Merge pull request 'reminder_verbs has no alarm verb, so an alarm never routes (V-627)' (#180) from task/627-reminder-verbs-has-no-alarm-verb-so-an-a into master

reminder_verbs held five words and none named an alarm, and ReminderGrammar
did not read the set anyway — it carried the literal напомни|remind me. So no
part of the cascade recognised разбуди, and the three alarm cases in the
fixture went to fact and act at over 0.89.

The lexicon addition alone moved nothing, measured at 66/91. Every consumer
reads the set after a reminder route already exists. Building the grammar's
alternation from the set is what scored: 66/91 to 69/91, three cases gained,
none lost, and each alarm now carries its time slot.

Longest-first ordering in the alternation is load-bearing. Go's regexp
alternation is leftmost-first, so напомнить after напомни would never match.

Found while training the V-546 intent head, where the same three cases went
to system.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0117tgnmbgZpHVV3XSNw8Qua

cmd/labelgen/main.go

@@ -0,0 +1,113 @@
+// Command labelgen labels utterances with the stage 0 grammars and prints JSONL.
+//
+// docs/plans/18-routing-heads-on-e5-small.md calls the labeled set the whole
+// project, and it names the stage 0 grammars as the high-precision label
+// functions to start from. This runs them — the real ones, in the real
+// buildRouter order — rather than a reimplementation, so a rule change moves
+// the training data with it.
+//
+// A grammar that declines leaves the line unlabeled. Those go to the model, and
+// keeping them is the point: a set labeled only by the rules teaches only the
+// rules.
+//
+//	go run ./cmd/labelgen < utterances.txt > labeled.jsonl
+//
+// The wakeword-act grammar is absent, because its allowlist is the deployment's
+// enabled tool names and this tool has no deployment. Every other rule is here.
+package main
+
+import (
+	"bufio"
+	"encoding/json"
+	"fmt"
+	"os"
+	"strings"
+
+	"github.com/kami/maven/internal/router"
+)
+
+// label is one output row. The grammar name rides along so a reviewer can see
+// which rule made the claim, and so a rule that turns out to be wrong can have
+// its rows pulled without re-running everything.
+type label struct {
+	Utterance string `json:"utterance"`
+	Intent    string `json:"intent,omitempty"`
+	Grammar   string `json:"grammar,omitempty"`
+	Key       string `json:"key,omitempty"`
+	Value     string `json:"value,omitempty"`
+	Fn        string `json:"fn,omitempty"`
+	Text      string `json:"text,omitempty"`
+	Labeled   bool   `json:"labeled"`
+}
+
+// grammars mirrors buildRouter's order in cmd/mavend/voicewire.go. Order is
+// load-bearing there and so it is here: the agenda rules must sit after the
+// clock rules, Praxis before the capture marker, the narrative rules last.
+func grammars() []router.Grammar {
+	var g []router.Grammar
+	g = append(g, router.SystemTimeDateGrammars()...)
+	g = append(g, router.AgendaQueryGrammars()...)
+	g = append(g, router.FeedQueryGrammar())
+	g = append(g, router.TaskListGrammar())
+	g = append(g, router.ListGrammars()...)
+	g = append(g, router.ReminderGrammar())
+	g = append(g, router.PraxisGrammars()...)
+	g = append(g, router.TaskCaptureGrammar())
+	g = append(g, router.NarrativeQueryGrammars()...)
+	return g
+}
+
+func match(gs []router.Grammar, utterance string) label {
+	out := label{Utterance: utterance}
+	for _, g := range gs {
+		m := g.Pattern.FindStringSubmatch(utterance)
+		if m == nil {
+			continue
+		}
+		d, ok := g.Build(m)
+		if !ok {
+			continue // the rule saw its shape and declined it
+		}
+		out.Intent = string(d.Intent)
+		out.Grammar = g.Name
+		out.Key = d.Slots.Key
+		out.Value = d.Slots.Value
+		out.Fn = d.Slots.Fn
+		out.Text = d.Slots.Text
+		out.Labeled = true
+		return out
+	}
+	return out
+}
+
+func main() {
+	gs := grammars()
+	in := bufio.NewScanner(os.Stdin)
+	in.Buffer(make([]byte, 0, 64*1024), 1024*1024)
+	out := bufio.NewWriter(os.Stdout)
+	defer out.Flush()
+
+	enc := json.NewEncoder(out)
+	var seen, labeled int
+	for in.Scan() {
+		line := strings.TrimSpace(in.Text())
+		if line == "" || strings.HasPrefix(line, "#") {
+			continue
+		}
+		seen++
+		l := match(gs, line)
+		if l.Labeled {
+			labeled++
+		}
+		if err := enc.Encode(l); err != nil {
+			fmt.Fprintln(os.Stderr, "labelgen:", err)
+			os.Exit(1)
+		}
+	}
+	if err := in.Err(); err != nil {
+		fmt.Fprintln(os.Stderr, "labelgen:", err)
+		os.Exit(1)
+	}
+	// Coverage on stderr, so the count is visible without polluting the JSONL.
+	fmt.Fprintf(os.Stderr, "labelgen: %d/%d labeled by %d grammars\n", labeled, seen, len(gs))
+}

