Give the stage 0 grammar set one home (V-693)

buildRouter held the real set and baselineGrammars in eval_test.go restated it
by hand, in the daemon's order, with its own comment saying so. Three test files
score against the fixture and nothing compared the two lists. They had already
drifted: BareCaptureGrammar went into the daemon with V-557 and never into the
fixture, so every routing measurement since has scored a set nobody runs. That
is the failure CLAUDE.md warns about by name, and a diff test would have caught
it one grammar late.

The list moves to router.StageZeroGrammars in internal/router/stagezero.go, with
the ordering comments, which are the load-bearing part. buildRouter and the
fixture both call it. One list cannot drift from itself.

Measured before and after on the 96-case fixture: classifier+onnx 72/96, 75.0%
intent, 33.3% destination, identical either way, and the deterministic claim and
reach hash ratchets do not move. So the missing grammar cost no measurable
accuracy. That is the point rather than a reprieve: the fixture had been scoring
the wrong set for four days and nothing could say so.

The invariants caveat is deleted, both entries, since V-692 landed the other
guard in the previous commit. The reasoning for both now sits in docs/routing.md
beside the subsystem, which is where a fix's durable record belongs.

Unrelated and pre-existing: TestONNXPersonalBoundary fails on "я рассказывал
тебе про байкал?" (personal 0.9068, world 0.9413) at the merge base too.

.gitignore

@@ -70,3 +70,5 @@ coverage.out
 
 # root .env — MAVEN_AMBIENT_TOKEN and friends, same class as deploy/telegram.env
 .env
+# silero-vad, downloaded (see AGENTS.md)
+/models/vad/

AGENTS.md

@@ -95,6 +95,22 @@ model: the code puts `query: ` in front of a question and `passage: ` in front
 of a stored note, which is how e5 was trained. The quantized file is the one
 that is downloaded, deployed and measured.
 
+## Voice activity model for mavwaked
+
+`mavwaked` decides an utterance has started with silero-vad when `-vad-model`
+points at it, and with an energy threshold when it does not. The model is 2.3MB
+and is not committed:
+
+```sh
+mkdir -p models/vad
+curl -sL -o models/vad/silero_vad.onnx \
+  https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/silero_vad.onnx
+```
+
+It needs the same `libonnxruntime.so` the embedder needs, passed as `-onnx-lib`
+or read from `MAVEN_ONNX_LIB`. The measurement is
+`docs/evals/2026-08-09-silero-vad.md`, and the tests skip without the file.
+
 **Also need ONNX Runtime** (`libonnxruntime.so`):
 
 ```sh

CLAUDE.md

@@ -1,627 +1,211 @@
 # CLAUDE.md
 
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+Guidance for Claude Code (claude.ai/code) working in this repository.
 
-Maven is a self-hosted, privacy-first voice assistant (Russian + English). Go daemons
-talking over unix sockets; one resident small model for routing + phrasing; whisper.cpp STT, piper TTS.
-Deploy target is a Ryzen laptop (homesrv) with Vulkan offload to the Vega iGPU (`n_gpu_layers: 99`,
-compose passes `/dev/dri` + the render gid) — the resident model stays ≤1.7B either way.
+**This is a rules file.** It loads into every session, so it carries only what
+changes what an agent does. A measurement belongs in `docs/evals/`, dated and
+never edited after the day. A subsystem's reasoning belongs in its living doc
+under `docs/`. Read that doc before changing the subsystem.
 
-**Resident model:** currently **Qwen3-1.7B** (`UD-Q4_K_XL`), stock — not yet the CPT'd one.
-It replaced Qwen3.5-0.8B on 2026-07-31 because it measured better on both fixtures we have:
-67.5% vs 59.7% intent-only on the 77-case RU routing fixture, and 20/27 vs 11-17/27 on the
-talk fixture. See `docs/evals/2026-07-31-model-bakeoff.md`. It is a Thinking variant, so `n_ctx` is 4096
-— reasoning tokens need the room, and 4096 is what the scores above were measured at.
+| Read this | Before |
+|---|---|
+| `docs/routing.md` | touching `internal/router/` or `queryWalk` |
+| `docs/deployment.md` | touching a daemon, compose, a systemd unit or the web UI |
+| `docs/offload.md` | touching a daemon seam or adding a model caller |
+| `docs/world.md` | touching search, Kiwix or the world chain |
+| `docs/language.md` | changing a prompt contract or a Russian word list |
+| `docs/ecosystem.md` | touching Nexus, Praxis or Hexis |
+| `docs/rearchitecture.md`, `docs/design.md` | changing the shape of anything |
+| `docs/workflow.md` | the five stores, the doc tiers, the guards |
+| `docs/caveats/` | a known limit, its task id and its revisit trigger |
+| `docs/CLAUDE.md` | which tier a doc belongs in, and what each one holds |
+| `AGENTS.md` | local preview, screenshots, model downloads |
 
-The **target** is still the locally CPT'd **Qwen3-1.7B** (Vikunja #122, training in flight).
-Stock already speaks good Russian; what it gets wrong is the persona — it writes `я рад`,
-masculine, where Maven needs `рада`. That is what the CPT is for.
+## What Maven is
 
-**Do not bother with sub-500M models.** LFM2.5-230M and 350M were measured on 2026-07-31 and
-both are unusable in Russian: the 350M routes at 5.2% (worse than guessing) and answers
-"столица Франции?" with the invented non-word "Сторзит"; the 230M replies to Russian in
-Spanish. Their strong published IFEval/BFCL numbers are English-only. Model files live in
-`/mnt/hdd1/llms`, bind-mounted to `/opt/maven/models/llm` — which **shadows** the repo's
-`models/llm/`, so the LFM2.5 gguf sitting there is not loaded by anything. Swapping the resident
-model is a one-line change to `phraser.model_path` in `deploy/mavend.json`.
+A self-hosted, privacy-first voice assistant in Russian and English. Go daemons
+talk over unix sockets. One resident small model routes and phrases. whisper.cpp
+does speech-to-text and piper does text-to-speech.
 
-See `docs/rearchitecture.md` for the target architecture, `docs/design.md` for the folded design spec, and
-`AGENTS.md` for local-preview + model-download recipes.
+The resident model is **Qwen3-1.7B** (`UD-Q4_K_XL`) on homesrv, a Thinking
+variant at `n_ctx` 4096. Keep it at 1.7B or under. Sub-500M models are unusable
+in Russian (`docs/evals/2026-07-31-model-bakeoff.md`). Model files live in
+`/mnt/hdd1/llms`, bind-mounted over the repo's `models/llm/`, so a gguf sitting
+in the repo is loaded by nothing.
 
-**Model work is moving to the workstation** (owner's call, 2026-08-02). homesrv cannot grow a
-GPU and the workstation has 16GB of VRAM. So the resident model, STT and TTS become preferred
-remotes with a floor on homesrv. The workstation is never assumed up. Fall back silently when
-it would only do the job better. Name the gap when the 1.7B cannot do it at all. The embedder
-stays on homesrv permanently, because it backs that floor. It is multilingual-e5-small,
-quantized and asymmetric — `EmbedQuery` and `EmbedPassage` apply the `query:`/`passage:`
-prefixes it was trained with, and calling plain `Embed` on a note is a bug. It replaced
-MiniLM and bought ten points of recall@1 and 2.5× the speed; see
-`docs/evals/2026-08-04-recall-e5-small.md`. Read `docs/offload.md` before
-touching a daemon seam or adding a model caller. Vikunja #483 is the umbrella, #484 to #487
-are the work.
+The workstation is workpc and it holds the remote model and speech-to-text.
+**It is never assumed up.** **Fall back silently** when it would only do the job
+better. **Name the gap** when the resident model cannot do the job at all.
 
-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 the owner 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.
+**The embedder stays on homesrv permanently**, because it backs that floor.
+`EmbedQuery` and `EmbedPassage` apply the `query:` and `passage:` prefixes
+multilingual-e5-small was trained with. Calling plain `Embed` on a note is a bug.
 
-## Build & test
+## Build and test
 
-CGO daemons (`mavend`, `mavsttd`, `mavttsd`, `mavenclient`) need the vendored toolchain
-and libs wired through the Makefile — **do not** call `go build` on them bare, use `make`:
+CGO daemons (`mavend`, `mavsttd`, `mavttsd`, `mavenclient`) need the vendored
+toolchain wired through the Makefile. **Do not call `go build` on them bare**,
+and **do not hand-write the CGO preamble**. This box runs zsh, so an unquoted
+`-run Test*` dies on "no matches found" before `go` is reached. `make t` also
+carries `-count=1` and sets `MAVEN_ONNX_LIB`. Without that variable the four
+`TestONNX*` measurements self-skip and the run still prints `ok`.
 
 ```sh
-make build          # all 11 binaries
-make build-web      # single daemon (pure-Go ones: web/waked/poll/caldav build without CGO)
-make test           # go test -race across ./internal/... ./cmd/... with CGO env set
-```
-
-Run one package or one test with `make t`. **Do not hand-write the CGO preamble.**
-Past sessions pasted it about 390 times. That is where the shell-quoting failures
-came from. This box runs zsh, so an unquoted `-run Test*` or `--include=*.go`
-dies on "no matches found" before `go` is ever reached.
-
-```sh
-make t PKG=./internal/router/
-make t PKG=./cmd/mavend/ RUN=TestSimulator
+make build   # all 11 binaries. make build-web for one (web/waked/poll/caldav skip CGO)
+make test    # go test -race across ./internal/... ./cmd/... with CGO env set
 make t PKG=./internal/router/eval/ RUN='TestONNX' V=1   # V=1 for -v, RACE=0 to drop -race
+make analyze # staticcheck, deadcode and govulncheck. Not in `test`: all three need the network
 ```
 
-`t` carries `-race`, so a green `make t` cannot turn red under `make test`. It carries
-`-count=1`, so a cached PASS from before your edit is never mistaken for a result.
+**The static gates pass against a baseline, not against zero**
+(`scripts/analyzers/*.baseline`, reasoning in `docs/workflow.md`). A fix must
+delete its baseline entry, because the gate also fails on an entry whose finding
+is gone. **`make audit` is a git-grep inventory, not analysis.** Do not cite it
+as a reachability check.
