Item: rest}, true
+}
+
+// afterLongestPrefix matches the longest prefix in the table and returns what
+// follows it, trimmed. Lowercasing does not change the byte length of Russian
+// or English letters, so the index carries over to the original text.
+func afterLongestPrefix(text string, prefixes []string) (string, bool) {
+	trimmed := strings.TrimSpace(text)
+	lower := strings.ToLower(trimmed)
+	best := ""
+	for _, p := range prefixes {
+		if strings.HasPrefix(lower, p) && len(p) > len(best) {
+			best = p
+		}
+	}
+	if best == "" {
+		return "", false
+	}
+	return strings.Trim(trimmed[len(best):], listTrimCut), true
+}
+
+// takeListTag reads a list name off the front of the remainder and returns the
+// list plus what is left. A remainder naming no list is the default list, and
+// nothing is consumed — "добавь в список молоко" names no list and the item is
+// молоко.
+func takeListTag(rest string) (string, string) {
+	fields := strings.Fields(rest)
+	if len(fields) == 0 {
+		return "покупки", ""
+	}
+	head := strings.ToLower(strings.Trim(fields[0], listTrimCut))
+	// "в список покупок" leaves "покупок"; "в списке" leaves nothing.
+	if head == "список" || head == "списке" || head == "списка" || head == "list" {
+		fields = fields[1:]
+		if len(fields) == 0 {
+			return "покупки", ""
+		}
+		head = strings.ToLower(strings.Trim(fields[0], listTrimCut))
+	}
+	for _, s := range listStems {
+		if strings.HasPrefix(head, s.stem) {
+			return s.list, strings.Join(fields[1:], " ")
+		}
+	}
+	return "покупки", strings.Join(fields, " ")
+}
+
+// ListGrammars — stage 0 for the list (Vikunja #453).
+//
+// Both patterns match everything and the Build functions are the real filter,
+// the shape the wake-word act grammar already uses: the parsers above are the
+// definition of a list utterance and duplicating them as regexps would give
+// two answers to one question.
+//
+// Why stage 0 at all: an add and a read-back are deterministic and cheap, and
+// leaving them to the model means "добавь в список покупок молоко" lands as an
+// act or a fact on the turns the model has a bad day. The action handlers still
+// re-parse, so a list turn that arrives by any other route still works.
+func ListGrammars() []Grammar {
+	anything := regexp.MustCompile(`(?s)^(.*)$`)
+	return []Grammar{
+		{
+			Name:    "list-query",
+			Pattern: anything,
+			Build: func(m []string) (Decision, bool) {
+				if _, ok := ParseListQuery(m[1]); !ok {
+					return Decision{}, false
+				}
+				return Decision{Stage: 0, Intent: IntentQuery, Confidence: 1.0}, true
+			},
+		},
+		{
+			Name:    "list-capture",
+			Pattern: anything,
+			Build: func(m []string) (Decision, bool) {
+				text := m[1]
+				_, add := ParseListCapture(text)
+				_, clear := ParseListClear(text)
+				if !add && !clear {
+					return Decision{}, false
+				}
+				return Decision{
+					Stage:      0,
+					Intent:     IntentNote,
+					Confidence: 1.0,
+					Slots:      Slots{Text: strings.TrimSpace(text)},
+				}, true
+			},
+		},
+	}
+}

internal/router/list_test.go

@@ -0,0 +1,89 @@
+package router
+
+import "testing"
+
+func TestParseListCaptureReadsListAndItem(t *testing.T) {
+	cases := []struct {
+		utterance string
+		list      string
+		item      string
+	}{
+		{"добавь в список покупок молоко", "покупки", "молоко"},
+		{"добавь в список молоко", "покупки", "молоко"},
+		{"Добавь в покупки хлеб и яйца", "покупки", "хлеб и яйца"},
+		{"запиши в список аптеки бинт", "аптека", "бинт"},
+		{"добавь в список хозяйства лампочки.", "хозяйство", "лампочки"},
+		{"add to the shopping list milk", "покупки", "milk"},
+	}
+	for _, c := range cases {
+		got, ok := ParseListCapture(c.utterance)
+		if !ok {
+			t.Errorf("ParseListCapture(%q) did not claim it", c.utterance)
+			continue
+		}
+		if got.List != c.list || got.Item != c.item {
+			t.Errorf("ParseListCapture(%q) = %+v; want list %q item %q", c.utterance, got, c.list, c.item)
+		}
+	}
+}
+
+// A marker with no item is not a capture, and an utterance that only mentions
+// shopping is not one either.
+func TestParseListCapturePasses(t *testing.T) {
+	for _, u := range []string{
+		"добавь в список покупок",
+		"добавь в список",
+		"молоко закончилось",
+		"надо бы съездить в магазин",
+		"добавь в задачи купить молоко",
+	} {
+		if got, ok := ParseListCapture(u); ok {
+			t.Errorf("ParseListCapture(%q) claimed it as %+v", u, got)
+		}
+	}
+}
+
+func TestParseListQueryNamesTheList(t *testing.T) {
+	cases := []struct{ utterance, list string }{
+		{"что в списке покупок?", "покупки"},
+		{"что в списке", "покупки"},
+		{"что мне купить", "покупки"},
+		{"покажи список аптеки", "аптека"},
+		{"what's on the list", "покупки"},
+	}
+	for _, c := range cases {
+		list, ok := ParseListQuery(c.utterance)
+		if !ok {
+			t.Errorf("ParseListQuery(%q) did not claim it", c.utterance)
+			continue
+		}
+		if list != c.list {
+			t.Errorf("ParseListQuery(%q) = %q; want %q", c.utterance, list, c.list)
+		}
+	}
+	if _, ok := ParseListQuery("какие у меня задачи"); ok {
+		t.Error("ParseListQuery claimed a task question")
+	}
+}
+
+func TestParseListClearAndRemove(t *testing.T) {
+	if list, ok := ParseListClear("всё купил"); !ok || list != "покупки" {
+		t.Errorf("ParseListClear = %q, %v; want покупки, true", list, ok)
+	}
+	if list, ok := ParseListClear("очисти список аптеки"); !ok || list != "аптека" {
+		t.Errorf("ParseListClear = %q, %v; want аптека, true", list, ok)
+	}
+	if _, ok := ParseListClear("купил молоко"); ok {
+		t.Error("ParseListClear claimed a single item")
+	}
+	got, ok := ParseListRemove("вычеркни молоко")
+	if !ok || got.Item != "молоко" || got.List != "покупки" {
+		t.Errorf("ParseListRemove = %+v, %v; want молоко on покупки", got, ok)
+	}
+	if got, ok := ParseListRemove("убери из списка аптеки бинт"); !ok || got.Item != "бинт" || got.List != "аптека" {
+		t.Errorf("ParseListRemove = %+v, %v; want бинт on аптека", got, ok)
+	}
+	if _, ok := ParseListRemove("вычеркни"); ok {
+		t.Error("ParseListRemove claimed a marker with no item")
+	}
+}

internal/router/llmrouter.go

@@ -210,7 +210,11 @@ func (lr *LLMRouter) Route(ctx context.Context, utterance string, now time.Time)
 		d.Slots.HasKey = a.Key != ""
 	case IntentReminder:
 		d.Intent = IntentReminder
-		d.Slots.Text = firstNonEmpty(a.Text, utterance)
+		// No utterance fallback here, unlike every other intent below. The
+		// model returning no text for a reminder means it found no subject,
+		// and "напомни в 11" is not a subject. Leaving Text empty is what
+		// lets the gate turn that into a question (Vikunja #383).
+		d.Slots.Text = a.Text
 	case IntentNote:
 		d.Intent = IntentNote
 		d.Slots.Text = firstNonEmpty(a.Text, utterance)

internal/router/llmrouter_test.go

@@ -356,3 +356,35 @@ func TestRouterLLMFactWithResolvedKeyStaysConfident(t *testing.T) {
 		t.Fatalf("a fact the parser could key must not clarify: %+v", d)
 	}
 }
+
+// A reminder with a time and no subject must come back empty and gated, not
+// backfilled with the raw words. "напомни в 11" carries an hour and nothing to
+// say at that hour; parking the utterance in Text made the request look
+// complete, so the daemon set a reminder that fires saying "напомни в 11"
+// (Vikunja #383).
+func TestLLMReminderWithoutSubjectAsksInsteadOfGuessing(t *testing.T) {
+	r := newLLMTestRouter(t, `{"intent":"reminder"}`)
+	d, err := r.Route(context.Background(), "напомни в 11", refNow())
+	if err != nil {
+		t.Fatalf("route: %v", err)
+	}
+	if d.Slots.Text != "" {
+		t.Fatalf("subject backfilled from the utterance: %q", d.Slots.Text)
+	}
+	if !d.Clarify {
+		t.Fatalf("a subjectless reminder was accepted, confidence %v", d.Confidence)
+	}
+}
+
+// The gate is about the subject, not about reminders in general: one that has
+// both halves still runs without a question.
+func TestLLMReminderWithSubjectIsNotGated(t *testing.T) {
+	r := newLLMTestRouter(t, `{"intent":"reminder","text":"позвонить маме"}`)
+	d, err := r.Route(context.Background(), "напомни в 11 позвонить маме", refNow())
+	if err != nil {
+		t.Fatalf("route: %v", err)
+	}
+	if d.Clarify {
+		t.Fatalf("a complete reminder was sent back as a question: %+v", d.Slots)
+	}
+}