cmd/mavend/ecosystem_acts.go

@@ -139,9 +139,9 @@ func (h *reactiveHandler) handlePraxisAct(ctx context.Context, dec router.Decisi
 // praxisItemAction is the shared shape of the item-lifecycle capabilities: take
 // an item id from the value slot, call one Praxis endpoint, trace the result.
 type praxisItemAction struct {
-	verbs   []string
-	ask     string // reply when no item id was given
-	op      string // trace + log name of the operation
+	verbs []string
+	ask   string // reply when no item id was given
+	op    string // trace + log name of the operation
 	// failure is the first half of the reply when the Praxis call errors: which
 	// operation did not happen. ecosystemGap supplies the second half, which
 	// names Praxis and splits a refused token from an outage — those two used to

docs/evals/2026-08-06-alarm-verbs-reach-stage-0.md

@@ -0,0 +1,51 @@
+# Alarm verbs reach stage 0
+
+**06-08-2026. V-627.** Measured with `TestONNXBaseline`, 91-case RU routing fixture,
+classifier plus the ONNX embedder. No LLM arm in this run.
+
+## What was wrong
+
+`lexicon.ReminderVerbs` held five words and none of them named an alarm. `ReminderGrammar`
+in `internal/router/stage0.go` did not read the set at all: it carried the literal
+`напомни|remind me`. So no part of the cascade recognised `разбуди`.
+
+Three fixture cases ride on that. Under the classifier they went to fact and act at high
+confidence, so the failure was never a near miss:
+
+- `ru-rem-005` "разбуди меня в 6:30" to fact at 0.918
+- `ru-rem-009` "разбуди меня полвосьмого" to act at 0.941
+- `en-rem-002` "wake me at 6:15" to fact at 0.899
+
+Found while training the V-546 intent head, where the same three cases went to system. The
+head reads a spoken time with no known verb in front of it as a clock question. The
+classifier was making the same mistake in its own way.
+
+## The change
+
+Four alarm imperatives and bare `wake` join `reminder_verbs`. `ReminderGrammar` builds its
+alternation from the set, longest alternative first, and eats an optional `мне`, `меня` or
+`me` before the body.
+
+Longest-first is load-bearing. Go's alternation is leftmost-first rather than longest-match,
+so `напомнить` listed after `напомни` would never match.
+
+## Result
+
+**66/91 to 69/91, 72.5% to 75.8% full.** Three cases gained, none lost.
+
+All three are the alarms above, and each now carries its time slot, which it did not before.
+Clarify counts unchanged at 0 false and 8 missed. The two remaining system failures,
+`какое число завтра` and `какой день недели послезавтра`, failed at baseline too.
+
+## What this does not fix
+
+The lexicon addition on its own moved nothing. Measured before touching the grammar:
+**66/91**, exactly the baseline. Every consumer of `reminder_verbs` reads it after a reminder
+route already exists. A verb that cannot win the route is a verb nobody asks about. The
+grammar was the whole change.
+
+Lemma matching in `isReminderVerb` now covers `разбудил` as well as `разбуди`, because one
+lemma holds both. That is the trap `cmd/mavend/quiet_toggle.go` documents for `говори`. It
+is tolerable here and not in the quiet toggle. `isReminderVerb` runs only on an utterance
+already routed to reminder, and it decides where the subject starts. A quiet match flips a
+daemon-wide setting from any channel.