 
-It also sets `MAVEN_ONNX_LIB`, which the hand-written recipe did not. The four
-`TestONNX*` measurements self-skip when that variable is unset. The run still prints
-`ok`. So every targeted eval done the old way reported the hash ratchet while reading
-as a real embedder score.
+## The daemons
 
-Pure-Go packages (`router`, `memory`, `mavweb`, …) also run under a plain `go test ./pkg/`,
-but `make t` works everywhere and is one thing to remember.
+Eleven binaries under `cmd/`, wired socket-to-socket over `internal/ipc`, not
+linked. `mavend` is the core and owns the DB and the IPC socket.
+`deploy/mavend.json` sets sockets, model paths and the phraser and embedder
+blocks, with `${VAR}` expansion from gitignored `deploy/telegram.env`.
+**`docker-compose.yml` runs five**: `mavend`, `mavsttd`, `mavttsd`, `mavweb`,
+`mavpoll`. Count against compose, not against `make build`. `mavwaked` runs on
+workpc under systemd. `docs/deployment.md` says who else is absent and why.
 
-## The daemons (`cmd/`)
-
-| Binary | Role |
-|---|---|
-| `mavend` | **Core.** Router, phraser, memory, reminders, digestion tick. Owns the DB + IPC socket. |
-| `mavweb` | HTTP UI + PWA (`/dash`, `/history`, `/trace`, `/notifications`, `/tools`); WebAuthn auth. Connects to mavend's socket. |
-| `mavsttd` | Speech-to-text (whisper.cpp, CGO). |
-| `mavttsd` | Text-to-speech (piper subprocess). |
-| `mavwaked` | Wake-word / VAD gate. **Not on homesrv** — see below. |
-| `mavenclient` | Voice loop client (mic → stt → core → tts). **Not on homesrv** — see below. |
-| `mavpoll` | Environment poller: netdata alarms, uptime-kuma, zenmoney, wireguard presence. Writes facts, sends nothing. Telegram is `internal/delivery/telegramsink`, not this. |
-| `mavcaldav` | CalDAV calendar sync. |
-| `mavmaild` | Mail reader (IMAP, read-only). Holds the IMAP password; core never sees it. |
-| `mavgpud` | GPU supervisor. **Runs on workpc, not homesrv** — own unit, `deploy/mavgpud.service`. Keeps llama-server loaded while the card is free (V-488). Maven never asks it for anything, it reads `/health` through `llm.Pair`. |
-| `mavupdate` | Not a daemon. Operator CLI a human runs on the box to deploy a new build. |
-
-Two more binaries have no Makefile target and are built with `go run` or `go build` when
-they are needed. Neither is deployed.
-
-| Binary | Role |
-|---|---|
-| `mavseal` | Recovery tool. Encrypts a live tmpfs working copy back to the ciphertext file when mavend was killed before `defer st.Close()` sealed it. |
-| `labelgen` | Runs the stage 0 grammars over utterances and prints JSONL, the training data for the routing heads (V-546). |
-
-Daemons are wired socket-to-socket, not linked. `internal/ipc` is the client/server wire
-protocol; the config in `deploy/mavend.json` (with `${VAR}` env expansion from gitignored
-`deploy/telegram.env`) sets socket paths, model paths, and the phraser/embedder blocks.
-
-**`docker-compose.yml` runs five: `mavend`, `mavsttd`, `mavttsd`, `mavweb`, `mavpoll`.**
-Count against compose, not against the table. Four of the nine daemons are absent, and each
-absence has a different reason.
-
-`mavmaild` and `mavcaldav` are commented out in compose, each with the reason written
-beside it: the first needs a mail account, the second a CalDAV account, and this box has
-neither. `mavcaldav` used to appear nowhere at all, which was an oversight; it became a
-recorded decision on 07-08-2026 (V-644). Two things ride on that absence and the block
-names them. Agenda questions route to `IntentQuery` at stage 0 (V-498) and the `calendar`
-query source then reads a table nobody writes. And `loop.State.CalendarBusy` is fed by the
-same facts, so the gate's "do not nag mid-meeting" is permanently false. Its password is
-read from a file (`-pass-file`, and `-render-pass-file` for the render collection), never
-taken as a flag value, which is the rule `mavpoll` and `mavmaild` follow too.
-
-**`mavwaked` and `mavenclient` are absent by decision, not oversight** (Vikunja #463,
-`docs/plans/17-where-the-voice-loop-runs.md`).
-homesrv has a microphone — it is a laptop — but it is in the wrong room, so a wake-word
-daemon there listens to nobody. They belong on a client machine where the owner is standing.
-
-**That machine is workpc** (owner's correction, 2026-08-05). This section used to say no
-such machine existed, which was written when the workstation was only a model host. It is
-where he sits most of the day and it has the microphone. `ipc.Dial` already takes
-`tcp://host:port?token=...` through the netaddr seam, so the two daemons need deploying,
-not building. V-515 is that deployment.
-
-Until they are deployed, **the wake word and the VAD gate are covered by unit tests and by
-nothing else**, and push-to-talk through `/dash` is what QA actually covers. Note that
-deploying them does not by itself prove a wake word: `mavwaked` gates on energy and has no
-keyword model (V-487), so the loop runs open until that lands.
+- **Passwords are read from files, never taken as flag values.**
+- **The voice wire is plaintext with no auth.** mavend's voice port stays on
+  homesrv loopback and reaches workpc over ssh. Do not LAN-bind it.
+  `SurfaceVoice` caps acts at L0, and L0 does not cap reading.
+- **A GPU service added beside mavgpud goes in `cmd/mavgpud`, never in systemd.**
+  The card needs one owner. A second unit made mavgpud evict llama-server every
+  few seconds and took the model arm down for eight minutes.
 
 ## The ecosystem: Nexus, Praxis, Hexis
 
-Maven is one of four services. It owns conversation and personal memory. It does not
+Nexus identifies, Praxis observes, Hexis acts, Maven understands. Maven does not
 own identity, operational state, or execution. Full contract in
-`docs/ecosystem.md`.
+`docs/ecosystem.md`. All three are `nil` unless configured and each degrades
+alone. An outage means a named gap, never a broken turn or a guess.
 
-```text
-Nexus identifies.  Praxis observes.  Hexis acts.  Maven understands and coordinates.
-```
-
-| Service | Owns | Maven's client | Configured at |
-|---|---|---|---|
-| **Nexus** | Canonical entity ids, names, aliases, relationships. Projects, services, devices, people, pets, places. | `nexusClient` in `cmd/mavend/ecosystem.go`, `POST /api/v1/resolve` | `nexus.url` (`http://nexus:9740`) |
-| **Praxis** | Operational attention and item lifecycle. What needs looking at, what changed, what is still unresolved. | `praxisClient`, the HTTP tools API under `/api/v1/tools/` | `praxis.url` (`http://praxis:8989`) |
-| **Hexis** | The capability registry and the only path to executing anything. | vendored `github.com/kami/hexis/pkg/client` | `hexis.url` (`http://hexis:9741`) |
-
-All three are `nil` unless configured, and every one of them degrades on its own.
-An outage means a named gap in the answer, never a broken turn and never a guess.
-
-Rules that are not negotiable:
-
-- **No component reads another component's database.** Praxis attention comes over
-  HTTP, never from its SQLite file.
-- **Identity lives in Nexus.** Do not invent a local fact key for something Nexus
-  resolves. `actionFact` already sets `Subject`, and `cmd/mavend/factenrichment.go`
-  resolves it in the background against Nexus.
-- **Free text never reaches a mutating Hexis call.** Resolve to a canonical entity id
-  first. Ambiguous resolution asks the owner, it does not pick.
+- **No component reads another component's database.** Praxis attention comes
+  over HTTP, never from its SQLite file.
+- **Identity lives in Nexus.** Do not invent a local fact key for something
+  Nexus resolves. `cmd/mavend/factenrichment.go` resolves `actionFact.Subject`.
+- **Free text never reaches a mutating Hexis call.** Resolve to a canonical
+  entity id first. Ambiguous resolution asks the owner, it does not pick.
 - **LLM output is not authorization.** Confirmation binds capability id, target
-  entity, arguments, requester and expiry. See `cmd/mavend/confirm.go`.
-- **Praxis lifecycle words mean different things.** Surfaced is not acknowledged,
-  acknowledged is not resolved, execution success is not recovery. Reading an item
-  aloud calls `Surface`, never `Acknowledge`.
-- **No automatic attention-to-action path.** Digestion may summarise Praxis. It may
-  not call Hexis.
+  entity, arguments, requester and expiry (`cmd/mavend/confirm.go`).
+- **Praxis lifecycle words differ.** Surfaced is not acknowledged, acknowledged
+  is not resolved, execution success is not recovery. Reading an item aloud
+  calls `Surface`, never `Acknowledge`.
+- **No automatic attention-to-action path.** Digestion may summarise Praxis and
+  may not call Hexis.
+- Every cross-service call carries a correlation id minted once per action
+  (`withCorrelationID`), a contract version header, and `X-Requested-By: maven`.
 
-Every cross-service call carries a correlation id minted once per action
-(`withCorrelationID`), a contract version header, and `X-Requested-By: maven`.
+## Routing
 
-## Routing — read this before touching the router
+**Read `docs/routing.md` before touching `internal/router/` or `queryWalk`.** It
+carries the reasoning, the measurements and every rule's why. A route produces
+two decisions. **Intent** is one of seven values. **Source** is where the answer
+lives and is read on `IntentQuery` alone. Score them separately. The cascade is
+stage 0 grammars, then the routing heads, then the resident model, then the
+classifier. Every stage may decline and the next one answers.
 
-`internal/router/` has TWO layered engines. **The LLM router is now the default and it is
-on in deploy** — this section used to say it was wired `nil`, which stopped being true on
-2026-07-31.
+- **The classifier is the floor, not dead code.** It answers when the resident
+  model is off, absent, or erroring. **Any model error falls through.**
+- **The stage 0 set lives in `router.StageZeroGrammars`**, and both `buildRouter`
+  and the eval fixture call it. Add a grammar there, in the right place, and read
+  the comment above the line you insert after. Do not restate the list anywhere.