internal/router/question.go

@@ -0,0 +1,64 @@
+package router
+
+import "strings"
+
+// interrogatives — the question words that mark an utterance as asking rather
+// than telling. Tokenized, never substring: "что" inside "чтобы" and "как"
+// inside "какао" are not questions.
+var interrogatives = []string{
+	"что", "чего", "какой", "какая", "какое", "какие", "каких",
+	"кто", "кого", "кому", "чей", "почему", "зачем", "отчего",
+	"где", "куда", "откуда", "когда", "сколько", "как",
+	"what", "who", "whom", "why", "when", "where", "which", "how",
+}
+
+// narrativeRequests — "tell me about X" asks for knowledge Maven does not
+// hold about him. It carries no question mark and no interrogative, which is
+// how "расскажи про битву при Ватерлоо" reached the fact store (#470).
+var narrativeRequests = []string{
+	"расскажи", "объясни", "опиши", "перечисли",
+	"tell", "explain", "describe",
+}
+
+// captureVerbs — an explicit instruction to record something. These win over
+// every test below, because "запиши что я пил воду" contains an interrogative
+// and is still a capture: the word he said is "запиши".
+var captureVerbs = []string{
+	"запиши", "запомни", "отметь", "заметь", "добавь", "сохрани",
+	"note", "remember", "log", "save",
+}
+
+// IsQuestionShaped reports whether text asks for something rather than
+// records it. It is a deterministic offline test over tokens, so it costs
+// nothing and never depends on the model that produced the routing decision.
+//
+// It exists because a mis-routed question used to be persisted as a fact
+// about the owner, with the model's invented answer as the value (#470). The
+// predicate is deliberately blunt: refusing to store a question is cheap and
+// reversible, storing an invented fact about him is neither.
+func IsQuestionShaped(text string) bool {
+	t := strings.TrimSpace(text)
+	if t == "" {
+		return false
+	}
+	toks := planTokens(t)
+	for _, v := range captureVerbs {
+		if hasTok(toks, v) {
+			return false
+		}
+	}
+	if strings.HasSuffix(t, "?") {
+		return true
+	}
+	for _, w := range interrogatives {
+		if hasTok(toks, w) {
+			return true
+		}
+	}
+	for _, w := range narrativeRequests {
+		if hasTok(toks, w) {
+			return true
+		}
+	}
+	return false
+}

internal/router/question_test.go

@@ -0,0 +1,48 @@
+package router
+
+import "testing"
+
+func TestIsQuestionShaped(t *testing.T) {
+	// The seven utterances #470 recorded, plus the captures that must keep
+	// working. A capture misread as a question loses a fact; a question
+	// misread as a capture poisons recall, so the captures are the ones worth
+	// pinning here.
+	cases := []struct {
+		text string
+		want bool
+	}{
+		{"какая последняя версия языка Go?", true},
+		{"что дальше?", true},
+		{"расскажи про битву при Ватерлоо", true},
+		{"почему небо синее?", true},
+		{"какая столица Австралии?", true},
+		{"кто такой Никола Тесла?", true},
+		{"сколько стоит доллар", true},
+		{"who is the premier of Japan", true},
+		{"объясни линии Фраунгофера", true},
+
+		{"запиши что я пил воду", false},
+		{"запомни какая у меня машина", false},
+		{"отметь что я поужинал", false},
+		{"поужинал", false},
+		{"я выпил кофе", false},
+		{"вода", false},
+		{"привет", false},
+		{"", false},
+	}
+	for _, c := range cases {
+		if got := IsQuestionShaped(c.text); got != c.want {
+			t.Errorf("IsQuestionShaped(%q) = %v, want %v", c.text, got, c.want)
+		}
+	}
+}
+
+// Substring matching is what made the day-plan predicates wrong before, and
+// this predicate gates a write, so it gets the same guard.
+func TestIsQuestionShapedIsTokenized(t *testing.T) {
+	for _, text := range []string{"чтобы не забыть, я полил кактус", "какао выпил"} {
+		if IsQuestionShaped(text) {
+			t.Errorf("IsQuestionShaped(%q) = true; a question word inside a longer word is not a question", text)
+		}
+	}
+}

internal/router/router.go

@@ -147,7 +147,15 @@ func (r *Router) fillSlots(ctx context.Context, d *Decision, now time.Time) {
 			d.Slots.Fn, d.Slots.Args, d.Slots.HasFn = fn, args, true
 		}
 	}
-	if d.Slots.Text == "" {
+	// The extractor's Text is the raw utterance, which is the payload for a
+	// note, a query or a chat turn but not for a reminder — there Text is the
+	// subject, what she says at the hour. Backfilling it made Text impossible
+	// to be empty, so StillMissing never reported SlotText and "О чём
+	// напомнить?" was unaskable; the answer to a question she did manage to
+	// ask then overwrote the whole request instead of filling one gap
+	// (Vikunja #383). A reminder with no subject stays empty and is gated
+	// below into a question.
+	if d.Slots.Text == "" && d.Intent != IntentReminder {
 		d.Slots.Text = ex.Text
 	}
 	// Stage stays 1: it says who decided the route, and that was the LLM.
@@ -177,6 +185,12 @@ func (r *Router) gateLLMDecision(d *Decision) {
 	if d.Intent == IntentAct && !d.Slots.HasFn && d.Confidence > llmThinConfidence {
 		d.Confidence = llmThinConfidence
 	}
+	// A reminder with no subject: she knows when but not what to say then.
+	// Setting it anyway fires an empty reminder at the hour, which reads as a
+	// bug to him and cannot be repaired after the fact. Ask (Vikunja #383).
+	if d.Intent == IntentReminder && d.Slots.Text == "" && d.Confidence > llmThinConfidence {
+		d.Confidence = llmThinConfidence
+	}
 	if d.Confidence < r.threshold {
 		d.Clarify = true
 	}

internal/router/stage0.go

@@ -182,9 +182,38 @@ func AgendaQueryGrammars() []Grammar {
 			Pattern: regexp.MustCompile(`(?i)^\s*(что|чего|какие|сколько|во\s+сколько|когда)\s+у\s+меня(\s|[?!.]|$)`),
 			Build:   agendaQueryBuild,
 		},
+		{
+			// A plan noun aimed at a named day, with no possessive to anchor
+			// on: "какие планы на завтра", "что по делам в среду". The rule
+			// above wants "у меня" and this phrasing never has it, so
+			// "какие планы на завтра" answered "пока не умею" while "какие
+			// планы на сегодня" worked (Vikunja #471). The day word is what
+			// makes it an agenda question rather than a topic.
+			Name: "plan-day-query",
+			// Only "план" and "дел". A verb stem like "встреч" would take
+			// "встречаемся в среду", which is him telling her something, not
+			// asking.
+			Pattern: regexp.MustCompile(`(?i)(^|\s)(план|дел)[а-я]*\s+(на|в|во|по)\s+` + dayWordPattern + `(\s|[?!.]|$)`),
+			Build:   agendaQueryBuild,
+		},
+		{
+			// A named event with no calendar word at all: "когда планёрка?",
+			// "во сколько созвон". He is asking when something on his calendar
+			// happens, and the noun is the only signal. Closed list, so "когда
+			// битва при Ватерлоо" is still a world question.
+			Name:    "event-time-query",
+			Pattern: regexp.MustCompile(`(?i)^\s*(когда|во\s+сколько|в\s+котором\s+часу)\s+(будет\s+|у\s+нас\s+)?(планёрк|планерк|встреч|созвон|митинг|совещани|звонок|созвон|приём|прием|интервью|собеседовани|тренировк|урок|занятие|пара)[а-я]*(\s|[?!.]|$)`),
+			Build:   agendaQueryBuild,
+		},
 	}
 }
 
+// dayWordPattern — the day words an agenda question can name. Weekdays appear
+// in the accusative and prepositional forms the questions actually use ("в
+// среду", "на среде"), which is why the stems carry an inflection tail rather
+// than a fixed ending.
+const dayWordPattern = `(сегодня|завтра|послезавтра|выходн[а-я]+|недел[а-я]+|понедельник[а-я]*|вторник[а-я]*|сред[ауые][а-я]*|четверг[а-я]*|пятниц[ауые][а-я]*|суббот[ауые][а-я]*|воскресень[ея][а-я]*)`
+
 // agendaQueryBuild — shared Build for the agenda grammars. Confidence 1.0 on
 // the intent only: the utterance travels intact and the query chain's own
 // matchers decide the rest.

internal/store/facts.go

@@ -6,6 +6,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"log"
 	"strings"
 	"time"
 
@@ -46,6 +47,39 @@ func (s *Store) WriteFact(ctx context.Context, ts time.Time, kind FactKind, key,
 	return id, nil
 }
 
+// FactRecallText is the text a fact is indexed under and read back as (#493).
+//
+// It used to be the utterance that wrote the fact, so recall of ANY
+// voice-tapped fact answered with the sentence he said instead of the value
+// stored: `go_version = 1.20` was indexed as "какая последняя версия языка
+// Go?", and that question is what came back. The poisoned rows made the defect
+// visible; the shape was wrong for legitimate facts too.
+//
+// The key is spoken with its underscores dropped, because a key is written for
+// the store and this string is read out loud.
+func FactRecallText(key, value string) string {
+	spoken := strings.TrimSpace(strings.ReplaceAll(key, "_", " "))
+	v := strings.TrimSpace(DecodeFactValue(value))
+	switch {
+	case v == "":
+		return spoken
+	case spoken == "":
+		return v
+	}
+	return spoken + " — " + v
+}
+
+// DecodeFactValue unwraps a stored value for reading. The column holds raw json
+// when the writer serialized one (SetValue, CorrectValue) and a plain string
+// when it did not (a voice tap), so a reader that wants the text handles both.
+func DecodeFactValue(value string) string {
+	var s string
+	if err := json.Unmarshal([]byte(value), &s); err == nil {
+		return s
+	}
+	return value
+}
+
 // LatestFact returns the latest non-voided fact for key, or ErrNoFact.
 // "Non-voided" = no later row has voids_id pointing at it. We resolve this by
 // taking the newest row whose id is not referenced by any voids_id.
@@ -290,6 +324,20 @@ func (s *Store) CorrectValue(ctx context.Context, key, source string, value any,
 	if err != nil {
 		return 0, fmt.Errorf("last insert id: %w", err)
 	}
+	// The same repair a void needs, for the same reason (#493). A correction
+	// supersedes the value, and the vector still holds the old one, so recall
+	// kept answering with the value he had just corrected. Dropping it costs
+	// the key its recall vector until the fact is tapped again: this layer has
+	// no embedder, and a missing vector loses a question while a stale one
+	// answers it wrongly.
+	//
+	// Best-effort: the corrected row is committed, and a correction that lands
+	// beats one that fails on cleanup.
+	if n, derr := s.VectorMemory().DeletePrefix(ctx, "fact:"+key+":"); derr != nil {
+		log.Printf("store: correct %q: memory vectors survive: %v", key, derr)
+	} else if n > 0 {
+		log.Printf("store: correct %q: dropped %d superseded memory vector(s)", key, n)
+	}
 	return newID, nil
 }
 