docs/evals/2026-08-06-seed-labels-vs-router-prompt.md

@@ -0,0 +1,82 @@
+# Gemma as a label function, and what it found in the seeds
+
+**06-08-2026. V-546.** Measured on workpc against gemma-4-12b-it-qat-UD-Q4_K_XL.
+
+`docs/plans/18-routing-heads-on-e5-small.md` puts the labeled set at 20k examples through
+gemma, costing 2 to 4 hours of the card. This is the check before spending that. Gemma
+labels the 344 hand-written classifier seeds. Agreement with the label a person already
+chose is a precision number rather than a guess.
+
+## What ran
+
+`cmd/labelgen` runs the stage 0 grammars. The real ones, in `buildRouter` order, minus
+`wakeword-act`, whose allowlist is a deployment's enabled tool names. It labels 62 of 339
+seed lines and leaves the rest.
+
+The remaining 277 went to gemma through the daemon's own `routeSystem` prompt and
+`routeGrammar`, both extracted from `internal/router/llmrouter.go` at run time rather than
+retyped. Temperature 0.
+
+## Cost
+
+**334ms per call, 0 unparsed of 277.** The GBNF held every time. At that rate the plan's
+20k examples is under two hours of card, which matches its estimate.
+
+## The stage 0 rules as label functions
+
+Agreement between the grammar's label and the seed file the line came from:
+
+| seed intent | agree |
+|---|---|
+| reminder | 37/37 |
+| query | 9/10 |
+| system | 7/8 |
+| act | 2/2 |
+| chat | 0/4 |
+| note | 0/1 |
+
+`ReminderGrammar` at 37/37 is the evidence the plan wanted. The chat column is a defect
+rather than a disagreement: `chatNarrativeTopics` is Russian-only, so `tell me about
+yourself` survives the decline and routes IntentQuery with topic `yourself`. Filed as
+V-625, which also records that `как дела у сервера` appears verbatim in two seed files
+under two intents.
+
+## Gemma against the seeds
+
+**197/277, 71.1%.** By intent:
+
+| seed intent | agree |
+|---|---|
+| note | 33/33 |
+| act | 57/64 |
+| fact | 37/40 |
+| query | 51/54 |
+| chat | 15/35 |
+| system | 4/43 |
+| reminder | 0/8 |
+
+The number is not gemma's error rate. Reading the 80 disagreements, most are the seed files
+and the prompt holding different definitions of the same intent. Three boundaries carry 42
+of them, and V-626 is the fix:
+
+- **system, 26 lines.** The prompt restricts system to the clock, the calendar date and the
+  assistant itself. The seeds also put sensor and host state there. That is the V-374 edit
+  of 31-07-2026, which the seeds never received.
+- **world questions, 8 lines.** `почему небо голубое`, `why is the sky blue`. Written when
+  chat was the only honest destination for a question nothing could answer, and external
+  search now answers them.
+- **bare verbs, 8 lines.** `поставь напоминание` with nothing to remind about. The prompt
+  calls that unknown. This one is not staleness. A nearest-neighbour centroid wants the
+  bare verb phrase, and that is what a seed file is for.
+
+Four intents have not been redefined since the seeds were written: note, fact, query and
+act. They agree at 178 of 191.
+
+## What this says about the plan
+
+Gemma is usable as a label function on those four and not on system, chat or a bare verb.
+The plan already budgets a day of the owner reading the set. This says where to spend it.
+
+It also says the two engines in the cascade are being taught different rules on 80 lines.
+A routing measurement that swaps between the classifier and the router is measuring some of
+that disagreement rather than the models.