+- **Go's `\b` is ASCII-only** and never fires after a Cyrillic letter. A Russian
+  pattern needs an explicit `(\s|[?!.]|$)`.
+- **`PraxisGrammars()` is the only path to Praxis**, not a faster one.
+- **`voice.embedder.heads_path` must never point at `model_path`.** Recall
+  depends on the resident e5-small scoring what it scored. Fine-tune a copy.
+  Refused at config load since V-692, symlinks included.
+- **Routing traces are retained 14 days**, enforced on write and again on start.
+- **Bump `tokenizerRev` on any change to what `encodeWord` emits**, so a
+  tokenizer fix triggers `ReembedAll` the way swapping the model file does.
+- **A new rung in the `runTurn` ladder needs its name in `preRouteLadder`**
+  (`cmd/mavend/decisiontrace.go`), or it is missing from the decision record.
 
-- **LLM router (the intended design, docs/rearchitecture.md):** the resident Qwen3-1.7B (`llmrouter.go`)
-  emits GBNF-constrained structured JSON, and the SAME model phrases replies. Embedder is
-  demoted from a routing gate to a RAG hint. Wired at `voice.go:214` via
-  `pickLLMRouter(cfg.Voice.UseLLMRouter(), llmClient)`; the flag is `voice.llm_router`
-  (`config.go`), `DefaultLLMRouter` is **on**, and `deploy/mavend.json` sets it `true`.
-- **Classifier cascade (the failure floor, not dead code):** `classifier.go` +
-  `embedder.go` nearest-neighbour over frozen seed phrases. It runs when the LLM router is
-  off, when there is no llama-server to talk to (`pickLLMRouter` logs that and degrades),
-  and on any per-turn LLM error. Do not delete it — routing by seed similarity is the known
-  cause of weak RU query handling, but a turn must never break on the model.
+**`queryWalk` takes query sources out and moves none** (`actions_query.go`). That
+is the safety argument and it is not negotiable. The table's order is
+load-bearing and carries "the owner's data first, then the world".
+`SourceUnknown` is the floor and walks the whole chain. A named destination
+removes only the sources marked `guesses: true`, so a source that looks rather
+than guesses is always asked. **The personal boundary is the one exception and
+it is deliberate.** It guesses, so naming `SourceWorld` drops it. **Only a stage
+0 grammar may drop it** (owner's call, V-666). `queryWalk` reads
+`Decision.SourceAnchored` for the source marked `boundary: true` and no other.
 
-Cascade order: `stage0.go` exact-match fast-path → LLM router (when non-nil) → classifier
-fallback. Any LLM error falls through to the classifier so a turn never breaks on the model.
+Judge a routing change against the classifier (76.0% intent, 36.4% destination)
+and the resident model (80.2% intent), since those always answer. The fixture
+has grown from 77 cases to 96, so a number compares only to another number on
+the same fixture.
 
-**A stage-0 decision is slot-extracted too, since 06-08-2026** (V-572). `fillMatchedSlots`
-in `router.go` runs the stage-2 extractor over whatever a grammar built and fills only the
-slots it left empty — a matched value always wins, because the rule read a literal pattern
-and the extractor guesses. It did not run before, so `ReminderGrammar` handed the daemon
-`HasTime: false` for "напомни в 11:00 позвонить маме" and `missingFor` read the silence as
-absence and asked "Когда?". It applies to every grammar and is inert for all but the
-reminder: `Extract` fills Time, Fn and Key and nothing else, and the query, clock, agenda,
-feed, list, task and narrative rules all emit intents with no such slot. Benchmarked at
-20000x, a stage-0 query costs 3.7µs against 3.9µs before. **`Slots.Text` is deliberately not
-filled** — a grammar that left it empty meant it, and `agendaQueryBuild` hands the query
-chain the utterance itself. Fixture unchanged at 64/91, with "slots deferred to daemon"
-6 → 0.
+## Language: model output and Russian
 
-Measured on the 77-case RU fixture. **Re-measured 2026-08-02: the classifier scores 68.8%
-full accuracy at p50 16.6µs**, not the 36.8% at p50 31ms that stood here from
-`docs/evals/2026-07-31-model-bakeoff.md`. That older figure predates the stage 0 rules and the
-seed additions, both of which now score inside the classifier baseline. Qwen3-1.7B scores
-77.9% intent-only / 72.7% through the cascade. So the router buys about 4 points of accuracy,
-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.
+Both contracts are in `docs/language.md`. What must not be broken:
 
-**Re-measured 2026-08-05 on the fixture as it now stands, 91 cases** (V-320 item 2,
-`docs/evals/2026-08-05-routing-resident-model.md`): cascade + resident model scores
-**75.8% full / 80.2% intent-only at p50 1.19s / p95 1.65s**. That is a new baseline and not
-a movement, because 14 cases were added since the 77-case number above. The model alone
-scores 37.4% full against 61.5% intent-only, and the gap is slots rather than routing: it
-routes `reminder` and leaves the time to the daemon, which is what the contract asks. To
-re-run it, start a **second** llama-server on a fixed host port — the resident one binds
-`--port 0` inside the container and no host process can reach it.
+- **One parser for model text, `parseResponseMood`** in
+  `internal/phraser/parse.go`. Every phrasing path reaches it. Mood is an enum.
+- **The router prompt is a separate contract** over 7 intents, and
+  `llm/check_prompt_parity.py` keeps the Go and relabelling copies identical.
+- **Russian words are matched by three mechanisms and no fourth**:
+  `internal/lexicon` for closed classes, `internal/morph` for grammar, and
+  `cmd/mavend/topics.go` with the embedder for open sets. A regex whose output
+  is a fact or a route is the defect. A regex over structured input is not.
+- **Seeds are scoring data.** Editing one moves a recogniser and must be
+  re-measured against the `TestONNX*` tests, not eyeballed.
 
-**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.
+## Non-goals and hard constraints
 
-**The intended third engine is not a generative model** (owner's call, 05-08-2026, V-546,
-`docs/plans/18-routing-heads-on-e5-small.md`). Routing has a bounded output space, so it is
-classification, and the 118M multilingual-e5-small is already resident. Three heads on one
-forward pass: intent, mood, and BIO slot tags. Roughly 5e15 FLOPs to train, so 10 to 30
-minutes on the workstation. A 100M decoder from scratch is 10 to 20 GPU hours. Two things
-it buys that a decoder cannot. No grammar is needed, because a softmax cannot emit a value
-that does not exist. And max softmax is a calibratable confidence, where `Confidence: 1.0`
-was a hardcode. **Fine-tune a copy of the weights.** The resident embedder backs memory
-recall. Training it in place couples routing accuracy to recall@1, with nothing in the
-suite to name the trade.
+Not a nag, not autonomous.
+**The persona is feminine.** Russian self-reference takes feminine forms: `рада`
+not `рад`, `поняла` not `понял`. The owner is male and she speaks to him
+informally. Use "ты", singular, never "вы" or "ваш", and never "он" or "его".
+She talks TO the owner, not about him. Pet names such as "милый" are forbidden.
+The name "Ками" is not. `CheckAddress`, `CheckFeminine` and `CheckCringe` in
+`internal/phraser/eval/checks.go` enforce this, scored by `make eval-phrasing`.
 
-`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
-`gateLLMDecision` in `router.go`. Note the second half of that bug: the LLM branch never
-consulted `r.threshold` at all, so a correct low confidence would have been discarded anyway.
+**"Never phones home" is deprecated** (owner's call, 2026-07-31). She reads
+external sources, and `docs/world.md` carries that chain. What holds regardless:
 
-Re-measured on the fixture after the fix: **missed clarify 6/6 → 1**, at the cost of 3 false
-clarifies and 2.6pt of full accuracy (72.7% → 70.1%, intent-only 67.5% → 74.0%). Two of the
-three false clarifies are acts the model mis-routed and the gate caught — asking beats wrongly
-executing, so the fixture and the daemon disagree about what is correct there. The third,
-`"поужинал"`, was a real defect: the single-token rule was an English intuition and does not
-transfer to Russian, where one word is routinely a whole sentence.
-
-Narrowed 01-08-2026. `thinSingleToken` (`internal/router/singletoken.go`) still thins a bare
-one-word nominal — "вода", "бэкап" — but spares two classes: a closed lexicon of social and
-control singles ("привет", "спасибо", "стоп", "yes"), and any token carrying a Russian verb
-ending (past tense, 2nd person, reflexive), because a verb already contains its subject. Both
-tests are offline and cost nothing. Re-measured: **false clarifies 3 → 2, intent-only 74.0% →
-75.3%, full accuracy unchanged at 70.1%, missed clarify still 1.** The two remaining false
-clarifies are the act-with-no-allowlisted-fn arm of the gate, not this rule.
-
-Agenda questions taken off the model, 01-08-2026. `AgendaQueryGrammars` (`stage0.go`, wired
-after the clock rules in `buildRouter`) routes "что у меня сегодня", "во сколько у меня
-встреча" and anything naming a calendar to `IntentQuery` at stage 0. They were going to
-`IntentSystem`, where `replySystem` has no agenda arm and answered "пока не умею" — the
-fixture had said `query` since ru-query-019 was written. Measured: **full accuracy 70.1% →
-72.7%, intent-only 75.3% → 77.9%, calendar 0/2 → 2/2**, clarify counts unchanged. Note that
-Go's `\b` is ASCII-only and never fires after a Cyrillic letter; the pattern needs an
-explicit `(\s|[?!.]|$)`.
-
-Two more shapes taken off the model, 04-08-2026 (V-498). `rest-of-day-query` inside
-`AgendaQueryGrammars` claims "что дальше?" / "what's next", and `NarrativeQueryGrammar`
-(`stage0.go`, wired **last** in `buildRouter`, after the capture marker) claims "расскажи про
-X", "объясни X", "опиши X". Neither carries a question mark or an interrogative, so the model
-called both `IntentFact`; the write was caught downstream by `IsQuestionShaped`, so this was a
-latency and fixture defect, not a correctness one. The narrative rule reads the same
-`narrativeRequests` lexicon `IsQuestionShaped` reads, and declines `chatNarrativeTopics` — a
-joke, a bedtime story, herself — because the query chain has no source that answers those.