@@ -332,6 +380,20 @@ func (s *Store) VoidLatestFact(ctx context.Context, key, source string, ts time.
 	if err != nil {
 		return 0, 0, fmt.Errorf("void: last insert id: %w", err)
 	}
+	// The other half of the repair (#470). A fact reaches recall through a
+	// vector keyed `fact:<key>:<unix>`, holding the utterance that wrote it.
+	// Voiding the row alone left that vector answering questions, so revert
+	// reported success on a box that stayed broken. Deleting every vector for
+	// the key covers the earlier rows too: their values are superseded, and a
+	// superseded value has no business claiming a turn.
+	//
+	// Best-effort by design: the audit trail is already committed, and a fact
+	// that is voided but still recallable is better than a void that failed.
+	if n, derr := s.VectorMemory().DeletePrefix(ctx, "fact:"+key+":"); derr != nil {
+		log.Printf("store: void %q: memory vectors survive: %v", key, derr)
+	} else if n > 0 {
+		log.Printf("store: void %q: dropped %d memory vector(s)", key, n)
+	}
 	return oldID, newID, nil
 }
 

internal/store/factvectors.go

@@ -0,0 +1,177 @@
+package store
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"strconv"
+	"strings"
+	"time"
+)
+
+// metaKeyFactVectorShape names the shape the stored fact vectors were written
+// in. It exists so the repair below runs once per box instead of on every
+// start: the rows it fixes were written by a code path that no longer exists,
+// and once fixed nothing writes that shape again.
+const metaKeyFactVectorShape = "fact_vector_shape"
+
+// factVectorShapeFact is the shape FactRecallText produces. Anything else in
+// the marker (including nothing, which is every box written before #493) means
+// the fact vectors still hold utterances.
+const factVectorShapeFact = "fact-text (#493)"
+
+// FactVectorRepair is what one repair run did, for logging.
+type FactVectorRepair struct {
+	Skipped   bool // marker already matched — nothing to do
+	Rewritten int  // rows re-embedded from the fact they name
+	Dropped   int  // rows deleted: voided, superseded, or naming no fact at all
+	Kept      int  // rows already holding the right text
+	Took      time.Duration
+}
+
+// RepairFactVectors brings the fact rows of memory_vectors in line with the
+// facts they name, and is the operator recovery a poisoned box had no path to
+// (#470 point 4, #493).
+//
+// Three defects put wrong text in that index, and all three are write-path
+// fixes that do nothing for rows already stored:
+//
+//   - the indexed text was the utterance, so every fact row reads back a
+//     sentence rather than a value;
+//   - a void left its vector behind, so retracted junk kept answering;
+//   - a correction left its vector behind, so the superseded value did.
+//
+// So each fact row is resolved against the fact store and one of three things
+// happens. It is dropped when the key has no fact, when the newest row for the
+// key is a void marker, or when a newer vector for the same key exists — a
+// superseded value has no business claiming a turn. It is re-embedded when its
+// text is not what FactRecallText says the fact is. Otherwise it is left alone.
+//
+// Idempotent, and safe to interrupt: every step compares before writing and the
+// marker is written last, so a run that dies partway is simply redone.
+func (s *Store) RepairFactVectors(ctx context.Context, embed EmbedFunc) (FactVectorRepair, error) {
+	start := time.Now()
+	var res FactVectorRepair
+
+	shape, err := s.Meta(ctx, metaKeyFactVectorShape)
+	if err != nil {
+		return res, err
+	}
+	if shape == factVectorShapeFact {
+		res.Skipped = true
+		res.Took = time.Since(start)
+		return res, nil
+	}
+
+	rows, err := s.db.QueryContext(ctx, `SELECT id, meta FROM memory_vectors`)
+	if err != nil {
+		return res, fmt.Errorf("repair fact vectors: read: %w", err)
+	}
+	type factVec struct {
+		id, key string
+		meta    map[string]string
+		ts      int64
+	}
+	var vecs []factVec
+	newest := map[string]int64{} // key → newest ts seen for it
+	for rows.Next() {
+		var id, metaJSON string
+		if err := rows.Scan(&id, &metaJSON); err != nil {
+			rows.Close()
+			return res, fmt.Errorf("repair fact vectors: row: %w", err)
+		}
+		meta := map[string]string{}
+		if err := json.Unmarshal([]byte(metaJSON), &meta); err != nil {
+			rows.Close()
+			return res, fmt.Errorf("repair fact vectors: meta for %q: %w", id, err)
+		}
+		if meta["type"] != "fact" {
+			continue
+		}
+		key, ts, ok := splitFactVectorID(id)
+		if !ok {
+			continue
+		}
+		vecs = append(vecs, factVec{id: id, key: key, meta: meta, ts: ts})
+		if ts > newest[key] {
+			newest[key] = ts
+		}
+	}
+	rows.Close()
+	if err := rows.Err(); err != nil {
+		return res, fmt.Errorf("repair fact vectors: rows: %w", err)
+	}
+
+	for _, v := range vecs {
+		drop := v.ts < newest[v.key]
+		var want string
+		if !drop {
+			f, ferr := s.LatestFact(ctx, v.key)
+			switch {
+			case errors.Is(ferr, ErrNoFact):
+				drop = true
+			case ferr != nil:
+				return res, fmt.Errorf("repair fact vectors: fact %q: %w", v.key, ferr)
+			case DecodeFactValue(f.Value) == "voided":
+				drop = true
+			default:
+				want = FactRecallText(v.key, f.Value)
+			}
+		}
+		if drop {
+			if err := s.VectorMemory().Delete(ctx, v.id); err != nil {
+				return res, err
+			}
+			res.Dropped++
+			continue
+		}
+		if v.meta["text"] == want {
+			res.Kept++
+			continue
+		}
+		vec, err := embed(ctx, want)
+		if err != nil {
+			return res, fmt.Errorf("repair fact vectors: embed %q: %w", v.id, err)
+		}
+		// The whole meta blob is rewritten in Go rather than patched in SQL,
+		// because json_set needs the JSON1 extension and this store is opened
+		// through sqlcipher.
+		v.meta["text"] = want
+		metaJSON, err := json.Marshal(v.meta)
+		if err != nil {
+			return res, fmt.Errorf("repair fact vectors: meta %q: %w", v.id, err)
+		}
+		if _, err := s.db.ExecContext(ctx,
+			`UPDATE memory_vectors SET vec = ?, meta = ? WHERE id = ?`,
+			encodeVec(vec), string(metaJSON), v.id); err != nil {
+			return res, fmt.Errorf("repair fact vectors: write %q: %w", v.id, err)
+		}
+		res.Rewritten++
+	}
+
+	if err := s.SetMeta(ctx, metaKeyFactVectorShape, factVectorShapeFact); err != nil {
+		return res, err
+	}
+	res.Took = time.Since(start)
+	return res, nil
+}
+
+// splitFactVectorID reads the key and write time back out of a fact vector's
+// id, which the write path builds as `fact:<key>:<unix>`. A key may hold a
+// colon, the timestamp may not, so the split is from the right.
+func splitFactVectorID(id string) (key string, ts int64, ok bool) {
+	rest, found := strings.CutPrefix(id, "fact:")
+	if !found {
+		return "", 0, false
+	}
+	cut := strings.LastIndex(rest, ":")
+	if cut <= 0 {
+		return "", 0, false
+	}
+	ts, err := strconv.ParseInt(rest[cut+1:], 10, 64)
+	if err != nil {
+		return "", 0, false
+	}
+	return rest[:cut], ts, true
+}