docs/evals/2026-08-06-seeds-to-prompt-boundary.md

@@ -0,0 +1,58 @@
+# Moving the seed files onto the router prompt's boundaries
+
+**06-08-2026. V-626.** Measured with `TestONNXBaseline`, 91-case RU routing fixture,
+classifier plus the ONNX embedder. No LLM arm in this run.
+
+`docs/evals/2026-08-06-seed-labels-vs-router-prompt.md` found three intent boundaries where
+`models/seeds` and `routeSystem` disagree. This applies two of them and rejects the third,
+because the third was measured and it costs a case.
+
+## Baseline
+
+**64/91, 70.3% full.** Latency p50 22.9ms.
+
+## What moved
+
+**Sensor and host state, system to query. 26 lines.** `какая температура воздуха`,
+`сколько памяти занято`, `какой статус сервисов`. The prompt restricts system to the clock,
+the calendar date and the assistant itself, which is the V-374 edit of 31-07-2026.
+
+**World questions, chat to query. 8 lines.** `почему небо голубое`, `why is the sky blue`,
+`как работает интернет`. Only the genuine world-knowledge lines. An opener about herself
+stays in chat. `как тебя зовут` is a question word by rule 4 and about the assistant by
+rule 8. The rules are ordered and rule 4 fires first, which reads wrong. That is a prompt
+question rather than a seed question.
+
+`system.txt` goes from 43 lines to 17 and `query.txt` from 64 to 98.
+
+## Result
+
+**66/91, 72.5% full.** Two cases gained, none lost.
+
+- `en-sys-002` "turn quiet mode back on", quiet 2/3 to 3/3
+- `ru-query-011` "почему сервер тормозит", homelab 5/6 to 6/6
+
+Clarify counts unchanged at 0 false and 8 missed. The eight missed clarifies are the
+`ambiguous` tag and this change does not touch them. `TestONNXRecall`, `TestONNXTopics`,
+`TestONNXPersonalBoundary` and `TestONNXClaimConfidenceDistribution` all pass.
+
+Thinning system to 17 lines did not hurt it. The two remaining system failures,
+`какое число завтра` and `какой день недели послезавтра`, both failed at baseline too.
+
+## The third boundary, measured and rejected
+
+`reminder.txt` holds eight bare verbs: `поставь напоминание`, `создай напоминание`,
+`set a reminder`. Rule 9 of the prompt calls an utterance with no named subject unknown.
+By the prompt they do not belong in a reminder seed set.
+
+Dropping them scores **65/91**, one below keeping them. `ru-rem-004` "поставь напоминание
+через полчаса" falls from reminder to fact, because the centroid loses the phrase the
+utterance is built from.
+
+So the seed file and the prompt are not stale against each other here. They have different
+jobs. A prompt classifies one utterance and can say it cannot. A nearest-neighbour centroid
+is a shape to be near, and a bare verb phrase is part of that shape. The eight lines stay.
+
+That distinction matters past this file. V-546 trains a classification head on labeled
+utterances rather than a centroid, and the head is the prompt's kind of thing. These eight
+lines are seed data and not training data.

internal/lexicon/lexicon_ru_v1.json

@@ -173,10 +173,11 @@
       ]
     },
     "reminder_verbs": {
-      "note": "The imperatives that mean \"remind me\", in the forms he speaks. The same kind of set as capture_verbs and decided the same way: it is her vocabulary, not a discovery about Russian (Vikunja #530).",
+      "note": "The imperatives that mean \"remind me\", in the forms he speaks. The same kind of set as capture_verbs and decided the same way: it is her vocabulary, not a discovery about Russian (Vikunja #530). The alarm verbs joined them in V-627. \"разбуди меня в 6:30\" is a reminder that fires at the hour he gets up, and the set knew no form of it, so an alarm reached IntentReminder only by resembling one to the embedder.",
       "words": [
         "напомни", "напомните", "напомнить", "напоминай",
-        "remind"
+        "разбуди", "разбудите", "разбудить", "буди",
+        "remind", "wake"
       ]
     },
     "half_hour": {

internal/mcp/mcp_test.go

@@ -662,7 +662,7 @@ func (t *emptyFrameTransport) Call(ctx context.Context, req *rpcRequest) (*rpcRe
 }
 
 func (t *emptyFrameTransport) Notify(context.Context, string, any) error { return nil }
-func (t *emptyFrameTransport) Close() error                             { return nil }
+func (t *emptyFrameTransport) Close() error                              { return nil }
 
 func TestResultlessResponseIsNotSuccess(t *testing.T) {
 	c := newClient("empty", &emptyFrameTransport{})

internal/router/stage0.go

@@ -2,6 +2,7 @@ package router
 
 import (
 	"regexp"
+	"sort"
 	"strings"
 	"unicode"
 