-New fixture cases ru-query-024 and ru-query-025. Classifier + ONNX baseline **56/80 (70.0%) →
-58/82 (70.7%)**, no case regressed, no new false clarify. The LLM arm was not measured (no
-llama-server in that run), so judge it again before quoting a cascade number.
-
-Praxis taken off the model, 05-08-2026 (V-516). `PraxisGrammars()`
-(`internal/router/praxis.go`, wired in `buildRouter` before the capture marker because
-"отметь" is a capture verb) fills `Slots.Fn` with a Praxis capability name.
-**These grammars are the only path to Praxis, not a faster one.** Measured
-2026-08-05 with the resident model as router (V-517,
-`docs/evals/2026-08-05-reach-llm-router.md`): the model alone reaches Praxis
-**0/12**, the same as the classifier alone, because nothing in the router
-prompt names a Praxis capability and there is no string for it to write.
-Through the cascade it is 11/12. Deleting these rules costs every point. Praxis reach
-was **0/12 and structurally so**: `handlePraxisAct` compares `Slots.Fn` to a capability
-alias, and that slot is filled from the deployment's enabled tool names, which no Praxis
-alias is on. Measured **16/30 → 27/30 overall, praxis 0/12 → 11/12, lifecycle 0/5 → 5/5**
-(`docs/evals/2026-08-05-praxis-reach.md`). Two rules to know before editing: a **stative**
-lifecycle word ("готово", "принято") needs an item named beside it, while a bare
-**imperative** ("закрывай") may ask which one. The bare arm additionally requires that
-the sentence name no object of its own, or "закрой шторы в комнате" goes to Praxis instead
-of the house. A demonstrative ("отметь это как сделанное") resolves against
-`h.surfacedItems` only when exactly one item was spoken. Otherwise the turn goes back to
-the cascade rather than transitioning the wrong item.
-
-**Who claimed a turn is now recorded, and so is who did not** (V-564, umbrella
-V-558). Arbitration between the claimants on the utterance stream is order,
-hardcoded in the pre-route resolver ladder, in `buildRouter` and in
-`querySources`. `internal/decision` records one `Record` per turn: every
-claimant, what it would have made the turn, the score it reported, and whether
-it won, declined, lost on score, was thinned by a gate or was **never asked**.
-The record rides the context, the same seam `querysource.go` uses, so a claim
-site cannot change a route and a context with no record costs nothing. It is
-installed in `runTurn`, so the mic, telegram and the web all leave the same
-trail. Storage is a 25-turn in-memory ring on the handler (`decision.Ring`),
-read over `ipc.TurnDecisions` and rendered as the second table on `/trace`.
-**It also persists, since 06-08-2026, and that reverses what this section used to
-say** (V-629, `docs/plans/21-persisting-the-routing-trace.md`). The old rule was
-that nothing persists, because a turn record is read minutes later or never. The
-owner reversed it: the routing heads (V-546) cannot be fitted or calibrated
-without real utterances, and 9 of the 31 modes in `internal/modes` have no seed
-example at all. The ring did not move. It is still what `/trace` reads and still
-what a test with no store gets. `cmd/mavend/routingtrace.go` is a second sink
-beside it, writing `routing_traces` (migration #23). The utterance is stored in
-clear, because a 384-dimension vector of a short sentence is substantially
-recoverable and storing vectors instead would be a privacy claim we cannot
-support. What makes it safe is the same thing that makes the fact store safe.
-Retention is 14 days, enforced on write and again on start, so a box that goes
-quiet does not keep every row. Nothing reads it outward, and the rule
-that his notes and facts are never search input covers this table. `Store.Wipe`
-deletes it with everything else. A correction (V-630) is promoted out into a
-seed-shaped row in `routing_labels` (migration #24) and kept, because a label is
-not a transcript. The transcript still expires. The gesture that writes one is
-two buttons beside the reply on `/chat`, reached over `ipc.CorrectTurn` and the
-trace id that now rides back on `ipc.ChatReply`. A turn marked wrong with no
-target is a usable negative, so naming the intent is never required. The target
-is one of the seven intents and never free text. **All three reaches offer it as
-of 06-08-2026**, and this section used to say only `/chat` did. Voice is the
-`repair` rung, which has read spoken corrections since V-455 and now writes the
-durable label beside the classifier seed it always wrote; a spoken negative with
-no target is its own rung, `repair-negative` (V-636, `docs/plans/22-correcting-a-turn.md`).
-Telegram is an inline keyboard under the reply, and it needed the chat to become
-readable first — **telegram is no longer outbound only** (V-637,
-`docs/plans/23-inbound-telegram.md`). The poller is dark unless the `telegram`
-block says `intake`, it long-polls because the box takes no inbound connections,
-it accepts `chat_id` and no other sender, and it drops whatever queued while the
-daemon was down. It reaches the daemon through `ipc.CoreAPI` alone, so a chat
-turn takes the path `POST /api/chat` takes. Note that the turn source is still
-`tap:text` for both, so provenance cannot tell a chat turn from a typed one.
-Adding a rung to the ladder
-in `runTurn` means adding its name to `preRouteLadder` in
-`cmd/mavend/decisiontrace.go`, or that rung is silently missing from the record.
-
-**A route now says where the answer lives, not only that the turn is a question**
-(V-655, 07-08-2026). `query` was a shrug. The cascade sorted an utterance into one of
-seven intents, with stage 0, the resident model and the classifier behind it. Then
-`IntentQuery` handed the turn to `querySources` in the daemon. That is twenty-two branches
-deciding by seed similarity in a fixed order. It has no fixture and no accuracy
-number, no model arm and no floor. `Decision.Source` (`internal/router/source.go`) is
-the second half of the route. Twelve destinations, not twenty-two. The three recall
-passes plus `fact-by-key` are one destination from outside. So are search, Kiwix and
-the URL reader.
-
-**`SourceUnknown` is a real value and it is the floor.** Nothing named a destination,
-so the daemon walks the whole chain. That is byte-for-byte what shipped before the
-field existed. The classifier arm names nothing, so a box whose model is down routes
-queries exactly as it did.
-
-`queryWalk` in `cmd/mavend/actions_query.go` takes sources **out** and moves none.
-That is the safety argument and it is not negotiable. The table's order is
-load-bearing. Every comment on it argues a reason between two sources, and above all
-it carries "the owner's data first, then the world". Naming `SourceWorld` does not
-send the turn outside. His notes, his facts and the personal boundary still run first.
-
-What comes out is only the sources that **guess**. Those decide a turn is theirs by
-cosine against frozen seeds, then answer whatever they claimed. They hold no table
-that could come back empty. Weather is the pure case and has no local data at
-all. It was measured on the box on 2026-08-07
-(`docs/evals/2026-08-07-week-of-usage.md` section 4). It answered both "что такое
-TCP?" and "сколько будет 17 на 23?" with "для какого города?". The feed answered "какой у меня любимый язык?" with kernel headlines.
-The personal boundary answered "кто такой Линус Торвальдс?" with "не нашла у тебя
-такой записи". A source that guesses is marked `guesses: true` in the table. One that
-looks is not, and it is always asked.
-
-Stage 0 fills the destination where a rule already knows it. `WorldQueryGrammars()`
-(`internal/router/worldquery.go`) claims "что такое X" and "сколько будет 17 на 23".
-It is wired after the agenda rules and **before** the feed and list rules.
-"что такое лента" is a definition question, and the feed rule would take it on the
-noun alone.
-`calendar-query` and `event-time-query` name the calendar. The possessive agenda rules
-deliberately do not. "что у меня в списке покупок" matches `agenda-query`, and naming
-the calendar there would take the list source off the turn.
-
-Fixture unchanged at **69/91 classifier+ONNX**, measured both sides. That is the
-expected result, because it scores intent and no case here changes intent.
-
-**The destination has its own fixture and its own number as of 08-08-2026**
-(V-659, `docs/evals/2026-08-08-destination-fixture.md`). This section used to say
-it had neither. `want_source` on `eval.Case` is a pointer, because the destination
-has three states and a bare string has two. Absent is every intent but query,
-which never reaches `queryWalk`. Present and empty is the `SourceUnknown`
-contract: name nothing and walk the chain. Present and named is a destination the
-route must produce. Thirty-three of ninety-six cases carry one.
-
-A destination miss does **not** fail the case. It lands in `Outcome.SourceReason`
-and never in `Reasons`, so `Accuracy` and `IntentAccuracy` mean what they meant
-and `SourceAccuracy` is a second number over the labelled cases only. Intent and
-destination are two decisions, and one number hides which one moved. A route that
-lost its intent scores no destination hit, or a clarify would satisfy an empty
-label for free.
-
-Measured classifier+ONNX: intent **73/96 (76.0%)**, destination **12/33 (36.4%)**.
-The split is the finding. World is 5/5, because a stage 0 rule names it. The
-`SourceUnknown` floor is 5/7. Calendar is 2/6, because the possessive agenda
-rules deliberately do not name it. And **recall is 0/15, because nothing
-anywhere names it**. Those turns are still answered, since the chain walks
-recall early. Recall is the number the fourth head has to move.
-
-Seven cases assert the floor and six of them are homelab operations. They
-cluster because `SourceRecall`, `SourceNetwork` and `SourceAttention` overlap on
-every question about the box. `mavpoll` writes its netdata and uptime-kuma
-observations into the fact store recall reads. That is a finding about the enum,
-not a gap in the labelling.
-
-`baselineGrammars` in `eval_test.go` mirrors `buildRouter` and had drifted:
-`WorldQueryGrammars` was wired into the daemon by V-655 and not into the mirror,
-so the fixture scored a grammar set nobody runs. Fixed by V-659, worth 3 points of
-destination and nothing else. Check that function when adding a grammar.
-
-**The model arm landed the same day** (V-660,
-`docs/evals/2026-08-08-destination-model-arm.md`). `routeGrammar` carries a
-`source` rule closed over `router.Sources` plus the empty floor, so the model
-cannot emit a destination that does not exist. The prompt lists the twelve in
-Russian and says `""` is a normal answer to give often. `LLMRouter.Route` reads it
-back through `ValidSource` and on `IntentQuery` alone. Against gemma-4-12b on the
-workstation the cascade scores destination **24/33 (72.7%)** with intent unmoved
-at 84.4%, and **recall goes 0/15 to 14/15**. The resident Qwen3-1.7B is
-unmeasured, because it binds `--port 0` inside the container.