internal/store/factvectors_test.go

@@ -0,0 +1,151 @@
+package store
+
+import (
+	"context"
+	"database/sql"
+	"testing"
+	"time"
+)
+
+// The write-path half of #493: recall of a fact must read back the fact, not
+// the sentence he happened to say.
+func TestFactRecallText(t *testing.T) {
+	for _, tc := range []struct {
+		name, key, value, want string
+	}{
+		{"json value", "go_version", `"1.20"`, "go version — 1.20"},
+		{"plain value", "water", "выпил", "water — выпил"},
+		{"no value", "shower", "", "shower"},
+		{"underscores are spoken as spaces", "espresso_machine", `"чистая"`, "espresso machine — чистая"},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			if got := FactRecallText(tc.key, tc.value); got != tc.want {
+				t.Fatalf("FactRecallText(%q, %q) = %q; want %q", tc.key, tc.value, got, tc.want)
+			}
+		})
+	}
+}
+
+// A correction left the superseded value in the index, so recall answered with
+// the value he had just corrected (#493).
+func TestCorrectValueDropsMemoryVectors(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+	now := time.Now()
+	mem := s.VectorMemory()
+
+	if _, err := s.WriteFact(ctx, now, KindSelf, "go_version", `"1.20"`, "tap:voice", 1.0, sql.NullInt64{}); err != nil {
+		t.Fatalf("WriteFact: %v", err)
+	}
+	if err := mem.Insert(ctx, "fact:go_version:1", []float32{1, 0, 0}, map[string]string{
+		"type": "fact", "text": "go version — 1.20",
+	}); err != nil {
+		t.Fatalf("Insert: %v", err)
+	}
+	if _, err := s.CorrectValue(ctx, "go_version", "feedback", "1.25", now.Add(time.Minute)); err != nil {
+		t.Fatalf("CorrectValue: %v", err)
+	}
+	got, err := mem.ByPrefix(ctx, "fact:")
+	if err != nil {
+		t.Fatalf("ByPrefix: %v", err)
+	}
+	if len(got) != 0 {
+		t.Fatalf("after the correction the index still holds %+v; the superseded value must not answer", got)
+	}
+}
+
+// The recovery path a poisoned box had none of (#470 point 4, #493): rows
+// written before the fix hold utterances, voided junk and superseded values,
+// and no write-path change reaches any of them.
+func TestRepairFactVectors(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+	now := time.Now()
+	mem := s.VectorMemory()
+	embed := func(ctx context.Context, text string) ([]float32, error) {
+		return []float32{float32(len(text)), 1, 0}, nil
+	}
+
+	// A live fact indexed under the question that wrote it — the defect.
+	if _, err := s.WriteFact(ctx, now, KindSelf, "water", `"выпил"`, "tap:voice", 1.0, sql.NullInt64{}); err != nil {
+		t.Fatalf("WriteFact water: %v", err)
+	}
+	if err := mem.Insert(ctx, "fact:water:100", []float32{9, 9, 9}, map[string]string{
+		"type": "fact", "source": "voice", "text": "запиши что я пил воду",
+	}); err != nil {
+		t.Fatalf("Insert water: %v", err)
+	}
+	// A voided fact whose vector survived the void.
+	if _, err := s.WriteFact(ctx, now, KindSelf, "go_version", `"1.20"`, "tap:voice", 1.0, sql.NullInt64{}); err != nil {
+		t.Fatalf("WriteFact go_version: %v", err)
+	}
+	if _, _, err := s.VoidLatestFact(ctx, "go_version", "feedback", now.Add(time.Minute)); err != nil {
+		t.Fatalf("VoidLatestFact: %v", err)
+	}
+	if err := mem.Insert(ctx, "fact:go_version:100", []float32{9, 9, 9}, map[string]string{
+		"type": "fact", "text": "какая последняя версия языка Go?",
+	}); err != nil {
+		t.Fatalf("Insert go_version: %v", err)
+	}
+	// A key with two vectors: only the newest may answer.
+	if _, err := s.WriteFact(ctx, now, KindSelf, "mood", `"устал"`, "tap:voice", 1.0, sql.NullInt64{}); err != nil {
+		t.Fatalf("WriteFact mood: %v", err)
+	}
+	for _, ts := range []string{"100", "200"} {
+		if err := mem.Insert(ctx, "fact:mood:"+ts, []float32{9, 9, 9}, map[string]string{
+			"type": "fact", "text": "мне грустно",
+		}); err != nil {
+			t.Fatalf("Insert mood %s: %v", ts, err)
+		}
+	}
+	// A note must be left entirely alone.
+	if err := mem.Insert(ctx, "note:7", []float32{5, 5, 5}, map[string]string{
+		"type": "note", "text": "сеть тормозит по вечерам",
+	}); err != nil {
+		t.Fatalf("Insert note: %v", err)
+	}
+
+	res, err := s.RepairFactVectors(ctx, embed)
+	if err != nil {
+		t.Fatalf("RepairFactVectors: %v", err)
+	}
+	if res.Rewritten != 2 || res.Dropped != 2 {
+		t.Fatalf("repair reported %+v; want 2 rewritten (water, newest mood) and 2 dropped (voided go_version, superseded mood)", res)
+	}
+
+	got, err := mem.ByPrefix(ctx, "fact:")
+	if err != nil {
+		t.Fatalf("ByPrefix: %v", err)
+	}
+	texts := map[string]string{}
+	for _, r := range got {
+		texts[r.ID] = r.Meta["text"]
+	}
+	if len(texts) != 2 {
+		t.Fatalf("the index holds %+v; want only fact:water:100 and fact:mood:200", texts)
+	}
+	if texts["fact:water:100"] != "water — выпил" {
+		t.Fatalf("water reads back %q; want the fact, not the utterance", texts["fact:water:100"])
+	}
+	if texts["fact:mood:200"] != "mood — устал" {
+		t.Fatalf("mood reads back %q", texts["fact:mood:200"])
+	}
+	// Provenance the row already carried must survive the rewrite.
+	for _, r := range got {
+		if r.ID == "fact:water:100" && r.Meta["source"] != "voice" {
+			t.Fatalf("water lost its source meta: %+v", r.Meta)
+		}
+	}
+	if notes, err := mem.ByPrefix(ctx, "note:"); err != nil || len(notes) != 1 {
+		t.Fatalf("the note row was touched: %+v (err %v)", notes, err)
+	}
+
+	// Marker written, so a second run is free and changes nothing.
+	again, err := s.RepairFactVectors(ctx, embed)
+	if err != nil {
+		t.Fatalf("second RepairFactVectors: %v", err)
+	}
+	if !again.Skipped {
+		t.Fatalf("second run did work: %+v; the marker must make it a no-op", again)
+	}
+}

internal/store/listitems.go

@@ -0,0 +1,194 @@
+package store
+
+import (
+	"context"
+	"database/sql"
+	"errors"
+	"fmt"
+	"strings"
+	"time"
+)
+
+// List items — the fourth append-only shape (Vikunja #453).
+//
+// A list is a standing set of short strings under a tag: покупки, аптека,
+// хозяйство. It is not work and it is not a claim about the world, which is
+// why it is neither a task nor a fact. Nothing here is prioritised, nothing
+// nudges about it, and the digestion worker does not read it. The only two
+// things a list does are grow and shrink.
+//
+// The consequence that made it worth a table: because no predicate touches a
+// list item, several people adding to the same list at once cost nothing. There
+// is no ranking to disagree about and no lifecycle beyond crossed-off.
+const (
+	// ListItemOpen — on the list.
+	ListItemOpen = "open"
+	// ListItemDone — bought, taken, crossed off.
+	ListItemDone = "done"
+	// ListItemDropped — removed without being got.
+	ListItemDropped = "dropped"
+)
+
+// DefaultList — the list a capture lands on when he names none. Almost every
+// spoken list item is groceries, and asking "в какой список?" for the common
+// case would be a nag.
+const DefaultList = "покупки"
+
+// ListItem — one line on one list.
+type ListItem struct {
+	ID         int64
+	CreatedTs  time.Time
+	List       string
+	Item       string
+	Source     string
+	Status     string
+	ResolvedTs *time.Time
+}
+
+var (
+	ErrListItemNotFound = errors.New("store: list item not found")
+	ErrListItemEmpty    = errors.New("store: list item is empty")
+	ErrListItemStatus   = errors.New("store: invalid list item status")
+)
+
+// NormalizeListName folds a list tag to its dedupe form. Lists are named out
+// loud, so "Покупки" and "покупки " are the same list.
+func NormalizeListName(s string) string {
+	n := NormalizeTaskText(s)
+	if n == "" {
+		return DefaultList
+	}
+	return n
+}
+
+// AddListItem puts an item on a list, or returns the existing row when the same
+// item is already on it. Created says which happened, so the caller can say
+// "уже есть" instead of pretending it wrote something.
+func (s *Store) AddListItem(ctx context.Context, li ListItem) (CaptureResult, error) {
+	item := strings.TrimSpace(li.Item)
+	if item == "" {
+		return CaptureResult{}, ErrListItemEmpty
+	}
+	list := NormalizeListName(li.List)
+	norm := NormalizeTaskText(item)
+	created := li.CreatedTs
+	if created.IsZero() {
+		created = time.Now()
+	}
+	res, err := s.db.ExecContext(ctx,
+		`INSERT INTO list_items (created_ts, list, item, norm, source, status)
+		 VALUES (?,?,?,?,?,?)
+		 ON CONFLICT DO NOTHING`,
+		created.UnixMilli(), list, item, norm, li.Source, ListItemOpen)
+	if err != nil {
+		return CaptureResult{}, fmt.Errorf("add list item: %w", err)
+	}
+	n, err := res.RowsAffected()
+	if err != nil {
+		return CaptureResult{}, fmt.Errorf("add list item: rows affected: %w", err)
+	}
+	if n > 0 {
+		id, err := res.LastInsertId()
+		if err != nil {
+			return CaptureResult{}, fmt.Errorf("add list item: last insert id: %w", err)
+		}
+		return CaptureResult{ID: id, Created: true}, nil
+	}
+	var id int64
+	err = s.db.QueryRowContext(ctx,
+		`SELECT id FROM list_items WHERE list = ? AND norm = ? AND status = ?`,
+		list, norm, ListItemOpen).Scan(&id)
+	if errors.Is(err, sql.ErrNoRows) {
+		return CaptureResult{}, ErrListItemNotFound
+	}
+	if err != nil {
+		return CaptureResult{}, fmt.Errorf("add list item: lookup: %w", err)
+	}
+	return CaptureResult{ID: id}, nil
+}
+
+// ListItems reads one list in the order it was added. An empty status reads the
+// open items, which is what reading the list aloud means.
+func (s *Store) ListItems(ctx context.Context, list, status string) ([]ListItem, error) {
+	if status == "" {
+		status = ListItemOpen
+	}
+	rows, err := s.db.QueryContext(ctx,
+		`SELECT id, created_ts, list, item, source, status, resolved_ts
+		   FROM list_items WHERE list = ? AND status = ?
+		  ORDER BY created_ts, id`,
+		NormalizeListName(list), status)
+	if err != nil {
+		return nil, fmt.Errorf("list items: %w", err)
+	}
+	defer rows.Close()
+	var out []ListItem
+	for rows.Next() {
+		var (
+			li       ListItem
+			created  int64
+			resolved sql.NullInt64
+		)
+		if err := rows.Scan(&li.ID, &created, &li.List, &li.Item, &li.Source, &li.Status, &resolved); err != nil {
+			return nil, fmt.Errorf("list items: scan: %w", err)
+		}
+		li.CreatedTs = time.UnixMilli(created)
+		if resolved.Valid {
+			t := time.UnixMilli(resolved.Int64)
+			li.ResolvedTs = &t
+		}
+		out = append(out, li)
+	}
+	if err := rows.Err(); err != nil {
+		return nil, fmt.Errorf("list items: %w", err)
+	}
+	return out, nil
+}
+
+// SetListItemStatus crosses an item off, or removes it. Moving an item that is
+// already resolved is not an error — crossing off twice is the same list.
+func (s *Store) SetListItemStatus(ctx context.Context, id int64, status string, at time.Time) error {
+	if status != ListItemOpen && status != ListItemDone && status != ListItemDropped {
+		return fmt.Errorf("%w: %q", ErrListItemStatus, status)
+	}
+	var resolved sql.NullInt64
+	if status != ListItemOpen {
+		if at.IsZero() {
+			at = time.Now()
+		}
+		resolved = sql.NullInt64{Int64: at.UnixMilli(), Valid: true}
+	}
+	res, err := s.db.ExecContext(ctx,
+		`UPDATE list_items SET status = ?, resolved_ts = ? WHERE id = ?`,
+		status, resolved, id)
+	if err != nil {
+		return fmt.Errorf("set list item status: %w", err)
+	}
+	n, err := res.RowsAffected()
+	if err != nil {
+		return fmt.Errorf("set list item status: rows affected: %w", err)
+	}
+	if n == 0 {
+		return ErrListItemNotFound
+	}
+	return nil
+}
+
+// ClearList crosses off every open item on a list and reports how many. This is
+// "всё купил", which is one sentence and must not become one turn per item.
+func (s *Store) ClearList(ctx context.Context, list string, at time.Time) (int, error) {
+	if at.IsZero() {
+		at = time.Now()
+	}
+	res, err := s.db.ExecContext(ctx,
+		`UPDATE list_items SET status = ?, resolved_ts = ? WHERE list = ? AND status = ?`,
+		ListItemDone, at.UnixMilli(), NormalizeListName(list), ListItemOpen)
+	if err != nil {
+		return 0, fmt.Errorf("clear list: %w", err)
+	}
+	n, err := res.RowsAffected()
+	if err != nil {
+		return 0, fmt.Errorf("clear list: rows affected: %w", err)
+	}
+	return int(n), nil
+}