@@ -94,10 +95,41 @@ func DefaultGrammars(actMatcher ActMatcher) []Grammar {
 // stage-0 decision too (fillMatchedSlots in router.go). Before that it did not,
 // so "напомни в 11:00 позвонить маме" reached the daemon with HasTime false and
 // was asked "Когда?" about an hour he had just said.
+//
+// The verb alternation is built from lexicon.ReminderVerbs rather than written
+// out (V-627). The literal here knew "напомни" and "remind me" and nothing
+// else, so "разбуди меня в 6:30" never reached stage 0 — and it does not reach
+// IntentReminder further down either, where the classifier calls it fact at
+// 0.918. An alarm is a reminder that fires at the hour he gets up, and the
+// verb that names one is her vocabulary, so it belongs in the lexicon with the
+// rest of it.
+//
+// Longest-first ordering matters: Go's regexp alternation is leftmost-first,
+// not longest-match, so "напомнить" listed after "напомни" would never match.
+var reminderVerbPattern = regexp.MustCompile(
+	`(?i)^\s*(?:` + longestFirstAlternation(lexicon.ReminderVerbs()) +
+		`)\s*(?:мне|меня|me)?[\s,:]+(.+)$`)
+
+// longestFirstAlternation joins a word set into a regexp alternation, longest
+// alternative first, with every member escaped.
+func longestFirstAlternation(set []string) string {
+	out := make([]string, 0, len(set))
+	for _, w := range set {
+		out = append(out, regexp.QuoteMeta(w))
+	}
+	sort.Slice(out, func(i, j int) bool {
+		if len(out[i]) != len(out[j]) {
+			return len(out[i]) > len(out[j])
+		}
+		return out[i] < out[j]
+	})
+	return strings.Join(out, "|")
+}
+
 func ReminderGrammar() Grammar {
 	return Grammar{
 		Name:    "reminder-wakeword",
-		Pattern: regexp.MustCompile(`(?i)^\s*(?:напомни|remind me)[\s,:]+(.+)$`),
+		Pattern: reminderVerbPattern,
 		Build: func(m []string) (Decision, bool) {
 			rest := strings.TrimSpace(m[1])
 			if rest == "" {

models/seeds/chat.txt

@@ -8,12 +8,9 @@
 как прошёл день
 расскажи про себя
 ты мне нравишься
-почему небо голубое
 о чём поговорим
 у тебя есть чувства
-что такое любовь
 расскажи историю
-как работает интернет
 шутка
 анекдот
 пошути
@@ -24,8 +21,6 @@
 что нового
 думаешь о чём-то
 расскажи про космос
-почему трава зелёная
-откуда берётся дождь
 что было интересного сегодня
 как тебя зовут
 сколько тебе лет
@@ -37,7 +32,4 @@ i'm bored
 what's up
 tell me a joke
 do you have feelings
-what is love
 tell me about yourself
-why is the sky blue
-how does the internet work

models/seeds/query.txt

@@ -62,3 +62,37 @@ what did I note about the garden
 будет дождь
 погода на сегодня
 weather in london
+сколько времени осталось до вечера
+какая температура воздуха
+сколько человек дома
+кто сейчас дома
+есть ли кто дома
+кто дома сейчас
+все ли дома
+сколько памяти занято
+какая загрузка процессора
+сколько свободного места на диске
+какой ip адрес у сервера
+всё ли работает
+сколько сервер работает без перезагрузки
+когда сервер запускался
+какая версия софта
+сколько аптайм
+какой статус сервисов
+все ли сервисы работают
+что с интернетом
+когда последний раз перезагружался
+сколько трафика сегодня
+какая скорость интернета
+сколько процессов запущено
+как загрузка системы
+сколько оперативной памяти свободно
+какая температура процессора
+почему небо голубое
+что такое любовь
+как работает интернет
+почему трава зелёная
+откуда берётся дождь
+what is love
+why is the sky blue
+how does the internet work

models/seeds/system.txt

@@ -5,36 +5,10 @@
 который час в Москве
 сколько сейчас времени
 который час у нас
-сколько времени осталось до вечера
 какой сегодня день недели
-какая температура воздуха
-сколько человек дома
-кто сейчас дома
-есть ли кто дома
-кто дома сейчас
-все ли дома
-сколько памяти занято
-какая загрузка процессора
-сколько свободного места на диске
-какой ip адрес у сервера
 как дела у сервера
-всё ли работает
-сколько сервер работает без перезагрузки
-когда сервер запускался
-какая версия софта
-сколько аптайм
-какой статус сервисов
-все ли сервисы работают
-что с интернетом
 интернет работает
-когда последний раз перезагружался
-сколько трафика сегодня
-какая скорость интернета
 загрузка сети
-сколько процессов запущено
-как загрузка системы
-сколько оперативной памяти свободно
-какая температура процессора
 тихий режим
 тихо
 не шуми
@@ -48,4 +22,4 @@
 quiet mode on
 quiet mode off
 quiet on
-quiet off
+quiet off