-
-**Stage 0 now costs four destination points.** It did not before. The four cases
-the cascade loses and the model alone wins are all calendar. The possessive
-agenda rules claim them first and name nothing on purpose. That caution was free
-while nothing downstream could name anything either. It is not free now, and the
-fix is the owner's call rather than a quiet edit.
-
-The last arm is V-546. Intent, mood and BIO slot tags were already three heads on
-one forward pass of the resident e5-small. Destination is a fourth head on the
-same pass, and 72.7% from a 12B teacher is the label source for training it.
-
-## LLM output contract
-
-All phrasing paths emit `{"response":"...","mood":"..."}`, with fallback to plain text when
-the model skips the JSON. **One parser, `parseResponseMood` in
-`internal/phraser/parse.go`**, and every path reaches it: the six `LLMPhraser` methods,
-`PhraseWorld`, and `Replier.PhraseReply`, which `cmd/mavend/replier_llm.go` wraps — that file
-holds the stub fallback and no parsing of its own. The legacy `{"body","summary"}` fallback
-was deleted on 2026-08-06 (V-397): it was the contract before `{"response","mood"}` replaced
-it, no prompt asks for that shape, the GBNF cannot emit it, and no test covered it.
-Mood is a fixed enum. Router prompt is a separate contract:
-`[{"intent":<enum>, key?, value?, text?, verb?}, ...]`, 7 intents (`fact, reminder,
-note, query, act, chat, system`). `llm/check_prompt_parity.py` in the training
-workspace enforces that the Go and relabelling prompts remain identical.
-
-## Russian patterns — three mechanisms, no fourth
-
-Hand-written Russian stem patterns were swept out on 2026-08-04 (owner's call: not a
-pattern, and the resident model cannot be asked per turn either). A regex whose output is a
-fact or a route is the defect; a regex over structured input — HTML, MIME, JSON, a URL, an
-argv list — is not. Before writing a Russian word list, pick one of these:
-
-- **`internal/lexicon`** — closed classes, in `lexicon_ru_v1.json`. Interrogatives,
-  capture verbs, reminder verbs, cardinals, day offsets, parts of day, weekdays, months,
-  spoken hours. Editing a word is a data change, and there is exactly one copy: months used
-  to live in three files. Cardinals carry the oblique forms, because a spoken time declines
-  and `в семь` / `к семи` are one hour.
-- **`internal/morph`** — grammar, from the vendored golem Russian dictionary. `IsVerbForm`
-  and `SameWord`. Note that lemma matching is BROADER than stem-plus-one-ending, so a verb
-  slot that means the imperative must be matched exactly — `говори` and `говорил` are one
-  lemma and only one of them is a command (`cmd/mavend/quiet_toggle.go`).
-- **`cmd/mavend/topics.go` and the embedder** — open sets, where the question is what a
-  turn is ABOUT. Frozen seeds per subject plus a real `other` class, scored against the
-  turn's own query vector. Same shape as the personal boundary in `personalboundary.go`,
-  with one difference: a topic must clear the runner-up by `topicMargin`, because a false
-  claim here spends a network scan rather than one honest "не знаю". The old keyword tests
-  stay as the offline floor and may remain narrow, since they are no longer the only answer.
-- **The ecosystem trio** — when the answer is not in the utterance at all. Identity is
-  Nexus's, never a local pattern.
-
-Seeds are scoring data. Editing one moves a recogniser and must be re-measured against the
-`TestONNX*` tests, not eyeballed.
-
-## Non-goals (hard constraints)
-
-Not a nag, not autonomous. Maven's persona is **feminine** — Russian
-self-reference must use feminine forms — `рада`, not `рад`; `поняла`, not `понял`. The owner
-is male and is addressed informally: "ты", singular, never "вы"/"ваш" and never "он"/"его"
-(she talks TO the owner, not about the owner). Pet names ("милый", "дорогой") are forbidden; the name
-("Ками") is not. The eval enforces this: `CheckAddress`, `CheckFeminine` and `CheckCringe` in
-`internal/phraser/eval/checks.go`, scored by `make eval-phrasing`.
-
-**"Never phones home" is DEPRECATED** (owner's call, 2026-07-31). It used to be a hard
-constraint and it is not one any more: a 0.8B — and a 1.7B — does not know enough to answer
-world questions, so she needs to read external sources. What replaces it:
-
-- **No telemetry, no cloud model, no third-party account.** That part never changes. Nothing
-  about Maven is reported to anyone, and inference stays on the box.
-- **The owner's data first, then the world.** Every source that reads the owner's facts, notes, calendar,
-  tasks or house runs before anything outside, and the personal boundary sits between them.
-  Reading beats recalling for a small model.
-- **In the world, live search leads and the ZIMs are the fallback** (owner's call,
-  2026-08-02). A self-hosted SearXNG (`search` block) answers first; the Kiwix ZIMs on
-  homesrv answer when the search is empty, unreachable, or the line is down.
-  **Verified with the line down on 2026-08-05** (V-508,
-  `docs/evals/2026-08-05-kiwix-offline-fallback.md`): a stopped SearXNG costs nothing,
-  the ZIM answers in the same turn budget. A blackholed host cost 8 seconds the owner waited
-  through. So the connect phase alone is capped at `dialTimeout` (1.5s), while a slow
-  instance that did connect keeps the full 8. **A Russian question reads
-  `wikipedia_ru_all_maxi_2026-02` verbatim** through `kiwix.book_ru`. The RU→EN rewriter
-  is the workaround for an English book and is skipped there. Kiwix catalog names come
-  from the filename, not the `<name>` field.
-  `Response.Empty()` is the whole gate and there is no quality threshold in front of it:
-  the three signals one could read were measured on 2026-08-05 and none of them separate a
-  real question from an invented one. Token overlap would cost "столица Франции" its
-  answer, because the answer is Париж and that word is not in the question. See
-  `docs/evals/2026-08-05-search-quality-signals.md` (V-539). **Which query source claimed
-  a turn is readable on `/chat`** as a badge beside the reply, carried on
-  `ipc.ChatReply.Source` and noted by `noteQuerySource` in `cmd/mavend/querysource.go`. It
-  rides the context, so `handleText` keeps the one string signature the mic, telegram and
-  the web share.
-- **External search is allowed and off unless configured**, like the weather and telegram
-  capabilities. The code default is still off. `deploy/mavend.json` now ships a `search`
-  block (owner's call, 2026-08-02), so it is on for this box and deleting the block turns
-  it off again.
-- **The owner's notes and facts are never search input.** Looking up why the sky is blue and
-  sending the owner's stored personal notes to an upstream engine are different acts. Only the utterance goes
-  out, never the persona block, history, or matched notes.
-
-## Web UI conventions
-
-Server-rendered pages share `cmd/mavweb/static/ui.css` (served at `/ui.css`) and the shell
-partial in `cmd/mavweb/shell.html`: a page opens with `{{template "shellTop" "<page-key>"}}`
-and closes with `{{template "shellBottom"}}`, and the key marks the active sidebar link.
-Every page is its own embedded `.html` file next to `main.go` — no page markup lives in Go,
-and the sidebar is data (`sidebarSections`, `pageIcon`) the template renders. No
-per-page `<style>` beyond true one-offs. Wrap every table in `<div class=scroll>` so wide
-data pans on a phone. Local preview + headless screenshot recipe is in `AGENTS.md`.
-
-## Vikunja
-
-This repo is project **Maven** (ID 2) in Vikunja. MCP: `http://localhost:9100/mcp` (or
-`http://192.168.1.104:9100/mcp` from workpc). Feature/bug/deploy tasks go there.
-
-Vikunja is the durable task store. A task holds the goal, the constraints and the
-assumption ledger. Work without a task id is work nobody can resume, so a session that
-has no id asks for one before it starts.
-
-The MCP tool schemas are deferred, so load the four you actually use in ONE call at the
-start of a session rather than one lookup per first use:
-
-```text
-ToolSearch("select:mcp__vikunja__list_tasks,mcp__vikunja__get_task_details,mcp__vikunja__create_task,mcp__vikunja__update_task")
-```
-
-**Close a finished task with `done: true` and nothing else** (owner's call, 07-08-2026).
-Do not write a completion summary into the description on the way out. It is lost anyway,
-and the durable record is the commit messages and the merged PR. Note that `update_task`
-carrying a `description` resets `done` to false, which is why a write-up ever took two
-calls.
+- **No telemetry, no cloud model, no third-party account.** Inference stays on
+  the box and nothing about Maven is reported to anyone.
+- **The owner's data first, then the world.** Every source reading his facts,
+  notes, calendar, tasks or house runs first, and the personal boundary sits
+  between them and anything outside.
+- **His notes and facts are never search input.** Only the utterance goes out,
+  never the persona block, the history, or matched notes.
+- **External search is allowed and off unless configured.** Deleting the
+  `search` block in `deploy/mavend.json` turns it off.
+- **`Response.Empty()` is the whole gate** on a world answer. There is no
+  quality threshold in front of it and four candidate signals all failed.
 
 ## Session workflow
 
-`~/.local/bin/task` owns the branch, the commit identity and the PR. One task, one
-session, one PR.
+`docs/workflow.md` carries the five stores, the doc tiers and the guards. One
+task, one session, one PR. `/pickup` opens a session and `/wrap` closes it. Wrap
+at roughly half context rather than letting the session compact.
 
 ```sh
 task start <vikunja-id>   # branch off origin/master, write TASK.md, fetch review comments
 task pr                   # push, open or refresh the PR, label Vikunja, notify
-task comments             # re-pull this branch's review comments into .task/
+ToolSearch("select:mcp__vikunja__list_tasks,mcp__vikunja__get_task_details,mcp__vikunja__create_task,mcp__vikunja__update_task")
 ```
 
-Around that, `/pickup` opens a session and `/wrap` closes it. Wrap at roughly half
-context rather than letting the session compact.