internal/store/listitems_test.go

@@ -0,0 +1,149 @@
+package store
+
+import (
+	"context"
+	"errors"
+	"testing"
+	"time"
+)
+
+var listNow = time.Date(2026, 8, 4, 12, 0, 0, 0, time.UTC)
+
+func TestAddListItemDedupesTheOpenList(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	first, err := s.AddListItem(ctx, ListItem{Item: "молоко", Source: "tap:voice", CreatedTs: listNow})
+	if err != nil {
+		t.Fatalf("add: %v", err)
+	}
+	if !first.Created {
+		t.Fatal("the first молоко did not create a row")
+	}
+	again, err := s.AddListItem(ctx, ListItem{Item: " Молоко ", Source: "tap:voice", CreatedTs: listNow})
+	if err != nil {
+		t.Fatalf("add again: %v", err)
+	}
+	if again.Created {
+		t.Error("молоко was added twice")
+	}
+	if again.ID != first.ID {
+		t.Errorf("second add points at %d; want the existing %d", again.ID, first.ID)
+	}
+	if _, err := s.AddListItem(ctx, ListItem{Item: "  "}); !errors.Is(err, ErrListItemEmpty) {
+		t.Errorf("empty item: %v; want ErrListItemEmpty", err)
+	}
+}
+
+// A crossed-off item does not block the next one: buying milk again next week
+// is a new line, the way saying an errand again is a new task.
+func TestCrossedOffItemComesBack(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	first, err := s.AddListItem(ctx, ListItem{Item: "молоко", CreatedTs: listNow})
+	if err != nil {
+		t.Fatalf("add: %v", err)
+	}
+	if err := s.SetListItemStatus(ctx, first.ID, ListItemDone, listNow); err != nil {
+		t.Fatalf("cross off: %v", err)
+	}
+	next, err := s.AddListItem(ctx, ListItem{Item: "молоко", CreatedTs: listNow.Add(time.Hour)})
+	if err != nil {
+		t.Fatalf("add after: %v", err)
+	}
+	if !next.Created || next.ID == first.ID {
+		t.Errorf("second молоко reused row %d; want a new one", next.ID)
+	}
+	open, err := s.ListItems(ctx, "", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(open) != 1 || open[0].ID != next.ID {
+		t.Errorf("open list %+v; want only the new row", open)
+	}
+}
+
+// Lists are separate stores under one table: the same word on two lists is two
+// items, and reading one never reads the other.
+func TestListsDoNotSeeEachOther(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	if _, err := s.AddListItem(ctx, ListItem{List: "покупки", Item: "вода", CreatedTs: listNow}); err != nil {
+		t.Fatalf("add: %v", err)
+	}
+	if _, err := s.AddListItem(ctx, ListItem{List: "Аптека", Item: "вода", CreatedTs: listNow}); err != nil {
+		t.Fatalf("add: %v", err)
+	}
+	for _, c := range []struct{ list, want string }{
+		{"покупки", "покупки"},
+		{"аптека", "аптека"},
+		{"", "покупки"},
+	} {
+		got, err := s.ListItems(ctx, c.list, "")
+		if err != nil {
+			t.Fatalf("list %q: %v", c.list, err)
+		}
+		if len(got) != 1 {
+			t.Fatalf("list %q has %d items; want 1", c.list, len(got))
+		}
+		if got[0].List != c.want {
+			t.Errorf("list %q returned tag %q; want %q", c.list, got[0].List, c.want)
+		}
+	}
+}
+
+func TestClearListCrossesOffEverythingOpen(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	for _, item := range []string{"молоко", "хлеб", "яйца"} {
+		if _, err := s.AddListItem(ctx, ListItem{Item: item, CreatedTs: listNow}); err != nil {
+			t.Fatalf("add %s: %v", item, err)
+		}
+	}
+	if _, err := s.AddListItem(ctx, ListItem{List: "аптека", Item: "бинт", CreatedTs: listNow}); err != nil {
+		t.Fatalf("add: %v", err)
+	}
+	n, err := s.ClearList(ctx, "покупки", listNow)
+	if err != nil {
+		t.Fatalf("clear: %v", err)
+	}
+	if n != 3 {
+		t.Errorf("cleared %d; want 3", n)
+	}
+	left, err := s.ListItems(ctx, "покупки", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(left) != 0 {
+		t.Errorf("%d items still open; want none", len(left))
+	}
+	done, err := s.ListItems(ctx, "покупки", ListItemDone)
+	if err != nil {
+		t.Fatalf("list done: %v", err)
+	}
+	if len(done) != 3 || done[0].ResolvedTs == nil {
+		t.Errorf("done list %+v; want 3 rows carrying a resolved time", done)
+	}
+	other, err := s.ListItems(ctx, "аптека", "")
+	if err != nil {
+		t.Fatalf("list: %v", err)
+	}
+	if len(other) != 1 {
+		t.Error("clearing покупки touched аптека")
+	}
+}
+
+func TestSetListItemStatusRejectsWhatIsNotAStatus(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	if err := s.SetListItemStatus(ctx, 1, "куплено", listNow); !errors.Is(err, ErrListItemStatus) {
+		t.Errorf("bad status: %v; want ErrListItemStatus", err)
+	}
+	if err := s.SetListItemStatus(ctx, 999, ListItemDone, listNow); !errors.Is(err, ErrListItemNotFound) {
+		t.Errorf("missing row: %v; want ErrListItemNotFound", err)
+	}
+}

internal/store/memory.go

@@ -148,6 +148,27 @@ func (m *MemoryStore) Delete(ctx context.Context, id string) error {
 	return nil
 }
 