-
-Five stores, and each one owns something the others must not hold:
-
-| Store | Holds | Lifetime |
-|---|---|---|
-| Vikunja task | goal, constraints, assumption ledger, status | durable |
-| `CLAUDE.md`, `AGENTS.md` | what an agent must know before touching code | durable |
-| `docs/` | design, measurements, decisions | durable |
-| `TASK.md` | the brief for this branch, written by `task start`, immutable | one branch |
-| `HANDOFF.md` | only what the next agent needs to resume | one session |
-
-`TASK.md` and `.task/` are excluded through `.git/info/exclude`. `HANDOFF.md` is
-gitignored and injected at session start. If a line in the handoff would still matter
-next week, it is in the wrong file.
-
-Docs are tiered by path, so staleness is visible from the filename. Files directly under
-`docs/` are living and carry a `Last verified: <date> @ <sha>` line. Files under
-`docs/evals/` are dated measurements and are never edited after the day, so a newer
-number is a new file. Files under `docs/archive/` are dead and read by nobody by default.
-
-## Git guards
-
-Two hooks in `.githooks/`, tracked, wired with `core.hooksPath`. Fresh clone:
-
-```sh
-git config core.hooksPath .githooks
-```
-
-- `pre-commit` refuses master, and refuses more than 300 changed lines in non-markdown
-  files. Markdown is exempt and may land as one batch.
-- `commit-msg` requires the subject to end with `(V-<id>)`. `V-` and not `#`, because
-  Gitea autolinks `#123` to a Gitea issue, which is the wrong tracker.
-
-Two more guards live outside the repo, in `~/.claude/hooks/`. `diff-budget.sh` blocks
-further edits past 600 changed lines on a `task/` branch. `prose_lint_hook.py` checks
-prose on every write. Both measure against `origin/master`, so a local master that is
-ahead of the remote makes the diff budget read high.
-
-`--no-verify` exists. Using it means saying why in the commit body.
+- This repo is Vikunja project **Maven** (ID 2), MCP at
+  `http://localhost:9100/mcp`, or `http://192.168.1.104:9100/mcp` from workpc.
+- **A session with no task id asks for one before it starts**, because work
+  without one is work nobody can resume.
+- **Close a finished task with `done: true` and nothing else** (owner's call,
+  2026-08-07). `update_task` carrying a `description` resets `done` to false.
+- **`pre-commit` refuses master** and more than 300 changed lines in
+  non-markdown files. Markdown is exempt and may land as one batch.
+- **`commit-msg` requires the subject to end with `(V-<id>)`.** `V-` and not
+  `#`, because Gitea autolinks `#123` to the wrong tracker.
+- **`diff-budget.sh` blocks edits past 600 changed lines** on a `task/` branch.
+- **`--no-verify` exists.** Using it means saying why in the commit body.

Makefile

@@ -4,7 +4,7 @@
 # `test` below fail on the two packages that have no test files. deps-go builds
 # the missing tools in, so the vendored tree is self-sufficient. Keep the version
 # here in step with the `go` directive in go.mod.
-GO_VERSION := 1.25.5
+GO_VERSION := 1.25.12
 GO       := $(shell pwd)/deps/go/go/bin/go
 export GOTOOLCHAIN := local
 GOFLAGS  :=
@@ -16,7 +16,7 @@ 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: t audit 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 deps-sentinel tidy eval-router eval-reach eval-recall eval-phrasing eval-models build-gpud
+.PHONY: t audit 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 deps-sentinel deps-vuln vuln deps-lint lint deadcode analyze tidy eval-router eval-reach eval-recall eval-phrasing eval-models build-gpud
 
 all: build
 
@@ -73,7 +73,7 @@ run-web: build-web
 # builds them on demand, but `go test -coverprofile` calls covdata through
 # base.Tool(), which only stats pkg/tool and exits. So build them in once here.
 GO_TARBALL := go$(GO_VERSION).linux-amd64.tar.gz
-GO_SHA256  := 9e9b755d63b36acf30c12a9a3fc379243714c1c6d3dd72861da637f336ebb35b
+GO_SHA256  := 234828b7a89e0e303d2556310ee549fbcf253d28de937bac3da13d6294262ac1
 deps-go: deps-sentinel
 	@mkdir -p deps/go
 	cd deps/go && curl -fLO 'https://go.dev/dl/$(GO_TARBALL)'
@@ -95,6 +95,70 @@ deps-sentinel:
 	@mkdir -p deps
 	@printf 'module github.com/kami/maven/deps\n\ngo 1.21\n' > deps/go.mod
 
+# vuln — the advisory gate the 2026-08-10 audit found missing (V-682). It reads
+# the published database over the network, so it is not part of `test`, which
+# has to pass on a box with no route out. Run it before a toolchain or
+# dependency bump lands, because that is what it grades: on 2026-08-11 the
+# pinned Go 1.25.5 and x/text 0.14.0 carried 20 reachable advisories and the
+# bumped pair carries none.
+#
+# govulncheck is a tool and not a dependency, so it is installed into deps/ like
+# the toolchain rather than added to go.mod. The version is pinned here for the
+# same reason GO_VERSION is: a gate that moves on its own is not a gate.
+GOVULNCHECK_VERSION := v1.6.0
+GOVULNCHECK := $(shell pwd)/deps/bin/govulncheck
+
+deps-vuln: deps-sentinel
+	@mkdir -p deps/bin
+	GOTOOLCHAIN=local GOBIN=$(shell pwd)/deps/bin \
+		$(GO) install golang.org/x/vuln/cmd/govulncheck@$(GOVULNCHECK_VERSION)
+
+# The CGO env is the same one `test` carries: govulncheck loads the packages,
+# and the four CGO daemons do not load without it.
+vuln: deps-vuln
+	CGO_CFLAGS="$(CGO_CFLAGS)" CGO_LDFLAGS="$(CGO_LDFLAGS)" LD_LIBRARY_PATH="$(shell pwd)/deps/lib" \
+		PATH="$(shell pwd)/deps/go/go/bin:$$PATH" GOTOOLCHAIN=local $(GOVULNCHECK) ./...
+
+# lint and deadcode — the other two analyzers the 2026-08-10 audit asked for
+# (V-694). They are not part of `test` for the same reason `vuln` is not: they
+# install over the network, and they are slow enough that a change to one Go
+# file should not pay for them.
+#
+# Neither reports zero, so neither fails on its own output. The accepted set
+# lives in scripts/analyzers/*.baseline and scripts/analyzer-gate.sh decides.
+# What is new fails, and so does a baseline entry whose finding is gone.
+#
+# deadcode runs with -test, so a test file is a root. Without it the report is
+# 172 lines, most of internal/router/eval, and none of it is a mistake.
+STATICCHECK_VERSION := v0.7.0
+DEADCODE_VERSION    := v0.48.0
+STATICCHECK := $(shell pwd)/deps/bin/staticcheck
+DEADCODE    := $(shell pwd)/deps/bin/deadcode
+
+deps-lint: deps-sentinel
+	@mkdir -p deps/bin
+	GOTOOLCHAIN=local GOBIN=$(shell pwd)/deps/bin \
+		$(GO) install honnef.co/go/tools/cmd/staticcheck@$(STATICCHECK_VERSION)
+	GOTOOLCHAIN=local GOBIN=$(shell pwd)/deps/bin \
+		$(GO) install golang.org/x/tools/cmd/deadcode@$(DEADCODE_VERSION)
+
+# Both load the packages, so both carry the CGO env `test` carries. Without it
+# the four CGO daemons do not load and the analyzer reports a build error
+# instead of a finding -- which analyzer-gate.sh fails on rather than filters.
+ANALYZER_ENV = CGO_CFLAGS="$(CGO_CFLAGS)" CGO_LDFLAGS="$(CGO_LDFLAGS)" \
+	LD_LIBRARY_PATH="$(shell pwd)/deps/lib" \
+	PATH="$(shell pwd)/deps/go/go/bin:$$PATH" GOTOOLCHAIN=local
+
+lint: deps-lint
+	@$(ANALYZER_ENV) $(STATICCHECK) ./... | scripts/analyzer-gate.sh staticcheck
+
+deadcode: deps-lint
+	@$(ANALYZER_ENV) $(DEADCODE) -test ./... | scripts/analyzer-gate.sh deadcode
+
+# Every static gate in one command. Not `check`, because it is not the thing to
+# run before a commit: vuln reads the network and all three are slow.
+analyze: lint deadcode vuln
+
 # Run the tidy the sentinel makes possible. Not part of `test`: it rewrites
 # go.mod, and a build target that edits the module file is a surprise.
 # vendor/ is committed, so a tidy that drops a requirement must be followed by

cmd/mavend/actions_query.go

@@ -13,6 +13,7 @@ import (
 	"github.com/kami/maven/internal/crawl"
 	"github.com/kami/maven/internal/decision"
 	"github.com/kami/maven/internal/ipc"
+	"github.com/kami/maven/internal/kiwix"
 	"github.com/kami/maven/internal/memory"
 	"github.com/kami/maven/internal/morning"
 	"github.com/kami/maven/internal/phraser"
@@ -79,6 +80,15 @@ type querySource struct {
 	// not named do not get to try. The lookups still run, because a named
 	// destination is evidence and not a promise.
 	guesses bool
+
+	// boundary — dropping this source widens what leaves the box, so only a
+	// literal pattern may do it (V-666, owner's call of 2026-08-09).
+	//
+	// Every other guesser costs an answer when it is wrongly taken off a turn.
+	// This one costs the rule that a question about him never reaches an
+	// upstream engine. A grammar read the words to name a destination. A model
+	// and a softmax both inferred one, and neither may spend that.
+	boundary bool
 }
 
 // querySources is the ordered chain actionQuery walks; first source to claim
@@ -156,7 +166,7 @@ var querySources = []querySource{
 	// below answers from the world's. A question about him that got this far
 	// has no answer in his data, and no outside source can supply one, so this
 	// stops the walk rather than let the encyclopedia and the model guess.
-	{name: "personal", answer: (*reactiveHandler).queryPersonal, dest: router.SourceRecall, guesses: true},
+	{name: "personal", answer: (*reactiveHandler).queryPersonal, dest: router.SourceRecall, guesses: true, boundary: true},
 	// The world, read live. Owner's ruling of 2026-08-02: a metasearch hit beats
 	// a frozen ZIM, so SearXNG asks before Kiwix does. Nothing of his is at
 	// stake by this point — the boundary above already stopped every question
@@ -198,12 +208,17 @@ var querySources = []querySource{
 // No destination named ⇒ the table exactly as written, which is what shipped
 // before the field existed. That is the floor. The classifier arm names
 // nothing, so a box whose model is down routes queries the way it always did.