+// DeletePrefix removes every vector whose id starts with prefix and returns
+// how many went. Same escaping as ByPrefix, so a key containing % or _ cannot
+// widen the delete.
+//
+// It exists for the repair half of a revert (#470). Voiding a fact row left
+// its vector in the index, so recall kept serving the voided fact's utterance
+// and the documented repair did not repair.
+func (m *MemoryStore) DeletePrefix(ctx context.Context, prefix string) (int64, error) {
+	pattern := escapeLike(prefix) + "%"
+	res, err := m.db.ExecContext(ctx,
+		`DELETE FROM memory_vectors WHERE id LIKE ? ESCAPE '\'`, pattern)
+	if err != nil {
+		return 0, fmt.Errorf("memory: delete prefix %q: %w", prefix, err)
+	}
+	n, err := res.RowsAffected()
+	if err != nil {
+		return 0, fmt.Errorf("memory: delete prefix %q: rows affected: %w", prefix, err)
+	}
+	return n, nil
+}
+
 // escapeLike neutralises the LIKE wildcards in a literal prefix.
 func escapeLike(s string) string {
 	r := strings.NewReplacer(`\`, `\\`, `%`, `\%`, `_`, `\_`)

internal/store/migrations.go

@@ -208,6 +208,38 @@ ALTER TABLE reminders ADD COLUMN next_fire_ts INTEGER;`, // #2
 	// list_tasks into something that writes without the row changing by one
 	// byte. The fingerprint is the declared shape at approval time, so a
 	// redefinition is a re-approval instead of a silent upgrade.
+	`DELETE FROM facts
+	  WHERE key LIKE 'calendar_event_%'
+	    AND replace(substr(key, 25), '-', '') = '';`,
+	// #18 — drop the calendar keys written while safeKey dropped Cyrillic
+	// (Vikunja #443). Everything after the date prefix was punctuation, so
+	// every Russian event on one day shared one key and only the last one
+	// survived. Deleting rather than rewriting: a calendar fact is derived
+	// data, the next poll writes the day again under keys that identify the
+	// event, and the old rows would otherwise be recited as extra meetings.
+	// The filter is exact — it keeps any key whose summary part still has a
+	// letter or a digit in it.
+	`CREATE TABLE IF NOT EXISTS list_items (
+		id          INTEGER PRIMARY KEY AUTOINCREMENT,
+		created_ts  INTEGER NOT NULL,
+		list        TEXT    NOT NULL,
+		item        TEXT    NOT NULL,
+		norm        TEXT    NOT NULL,
+		source      TEXT    NOT NULL,
+		status      TEXT    NOT NULL DEFAULT 'open' CHECK (status IN ('open','done','dropped')),
+		resolved_ts INTEGER
+	);
+	CREATE UNIQUE INDEX IF NOT EXISTS idx_list_items_live ON list_items (list, norm) WHERE status = 'open';
+	CREATE INDEX IF NOT EXISTS idx_list_items_list ON list_items (list, status, created_ts);`,
+	// #19 — standing lists (Vikunja #453). The fourth append-only shape, after
+	// facts, notes and tasks, and the reason it is its own table rather than a
+	// tag on tasks: milk on the shopping list is not work. Nothing prioritises
+	// it, nothing nudges about it, and the prioritiser must not start counting
+	// groceries as outstanding errands.
+	//
+	// The live-only unique index is the tasks one, per list: saying "молоко"
+	// twice before the shop keeps one row, saying it again next week after the
+	// last one was crossed off writes a new one.
 }
 
 // migrate applies every migration with a number greater than the DB's current

internal/store/migrations_test.go

@@ -47,3 +47,36 @@ func TestMigrateAppliesOnceAndIsIdempotent(t *testing.T) {
 		t.Fatalf("after re-migrate user_version = %d, want %d", v, want)
 	}
 }
+
+// Migration #18 clears the calendar keys written while safeKey dropped
+// Cyrillic. Those rows are indistinguishable from real events on read, so
+// leaving them would recite one meeting as several (Vikunja #443).
+func TestCollapsedCalendarKeysAreDropped(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+
+	rows := []string{
+		"calendar_event_20260804_--",             // "Встреча с Аней" under the old rule
+		"calendar_event_20260804_",               // a one-word Russian summary
+		"calendar_event_20260804_Встреча-с-Аней", // the new format
+		"calendar_event_20260804_Standup",        // an ASCII summary, always fine
+	}
+	for _, key := range rows {
+		if _, err := s.db.ExecContext(ctx,
+			`INSERT INTO facts (ts, kind, key, value, source, confidence) VALUES (0, 'env', ?, 'x', 'poll:caldav', 1.0)`,
+			key); err != nil {
+			t.Fatalf("seed %q: %v", key, err)
+		}
+	}
+	if _, err := s.db.ExecContext(ctx, migrations[17]); err != nil {
+		t.Fatalf("migration 18: %v", err)
+	}
+
+	var got int
+	if err := s.db.QueryRowContext(ctx, `SELECT count(*) FROM facts WHERE key LIKE 'calendar_event_%'`).Scan(&got); err != nil {
+		t.Fatal(err)
+	}
+	if got != 2 {
+		t.Fatalf("%d calendar rows left, want the 2 that identify their event", got)
+	}
+}

internal/store/voidvectors_test.go

@@ -0,0 +1,64 @@
+package store
+
+import (
+	"context"
+	"database/sql"
+	"testing"
+	"time"
+)
+
+// Stage 3 of #470: reverting a fact reported success and left the vector that
+// was answering questions, so the documented repair did not repair.
+func TestVoidLatestFactDropsMemoryVectors(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+	now := time.Now()
+	mem := s.VectorMemory()
+
+	if _, err := s.WriteFact(ctx, now, KindSelf, "go_version", `"1.20"`, "tap:voice", 1.0, sql.NullInt64{}); err != nil {
+		t.Fatalf("WriteFact: %v", err)
+	}
+	// The id shape actionFact writes: fact:<key>:<unix>.
+	if err := mem.Insert(ctx, "fact:go_version:1", []float32{1, 0, 0}, map[string]string{
+		"type": "fact", "text": "какая последняя версия языка Go?",
+	}); err != nil {
+		t.Fatalf("Insert: %v", err)
+	}
+	// A vector for another key must survive the void.
+	if err := mem.Insert(ctx, "fact:water:1", []float32{0, 1, 0}, map[string]string{
+		"type": "fact", "text": "запиши что я пил воду",
+	}); err != nil {
+		t.Fatalf("Insert: %v", err)
+	}
+
+	if _, _, err := s.VoidLatestFact(ctx, "go_version", "feedback", now.Add(time.Minute)); err != nil {
+		t.Fatalf("VoidLatestFact: %v", err)
+	}
+
+	got, err := mem.ByPrefix(ctx, "fact:")
+	if err != nil {
+		t.Fatalf("ByPrefix: %v", err)
+	}
+	if len(got) != 1 || got[0].ID != "fact:water:1" {
+		t.Fatalf("after the void the index holds %+v; want only fact:water:1", got)
+	}
+}
+
+func TestDeletePrefixDoesNotWidenOnWildcards(t *testing.T) {
+	ctx := context.Background()
+	s := newTestStore(t)
+	mem := s.VectorMemory()
+
+	for _, id := range []string{"fact:a_b:1", "fact:axb:1"} {
+		if err := mem.Insert(ctx, id, []float32{1, 0}, map[string]string{"type": "fact"}); err != nil {
+			t.Fatalf("Insert %q: %v", id, err)
+		}
+	}
+	n, err := mem.DeletePrefix(ctx, "fact:a_b:")
+	if err != nil {
+		t.Fatalf("DeletePrefix: %v", err)
+	}
+	if n != 1 {
+		t.Fatalf("deleted %d rows; the _ in the key must not match x", n)
+	}
+}

internal/tool/capability.go

@@ -0,0 +1,140 @@
+package tool
+
+import (
+	"sort"
+	"strings"
+
+	"github.com/kami/maven/internal/ipc"
+	"github.com/kami/maven/internal/mcp"
+	"github.com/kami/maven/internal/smarthome"
+)
+
+// Capability ids (Vikunja #452).
+//
+// A tool row is flat: one name, one scope, one enabled bit. Permission is
+// therefore per name, and nothing groups. Hexis has spoken dotted capability
+// ids since it existed, so the local surface was the odd one out — and the
+// flat shape gets expensive around fifteen rows, when "what can she do to the
+// house" stops being a question anyone can answer by reading a list.
+//
+// A capability id is scope.domain.action: homelab.docker.restart,
+// house.lock.unlock, mcp_vikunja.vikunja.delete_task.
+//
+// DERIVED, not stored, for the same reason the risk tier is (risk.go): a
+// derivation is one place to argue with, a column is whatever the last person
+// to enable the row happened to type. The name stays the primary key and
+// nothing about lookup or execution changes — this is a way to READ the
+// allowlist, not a second allowlist.
+type Capability struct {
+	Scope  string
+	Domain string
+	Action string
+}
+
+// String renders the dotted id. An empty segment becomes "unknown" rather than
+// collapsing, so an id always has three parts and a prefix match cannot
+// accidentally widen.
+func (c Capability) String() string {
+	return capSegment(c.Scope) + "." + capSegment(c.Domain) + "." + capSegment(c.Action)
+}
+
+func capSegment(s string) string {
+	s = strings.ToLower(strings.TrimSpace(s))
+	s = strings.ReplaceAll(s, ".", "_")
+	s = strings.ReplaceAll(s, " ", "_")
+	if s == "" {
+		return "unknown"
+	}
+	return s
+}
+
+// CapabilityOf derives the id of a tool row.
+//
+// The domain is the thing acted on and the action is what is done to it, read
+// off whichever dispatch shape the row uses:
+//
+//   - a house row: the Home Assistant entity domain and the service, so
+//     light.kitchen + turn_off becomes house.light.turn_off. Its scope is
+//     "house" whatever the row says, because the entity id is what decides
+//     what it touches.
+//   - an MCP row: the server handle and the remote tool name.
+//   - a process row: the program (path stripped) and its first subcommand, or
+//     the tool name when the argv carries no second word.
+func CapabilityOf(t ipc.Tool) Capability {
+	if entityID, service, ok := smarthome.ParseCmd(t.Cmd); ok {
+		domain := entityID
+		if i := strings.Index(entityID, "."); i > 0 {
+			domain = entityID[:i]
+		}
+		return Capability{Scope: "house", Domain: domain, Action: service}
+	}
+	if server, remote, ok := mcp.ParseCmd(t.Cmd); ok {
+		return Capability{Scope: "mcp_" + server, Domain: server, Action: remote}
+	}
+	scope := t.Scope
+	if scope == "" {
+		scope = "homelab"
+	}
+	if len(t.Cmd) == 0 {
+		// A proposal has no argv yet. It still gets an id, because "what did
+		// she ask for" is exactly the question the proposed list answers.
+		return Capability{Scope: scope, Domain: "unknown", Action: t.Name}
+	}
+	program := t.Cmd[0]
+	if i := strings.LastIndex(program, "/"); i >= 0 {
+		program = program[i+1:]
+	}
+	action := t.Name
+	if len(t.Cmd) > 1 && !strings.HasPrefix(t.Cmd[1], "-") {
+		action = t.Cmd[1]
+	}
+	return Capability{Scope: scope, Domain: program, Action: action}
+}
+
+// MatchCapability reports whether an id matches a pattern. A pattern is a
+// dotted id whose segments may be "*", and a pattern with fewer segments than
+// the id matches every id under it: "house" and "house.*" both cover
+// house.lock.unlock.
+//
+// Prefix widening is deliberate and one-directional. "house.lock" covers every
+// action on the locks; nothing lets a narrower id claim a wider pattern.
+func MatchCapability(pattern string, c Capability) bool {
+	want := strings.Split(strings.ToLower(strings.TrimSpace(pattern)), ".")
+	got := strings.Split(c.String(), ".")
+	if len(want) > len(got) {
+		return false
+	}
+	for i, w := range want {
+		if w == "*" || w == "" {
+			continue
+		}
+		if w != got[i] {
+			return false
+		}
+	}
+	return true
+}
+
+// GroupByDomain buckets rows by "scope.domain" and returns the buckets in a
+// stable order, which is what makes the allowlist readable past the point
+// where a flat list stops being.
+func GroupByDomain(tools []ipc.Tool) []CapabilityGroup {
+	byKey := map[string][]ipc.Tool{}
+	for _, t := range tools {
+		c := CapabilityOf(t)
+		byKey[capSegment(c.Scope)+"."+capSegment(c.Domain)] = append(byKey[capSegment(c.Scope)+"."+capSegment(c.Domain)], t)
+	}
+	out := make([]CapabilityGroup, 0, len(byKey))
+	for k, v := range byKey {
+		sort.Slice(v, func(i, j int) bool { return v[i].Name < v[j].Name })
+		out = append(out, CapabilityGroup{Prefix: k, Tools: v})
+	}
+	sort.Slice(out, func(i, j int) bool { return out[i].Prefix < out[j].Prefix })
+	return out
+}
+
+// CapabilityGroup — one scope.domain and the rows under it.
+type CapabilityGroup struct {
+	Prefix string
+	Tools  []ipc.Tool
+}

internal/tool/capability_test.go

@@ -0,0 +1,92 @@
+package tool
+
+import (
+	"testing"
+
+	"github.com/kami/maven/internal/ipc"
+)
+
+func TestCapabilityOfDescribesTheRow(t *testing.T) {
+	cases := []struct {
+		name string
+		tool ipc.Tool
+		want string
+	}{
+		{
+			"a process with a subcommand",
+			ipc.Tool{Name: "restart", Scope: "homelab", Cmd: []string{"docker", "restart"}},
+			"homelab.docker.restart",
+		},
+		{
+			"a program with a path and a flag",
+			ipc.Tool{Name: "backup", Scope: "homelab", Cmd: []string{"/usr/local/bin/borg", "-v"}},
+			"homelab.borg.backup",
+		},
+		{
+			"the house",
+			ipc.Tool{Name: "unlock_front", Cmd: []string{"smarthome", "lock.front_door", "unlock"}},
+			"house.lock.unlock",
+		},
+		{
+			"an mcp tool",
+			ipc.Tool{Name: "vikunja_delete_task", Cmd: []string{"mcp", "vikunja", "delete_task"}},
+			"mcp_vikunja.vikunja.delete_task",
+		},
+		{
+			"a proposal with no command yet",
+			ipc.Tool{Name: "перезапусти", Scope: "homelab"},
+			"homelab.unknown.перезапусти",
+		},
+	}
+	for _, c := range cases {
+		if got := CapabilityOf(c.tool).String(); got != c.want {
+			t.Errorf("%s: %q; want %q", c.name, got, c.want)
+		}
+	}
+}
+
+// A dotted id always has three segments, so a prefix pattern cannot widen by
+// accident onto a row whose scope happens to be empty.
+func TestCapabilityStringAlwaysHasThreeSegments(t *testing.T) {
+	if got := (Capability{}).String(); got != "unknown.unknown.unknown" {
+		t.Errorf("empty capability = %q", got)
+	}
+	if got := (Capability{Scope: "home lab", Domain: "a.b", Action: "X"}).String(); got != "home_lab.a_b.x" {
+		t.Errorf("segments not folded: %q", got)
+	}
+}
+
+func TestMatchCapabilityWidensOneWay(t *testing.T) {
+	c := CapabilityOf(ipc.Tool{Name: "unlock_front", Cmd: []string{"smarthome", "lock.front_door", "unlock"}})
+	for _, p := range []string{"house", "house.lock", "house.lock.unlock", "house.*.unlock", "*.lock"} {
+		if !MatchCapability(p, c) {
+			t.Errorf("%q did not match %s", p, c)
+		}
+	}
+	for _, p := range []string{"homelab", "house.light", "house.lock.lock", "house.lock.unlock.now"} {
+		if MatchCapability(p, c) {
+			t.Errorf("%q matched %s", p, c)
+		}
+	}
+}
+
+func TestGroupByDomainIsStable(t *testing.T) {
+	tools := []ipc.Tool{
+		{Name: "restart", Scope: "homelab", Cmd: []string{"docker", "restart"}},
+		{Name: "unlock_front", Cmd: []string{"smarthome", "lock.front_door", "unlock"}},
+		{Name: "logs", Scope: "homelab", Cmd: []string{"docker", "logs"}},
+	}
+	groups := GroupByDomain(tools)
+	if len(groups) != 2 {
+		t.Fatalf("%d groups; want 2", len(groups))
+	}
+	if groups[0].Prefix != "homelab.docker" || len(groups[0].Tools) != 2 {
+		t.Errorf("first group %+v; want homelab.docker with 2 rows", groups[0])
+	}
+	if groups[0].Tools[0].Name != "logs" {
+		t.Errorf("rows not sorted: %+v", groups[0].Tools)
+	}
+	if groups[1].Prefix != "house.lock" {
+		t.Errorf("second group %q; want house.lock", groups[1].Prefix)
+	}
+}

internal/tool/risk.go

@@ -0,0 +1,145 @@
+package tool
+
+import (
+	"strings"
+
+	"github.com/kami/maven/internal/ipc"
+	"github.com/kami/maven/internal/mcp"
+	"github.com/kami/maven/internal/smarthome"
+)
+
+// Risk tiers (Vikunja #449).
+//
+// What existed before this file was a mechanism and no policy: one
+// `Destructive` boolean per row, set by whoever ticked the checkbox on /tools.
+// Nothing said which acts are destructive, whether a confirmed act stays
+// confirmed, or what a new tool domain inherits — so every domain answered
+// those questions for itself, and two of them answered differently.
+//
+// The tiers below are the policy. They are derived from the row, not stored:
+// a derivation can be argued with and corrected in one place, while a column
+// is whatever the last person to enable the tool believed.
+//
+// The three questions, answered once:
+//
+//   - WHICH ACTS ARE DESTRUCTIVE. A house row always is, because there is no
+//     read-only way to turn the heating off. A row whose argv names one of the
+//     irreversible verbs always is, whatever the checkbox says. Everything else
+//     is what the row was enabled as.
+//   - DOES A CONFIRMED ACT STAY CONFIRMED. No. Never, at any tier. A
+//     confirmation binds one capability, one target and one argument list, and
+//     it expires with the parked turn (confirmTTL, 90s). "Same act again" is a
+//     new act and costs a new turn. A sticky confirm is a standing grant, and
+//     nothing on the voice path may hold one.
+//   - WHAT A NEW DOMAIN INHERITS. The default is TierDestructive, not
+//     TierSafe. A dispatch shape this file does not recognise gets the confirm
+//     turn — a new domain must argue its way DOWN to running freely, never up
+//     to needing a confirm.
+type Risk string
+
+const (
+	// TierSafe — a read, or a mutation the owner can undo by saying the
+	// opposite. Runs on first hearing.
+	TierSafe Risk = "safe"
+	// TierDestructive — it changes something real and undoing it takes work.
+	// One confirm turn, every time, never remembered.
+	TierDestructive Risk = "destructive"
+	// TierIrreversible — the thing it acts on does not come back: a wipe, a
+	// format, a delete with no bin behind it. A confirm turn is not enough,
+	// because the whole chain that proposed it — an STT guess, a router guess,
+	// a fuzzy allowlist match — has a spoken "да" as its only check. She names
+	// the gap and he runs it himself.
+	TierIrreversible Risk = "irreversible"
+)
+
+// Policy — what a tier requires of the act path.
+//
+// There is deliberately no "sticky for" field. Non-stickiness is the policy,
+// and a knob that could turn it off would be the thing to argue with instead
+// of the rule.
+type Policy struct {
+	// Confirm — the act does not run on first hearing.
+	Confirm bool
+	// VoiceMayRun — a spoken confirmation is enough authority to run it.
+	VoiceMayRun bool
+}
+
+// PolicyFor returns the requirements of a tier. An unknown tier is treated as
+// destructive, for the same reason the default derivation is.
+func PolicyFor(r Risk) Policy {
+	switch r {
+	case TierSafe:
+		return Policy{Confirm: false, VoiceMayRun: true}
+	case TierIrreversible:
+		return Policy{Confirm: true, VoiceMayRun: false}
+	default:
+		return Policy{Confirm: true, VoiceMayRun: true}
+	}
+}
+
+// irreversibleVerbs — argv heads and subcommands that destroy the thing they
+// name. Matched as whole argv elements, never as substrings: "rm" must not
+// fire on "/usr/bin/rmdir-report" and "drop" must not fire on "dropbox".
+//
+// The list is short on purpose. It is not a sandbox and it does not try to be
+// one — an enabled row can already run anything the daemon's user can run.
+// What it is, is the set of words that mean "and then it is gone", so that the
+// one act nobody can walk back is the one act a spoken "да" cannot authorise.
+var irreversibleVerbs = map[string]bool{
+	"rm": true, "rmdir": true, "shred": true, "srm": true,
+	"mkfs": true, "fdisk": true, "parted": true, "wipefs": true,
+	"dd": true, "format": true,
+	"drop": true, "drop-database": true, "destroy": true, "purge": true,
+	"prune": true, "truncate": true,
+}
+
+// RiskOf derives the tier of an enabled tool row.
+func RiskOf(t ipc.Tool) Risk {
+	if isIrreversible(t.Cmd) {
+		return TierIrreversible
+	}
+	// A house row is a physical change to the flat, and the confirm turn on it
+	// is structural rather than a column: /tools writes the checkbox straight
+	// through on enable, so unticking it once turned an unlock into a row that
+	// ran on first hearing. Nothing any surface writes removes the second turn
+	// from a physical device.
+	if _, _, ok := smarthome.ParseCmd(t.Cmd); ok {
+		return TierDestructive
+	}
+	// An MCP row is a call to somebody else's server. It is enabled with a
+	// fingerprint of what it declared at approval time (Vikunja #251), and the
+	// tier tracks the same flag every other row uses — the point of this branch
+	// is that it is NOT special-cased into running freely.
+	if _, _, ok := mcp.ParseCmd(t.Cmd); ok {
+		if t.Destructive {
+			return TierDestructive
+		}
+		return TierSafe
+	}
+	if t.Destructive {
+		return TierDestructive
+	}
+	if len(t.Cmd) == 0 {
+		// Not a shape this file knows how to read. The default is the confirm
+		// turn: a new domain argues its way down, not up.
+		return TierDestructive
+	}
+	return TierSafe
+}
+
+// isIrreversible reports whether any argv element is one of the verbs that
+// destroys what it names. Every element, not just the head: "sudo rm" and
+// "docker volume prune" both hide the verb behind a wrapper.
+func isIrreversible(cmd []string) bool {
+	for _, arg := range cmd {
+		word := strings.ToLower(strings.TrimSpace(arg))
+		// Take the last path element, so /bin/rm reads as rm.
+		if i := strings.LastIndex(word, "/"); i >= 0 {
+			word = word[i+1:]
+		}
+		if irreversibleVerbs[word] {
+			return true
+		}
+	}
+	return false
+}