-func queryWalk(dest router.Source) (walk, skipped []querySource) {
+// The personal boundary is the one exception, and anchored is what buys it
+// (V-666). A grammar matched a literal pattern to name the destination. The
+// routing heads and the resident model inferred one, and an inferred SourceWorld
+// takes the boundary off a question about him. That widens what is asked
+// upstream rather than costing a local answer, so those two keep it.
+func queryWalk(dest router.Source, anchored bool) (walk, skipped []querySource) {
 	if dest == router.SourceUnknown {
 		return querySources, nil
 	}
 	for _, s := range querySources {
-		if s.guesses && s.dest != dest {
+		if s.guesses && s.dest != dest && (anchored || !s.boundary) {
 			skipped = append(skipped, s)
 			continue
 		}
@@ -219,7 +234,7 @@ func (h *reactiveHandler) actionQuery(ctx context.Context, dec router.Decision)
 	// (V-564). Finish names everyone below the winner.
 	decision.Expect(ctx, decision.StageQuery, querySourceNames())
 	rec := decision.From(ctx)
-	walk, skipped := queryWalk(dec.Source)
+	walk, skipped := queryWalk(dec.Source, dec.SourceAnchored)
 	for _, src := range skipped {
 		rec.Note(decision.Claim{
 			Stage: decision.StageQuery, Claimant: src.name, Outcome: decision.NeverAsked,
@@ -912,6 +927,28 @@ func (h *reactiveHandler) queryKiwix(ctx context.Context, t *queryTurn) (string,
 		}
 	}
 
+	// The topic, not the sentence (V-668). Kiwix ranks by keyword overlap, so
+	// the question words outrank the one word that names the article: measured
+	// on 2026-08-09, "что такое TCP" returns "Перехват TCP-соединения" and
+	// "TCP" returns TCP. Only the verbatim path needs this. The rewriter
+	// already reduces a question to English keywords, and reducing twice would
+	// take the topic off the input it reads.
+	if verbatim {
+		if topic := kiwix.Topic(pattern); topic != "" {
+			// The article named exactly, before any ranking runs. A ZIM is
+			// addressable by title and a wrong title is a 404, so this either
+			// answers or costs one request that says nothing.
+			for _, cand := range kiwix.TitleCandidates(topic) {
+				page, err := h.kiwix.client.Article(ctxK, kiwix.TitlePath(book, cand), h.kiwix.runes)
+				if err == nil && page.Text != "" {
+					log.Printf("voice: kiwix: %q in %q → title hit %q", topic, book, page.Title)
+					return h.kiwixReply(ctx, t, page.Title, page.Text)
+				}
+			}
+			pattern = topic
+		}
+	}
+
 	hits, err := h.kiwix.client.Search(ctxK, pattern, book, h.kiwix.max)
 	if err != nil {
 		log.Printf("voice: kiwix: search %q: %v", pattern, err)
@@ -944,14 +981,18 @@ func (h *reactiveHandler) queryKiwix(ctx context.Context, t *queryTurn) (string,
 		}
 		page = crawl.Page{Title: top.Title, Text: top.Snippet}
 	}
-	// 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)
+	return h.kiwixReply(ctx, t, top.Title, page.Text)
+}
+
+// kiwixReply hands one article over the same way a note or a page is handed
+// over: context for the question he asked, not something to recite.
+func (h *reactiveHandler) kiwixReply(ctx context.Context, t *queryTurn, title, text string) (string, bool) {
+	snippet := title + "\n" + crawl.TrimRunes(text, h.kiwix.runes)
 	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.
-		return readBack(top.Title + " — " + page.Text), true
+		return readBack(title + " — " + text), true
 	}
 	return reply, true
 }

cmd/mavend/chat_degrade_test.go

@@ -19,7 +19,7 @@ func TestChatAnswersWithNoLlamaServer(t *testing.T) {
 	dead := llm.New("http://127.0.0.1:1", 500*time.Millisecond)
 	emb := router.NewHashEmbedder(1024)
 	h.recall.embedder = emb
-	h.router = buildRouter(emb, h.matcher, 0.55, pickLLMRouter(true, dead))
+	h.router = buildRouter(emb, h.matcher, 0.55, pickLLMRouter(true, dead), nil)
 	h.replier = newLLMReplier(dead, nil)
 
 	ctx := withDialogueID(context.Background(), dialogueIDFor(sourceText, "web"))

cmd/mavend/clarify.go

@@ -489,15 +489,19 @@ func (h *reactiveHandler) noteSuspended(ctx context.Context, q *dialogue.Pending
 	if !q.CanResume() {
 		h.clarifyStore.Delete(dialogueIDOf(ctx))
 		h.noteDropped(ctx)
-		log.Printf("voice: clarify — the question about %s stepped aside %d times; letting the request go", q.Missing[0], q.Suspends)
+		log.Printf("voice: clarify — letting the question about %s go: %d asides in a row, %d rides in all", q.Missing[0], q.Suspends, q.Rides)
 		return
 	}
 	q.Suspends++
+	// Rides is the same event counted without the reset (V-663). Incremented
+	// beside Suspends and never anywhere else, so the two cannot disagree about
+	// what happened, only about how much of it they remember.
+	q.Rides++
 	q.Asked = h.now()
 	h.clarifyStore.Put(dialogueIDOf(ctx), q)
 	rt.resume = question
 	rt.suspended = true
-	log.Printf("voice: clarify — is its own request; suspending the question about %s and resuming it in the same reply (suspend %d of %d)", q.Missing[0], q.Suspends, dialogue.MaxSuspends)
+	log.Printf("voice: clarify — is its own request; suspending the question about %s and resuming it in the same reply (suspend %d of %d, ride %d of %d)", q.Missing[0], q.Suspends, dialogue.MaxSuspends, q.Rides, dialogue.MaxRides)
 }
 
 // foldAnswerIntoUtterance appends an answered subject to the original words,
@@ -540,6 +544,11 @@ func (h *reactiveHandler) askRemainingGap(ctx context.Context, q *dialogue.Pendi
 	// Suspends is not carried, and by this point it is already zero: the answer
 	// path resets it (V-654). Left off the literal so the zero is stated where
 	// the struct is built, rather than inherited from a field nobody names.
+	//
+	// Rides IS carried, and that is the whole point of it (V-663). This is the
+	// same request under a second question, not a new one, so the turns it has
+	// already ridden still count against it. Dropping the field here is exactly
+	// the re-basing that let one question ride twenty-six replies.
 	h.clarifyStore.Put(dialogueIDOf(ctx), &dialogue.PendingQuestion{
 		Intent:      q.Intent,
 		Slots:       merged,
@@ -550,6 +559,7 @@ func (h *reactiveHandler) askRemainingGap(ctx context.Context, q *dialogue.Pendi
 		TTL:         clarifyTTL,
 		Attempts:    q.Attempts + 1,
 		MaxAttempts: q.MaxAttempts,
+		Rides:       q.Rides,
 	})
 	log.Printf("voice: clarify — one gap filled, still missing %s for intent=%s, asking again (attempt %d)", remaining[0], intent, q.Attempts+1)
 	return question, true

cmd/mavend/clarify_test.go

@@ -317,7 +317,7 @@ func TestClarifyExpiryIsAnnouncedAndWordsStillRoute(t *testing.T) {
 	h, _, now := newClarifyHandler(t)
 	emb := router.NewHashEmbedder(1024)
 	h.recall.embedder = emb
-	h.router = buildRouter(emb, h.matcher, 0.55, nil)
+	h.router = buildRouter(emb, h.matcher, 0.55, nil, nil)
 
 	if _, asked := h.askClarify(ctx, clarifyDec(router.IntentReminder, router.Slots{Text: "напомни"}, "напомни")); !asked {
 		t.Fatal("expected a question")
@@ -671,7 +671,7 @@ func TestUnresolvedActSaysItDoesNotKnowTheCommand(t *testing.T) {
 func newRoutingClarifyHandler(t *testing.T) (*reactiveHandler, *store.Store) {
 	t.Helper()
 	h, st, _ := newClarifyHandler(t)
-	h.router = buildRouter(router.NewHashEmbedder(1024), h.matcher, 0.55, nil)
+	h.router = buildRouter(router.NewHashEmbedder(1024), h.matcher, 0.55, nil, nil)
 	h.recall = recallWiring{embedder: router.NewHashEmbedder(1024), memStore: memory.NewInMemoryStore()}
 	return h, st
 }

cmd/mavend/decisiontrace_test.go

@@ -25,7 +25,7 @@ func traceHandler(t *testing.T, ring *decision.Ring) *reactiveHandler {
 	return &reactiveHandler{
 		api:       api,
 		recall:    recallWiring{embedder: emb, memStore: memory.NewInMemoryStore()},
-		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil),
+		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil, nil),
 		replier:   voice.NewStubReplier(),
 		now:       func() time.Time { return now },
 		dataStore: st,

cmd/mavend/dialogue_contract_test.go

@@ -171,7 +171,7 @@ func newDialogueHandler(t *testing.T) (*reactiveHandler, *store.Store, *time.Tim
 	// and never a coincidence (V-577, V-579). checkEnd refuses any reminder
 	// landing on it, and at 09:00 the row that answers "на 9" would trip that.