internal/tool/risk_test.go

@@ -0,0 +1,81 @@
+package tool
+
+import (
+	"context"
+	"errors"
+	"testing"
+
+	"github.com/kami/maven/internal/ipc"
+)
+
+func TestRiskOfReadsTheRow(t *testing.T) {
+	cases := []struct {
+		name string
+		tool ipc.Tool
+		want Risk
+	}{
+		{"a plain read", ipc.Tool{Cmd: []string{"systemctl", "status"}}, TierSafe},
+		{"the checkbox", ipc.Tool{Cmd: []string{"systemctl", "restart"}, Destructive: true}, TierDestructive},
+		{"a wipe", ipc.Tool{Cmd: []string{"rm", "-rf"}}, TierIrreversible},
+		{"a wipe behind a wrapper", ipc.Tool{Cmd: []string{"sudo", "/bin/rm"}}, TierIrreversible},
+		{"a prune behind a subcommand", ipc.Tool{Cmd: []string{"docker", "volume", "prune"}}, TierIrreversible},
+		{"the house", ipc.Tool{Cmd: []string{"smarthome", "light.kitchen", "turn_off"}}, TierDestructive},
+		{"the house with the box unticked", ipc.Tool{Cmd: []string{"smarthome", "lock.front", "unlock"}}, TierDestructive},
+		{"an mcp read", ipc.Tool{Cmd: []string{"mcp", "vikunja", "list_tasks"}}, TierSafe},
+		{"an mcp write", ipc.Tool{Cmd: []string{"mcp", "vikunja", "delete_task"}, Destructive: true}, TierDestructive},
+		{"a shape nobody wrote yet", ipc.Tool{}, TierDestructive},
+	}
+	for _, c := range cases {
+		if got := RiskOf(c.tool); got != c.want {
+			t.Errorf("%s: RiskOf = %q; want %q", c.name, got, c.want)
+		}
+	}
+}
+
+// The default is the confirm turn. A tier this file does not know is not a
+// tier that runs freely.
+func TestPolicyForDefaultsToConfirming(t *testing.T) {
+	for _, r := range []Risk{TierDestructive, Risk("whatever-lands-here-next")} {
+		p := PolicyFor(r)
+		if !p.Confirm || !p.VoiceMayRun {
+			t.Errorf("PolicyFor(%q) = %+v; want a confirm turn she may run", r, p)
+		}
+	}
+	if p := PolicyFor(TierSafe); p.Confirm || !p.VoiceMayRun {
+		t.Errorf("PolicyFor(safe) = %+v; want it to run", p)
+	}
+	if p := PolicyFor(TierIrreversible); !p.Confirm || p.VoiceMayRun {
+		t.Errorf("PolicyFor(irreversible) = %+v; want voice refused", p)
+	}
+}
+
+// An irreversible act is refused whether or not he said "да", because there is
+// no second answer that changes what it would do.
+func TestExecRefusesIrreversibleEvenConfirmed(t *testing.T) {
+	api := fakeAPI{tools: map[string]ipc.Tool{
+		"wipe": {Name: "wipe", Status: "enabled", Cmd: []string{"rm", "-rf"}, Destructive: true},
+	}}
+	e := NewExecutor(api, 0)
+	ran := false
+	e.run = func(context.Context, []string) (string, error) { ran = true; return "", nil }
+	for _, confirmed := range []bool{false, true} {
+		if _, err := e.Exec(context.Background(), "wipe", []string{"/data"}, confirmed); !errors.Is(err, ErrNeedsAuthedSurface) {
+			t.Errorf("confirmed=%v: %v; want ErrNeedsAuthedSurface", confirmed, err)
+		}
+	}
+	if ran {
+		t.Fatal("an irreversible act ran from the voice path")
+	}
+}
+
+// A row with no cmd at all is not a shape this file reads, and it must not
+// slide through as safe.
+func TestExecConfirmsAnUnreadableRow(t *testing.T) {
+	api := fakeAPI{tools: map[string]ipc.Tool{
+		"mystery": {Name: "mystery", Status: "enabled"},
+	}}
+	e := NewExecutor(api, 0)
+	if _, err := e.Exec(context.Background(), "mystery", nil, false); !errors.Is(err, ErrNeedsConfirm) {
+		t.Errorf("%v; want ErrNeedsConfirm", err)
+	}
+}

internal/tool/tool.go

@@ -67,6 +67,12 @@ var (
 	// proposal, and drafting a new proposal for a tool that already exists and
 	// is enabled is a lie about what is wrong.
 	ErrNotConnected = errors.New("tool is enabled but its backend is not connected")
+	// ErrNeedsAuthedSurface — the row is enabled and the act is understood,
+	// and its tier is one a spoken "да" may not authorise (risk.go,
+	// TierIrreversible). Held apart from ErrNeedsConfirm because there is no
+	// confirm turn that would help: asking again would imply the second answer
+	// changes the outcome.
+	ErrNeedsAuthedSurface = errors.New("tool is irreversible and voice may not authorise it")
 )
 
 // MCPCaller is the seam for an act that is an MCP tool call rather than a
@@ -121,7 +127,12 @@ func (e *Executor) WithHome(h HomeCaller) *Executor {
 // Exec looks up name in the store and runs Cmd+args as argv (no shell).
 // confirmed=true is the second turn of a destructive act (the user said "да");
 // it bypasses the ErrNeedsConfirm gate. Non-enabled ⇒ ErrNotEnabled; a
-// destructive tool with confirmed=false ⇒ ErrNeedsConfirm.
+// destructive tool with confirmed=false ⇒ ErrNeedsConfirm; an irreversible one
+// ⇒ ErrNeedsAuthedSurface, confirmed or not.
+//
+// Exec IS the voice path. Nothing else calls it, which is why the tier check
+// needs no surface argument: the authority it can offer a tool is a spoken
+// "да", and TierIrreversible says that is not enough.
 func (e *Executor) Exec(ctx context.Context, name string, args []string, confirmed bool) (string, error) {
 	t, err := e.api.LookupTool(ctx, name)
 	if errors.Is(err, ipc.ErrToolNotFound) {
@@ -133,7 +144,15 @@ func (e *Executor) Exec(ctx context.Context, name string, args []string, confirm
 	if t.Status != "enabled" {
 		return "", ErrNotEnabled
 	}
-	if t.Destructive && !confirmed {
+	// The tier decides, not the column (Vikunja #449). RiskOf reads the row and
+	// answers the three questions the boolean never did: which acts are
+	// destructive, whether a confirm sticks (it never does), and what an
+	// unrecognised shape inherits (the confirm turn).
+	policy := PolicyFor(RiskOf(t))
+	if !policy.VoiceMayRun {
+		return "", ErrNeedsAuthedSurface
+	}
+	if policy.Confirm && !confirmed {
 		return "", ErrNeedsConfirm
 	}
 	// An MCP row is a call to a configured server, not a process. Everything