 	*now = time.Date(2026, 7, 31, 9, 17, 0, 0, time.UTC)
-	h.router = buildRouter(router.NewHashEmbedder(1024), h.matcher, 0.55, nil)
+	h.router = buildRouter(router.NewHashEmbedder(1024), h.matcher, 0.55, nil, nil)
 	h.recall = recallWiring{embedder: router.NewHashEmbedder(1024), memStore: memory.NewInMemoryStore()}
 	return h, st, now
 }

cmd/mavend/fact_subject_test.go

@@ -24,7 +24,7 @@ func TestApplyAction_FactCapture_QueuesEntityResolution(t *testing.T) {
 
 	emb := router.NewHashEmbedder(1024)
 	matcher := tool.NewMatcher(api)
-	rtr := buildRouter(emb, matcher, 0.55, nil)
+	rtr := buildRouter(emb, matcher, 0.55, nil, nil)
 
 	h := &reactiveHandler{
 		api:       api,

cmd/mavend/factgate_test.go

@@ -20,7 +20,7 @@ func newFactGateHandler(t *testing.T, now time.Time) (*reactiveHandler, ipc.Core
 	h := &reactiveHandler{
 		api:       api,
 		recall:    recallWiring{embedder: emb, memStore: memory.NewInMemoryStore()},
-		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil),
+		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil, nil),
 		replier:   voice.NewStubReplier(),
 		now:       func() time.Time { return now },
 		dataStore: st,

cmd/mavend/notefragment_test.go

@@ -45,7 +45,7 @@ func newNoteHandler(t *testing.T) (*reactiveHandler, *store.Store) {
 	h := &reactiveHandler{
 		api:       api,
 		recall:    recallWiring{embedder: emb, memStore: memory.NewInMemoryStore()},
-		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil),
+		router:    buildRouter(emb, tool.NewMatcher(api), 0.55, nil, nil),
 		replier:   voice.NewStubReplier(),
 		now:       func() time.Time { return now },
 		dataStore: st,

cmd/mavend/querywalk_test.go

@@ -10,7 +10,7 @@ import (
 // whose model is down names nothing, and naming nothing has to walk the chain
 // the way it walked before the field existed.
 func TestNoDestinationWalksTheWholeChain(t *testing.T) {
-	walk, skipped := queryWalk(router.SourceUnknown)
+	walk, skipped := queryWalk(router.SourceUnknown, false)
 	if len(skipped) != 0 {
 		t.Errorf("skipped %d sources with no destination named, want none", len(skipped))
 	}
@@ -32,15 +32,16 @@ func TestANamedDestinationSilencesTheOtherGuessers(t *testing.T) {
 		dest      router.Source
 		utterance string
 		silenced  string
+		anchored  bool // a stage 0 grammar named the destination
 	}{
-		{router.SourceWorld, "что такое TCP?", "weather"},
-		{router.SourceWorld, "сколько будет 17 на 23?", "weather"},
-		{router.SourceWorld, "кто такой Линус Торвальдс?", "personal"},
-		{router.SourceRecall, "какой у меня любимый язык?", "feeds"},
-		{router.SourceCalendar, "что в календаре на завтра?", "weather"},
+		{router.SourceWorld, "что такое TCP?", "weather", true},
+		{router.SourceWorld, "сколько будет 17 на 23?", "weather", true},
+		{router.SourceWorld, "кто такой Линус Торвальдс?", "personal", true},
+		{router.SourceRecall, "какой у меня любимый язык?", "feeds", false},
+		{router.SourceCalendar, "что в календаре на завтра?", "weather", true},
 	}
 	for _, c := range cases {
-		walk, skipped := queryWalk(c.dest)
+		walk, skipped := queryWalk(c.dest, c.anchored)
 		if inWalk(walk, c.silenced) {
 			t.Errorf("%q named %q: %q is still asked", c.utterance, c.dest, c.silenced)
 		}
@@ -56,7 +57,7 @@ func TestANamedDestinationSilencesTheOtherGuessers(t *testing.T) {
 // data first, then the world", and a destination a model wrote must not be able
 // to reverse it.
 func TestNamingTheWorldStillReadsHisDataFirst(t *testing.T) {
-	walk, _ := queryWalk(router.SourceWorld)
+	walk, _ := queryWalk(router.SourceWorld, true)
 	for _, look := range []string{"fact-by-key", "embed", "memory", "notes"} {
 		if !inWalk(walk, look) {
 			t.Errorf("%q was dropped; only the sources that guess may be dropped", look)
@@ -74,7 +75,7 @@ func TestNamingTheWorldStillReadsHisDataFirst(t *testing.T) {
 // makes "какой у меня любимый язык?" answer "не нашла у тебя такой записи"
 // rather than reaching SearXNG once nothing local had it.
 func TestNamingRecallKeepsTheBoundary(t *testing.T) {
-	walk, _ := queryWalk(router.SourceRecall)
+	walk, _ := queryWalk(router.SourceRecall, true)
 	if !inWalk(walk, "personal") {
 		t.Fatal("the personal boundary was skipped on a turn named for his own data")
 	}
@@ -83,12 +84,33 @@ func TestNamingRecallKeepsTheBoundary(t *testing.T) {
 	}
 }
 
+// The owner's call of 2026-08-09 (V-666): only a stage 0 grammar may take the
+// personal boundary off a turn. The routing heads and the resident model both
+// name a destination by inference, and an inferred SourceWorld would send a
+// question about him upstream. Every other guesser still goes.
+func TestOnlyAGrammarMayDropTheBoundary(t *testing.T) {
+	walk, skipped := queryWalk(router.SourceWorld, false)
+	if !inWalk(walk, "personal") {
+		t.Error("an inferred destination took the boundary off the turn")
+	}
+	if !inWalk(skipped, "weather") {
+		t.Error("weather is still asked; the rule covers the boundary alone")
+	}
+	if posOf(walk, "personal") > posOf(walk, "search") {
+		t.Error("the boundary no longer sits in front of the world")
+	}
+	if anchored, _ := queryWalk(router.SourceWorld, true); inWalk(anchored, "personal") {
+		t.Error(`a grammar named the world and the boundary stayed: ` +
+			`"кто такой Линус Торвальдс?" is answered "не нашла у тебя такой записи" again`)
+	}
+}
+
 // Whatever the destination, the walk is a subsequence of the table. Every
 // comment on that table argues an order between two sources, and none of those
 // reasons is about this field.
 func TestTheWalkNeverReordersTheTable(t *testing.T) {
 	for _, dest := range append([]router.Source{router.SourceUnknown}, router.Sources...) {
-		walk, skipped := queryWalk(dest)
+		walk, skipped := queryWalk(dest, true)
 		if len(walk)+len(skipped) != len(querySources) {
 			t.Errorf("%q: %d walked + %d skipped, want %d", dest, len(walk), len(skipped), len(querySources))
 		}

cmd/mavend/reactive_notes_test.go

@@ -22,7 +22,7 @@ func TestReactiveNotesReminders(t *testing.T) {
 
 	emb := router.NewHashEmbedder(1024)
 	matcher := tool.NewMatcher(api)
-	rtr := buildRouter(emb, matcher, 0.55, nil)
+	rtr := buildRouter(emb, matcher, 0.55, nil, nil)
 
 	h := &reactiveHandler{
 		api:       api,
@@ -104,7 +104,7 @@ func TestSpokenTaskCaptureFilesATask(t *testing.T) {
 	h := &reactiveHandler{
 		api:       api,
 		recall:    recallWiring{embedder: emb, memStore: memory.NewInMemoryStore()},
-		router:    buildRouter(emb, matcher, 0.55, nil),
+		router:    buildRouter(emb, matcher, 0.55, nil, nil),
 		replier:   voice.NewStubReplier(),
 		now:       func() time.Time { return now },
 		dataStore: st,

cmd/mavend/simulator_test.go

@@ -474,7 +474,7 @@ func newSimWorld(t *testing.T, sc scenario) *simWorld {
 	// used to be built on a nil API, which meant any scenario that produced an
 	// act panicked the moment the matcher was consulted.
 	matcher := tool.NewMatcher(api)
-	rtr := buildRouter(emb, matcher, config.DefaultRouterThreshold, router.NewLLMRouter(scripted))
+	rtr := buildRouter(emb, matcher, config.DefaultRouterThreshold, router.NewLLMRouter(scripted), nil)
 
 	w.handler = &reactiveHandler{
 		stt:    simTranscriber{},

cmd/mavend/sttseam_test.go

@@ -0,0 +1,43 @@
+package main
+
+import (
+	"testing"
+
+	"github.com/kami/maven/internal/config"
+	"github.com/kami/maven/internal/stt"
+)
+
+// A box with no workstation.stt block transcribes exactly as it did before the
+// seam existed: the floor is handed back untouched, and nothing probes.
+func TestSttSeamWithNoBlockIsTheFloor(t *testing.T) {
+	floor := stt.NewStub()
+	got, pair := sttSeam(&config.Config{}, floor)
+	if pair != nil {
+		t.Fatal("no block must build no pair")
+	}
+	if got != stt.Transcriber(floor) {
+		t.Fatal("no block must hand back the floor itself")
+	}
+}
+
+func TestSttSeamPrefersTheWorkstation(t *testing.T) {
+	cfg := &config.Config{Workstation: &config.WorkstationConfig{
+		URL: "http://127.0.0.1:1",
+		Stt: &config.WorkstationSttConfig{
+			URL:    "http://127.0.0.1:2/transcribe",
+			Health: "http://127.0.0.1:2/health",
+		},
+	}}
+	got, pair := sttSeam(cfg, stt.NewStub())
+	if pair == nil {
+		t.Fatal("a configured block must build a pair")
+	}
+	defer pair.Stop()
+	if got != stt.Transcriber(pair) {
+		t.Fatal("the pair is what callers must transcribe through")
+	}
+	// Nothing answers on port 2, so the seam is the floor until it does.
+	if pair.Available() {
+		t.Fatal("an unreachable workstation must not be available")
+	}
+}

cmd/mavend/turnrole.go

@@ -186,6 +186,28 @@ func carriesReminderVerb(text string) bool {
 	return false
 }
 
+// isPleasantry matches the WHOLE utterance against lexicon.Pleasantries, after
+// lowercasing and dropping the punctuation a greeting carries.
+//
+// Whole utterance and not tokens. Every token rule tried here was wrong on
+// something: "вечер" answers "это утра или вечера?", "нет" answers a confirm,
+// and "спокойной" alone is not an utterance at all. A greeting is a fixed
+// phrase, so matching it as one costs nothing and claims nothing else.
+func isPleasantry(text string) bool {
+	t := strings.ToLower(strings.TrimSpace(text))
+	t = strings.Trim(t, " .,!?…")
+	t = strings.Join(strings.Fields(t), " ")
+	if t == "" {
+		return false
+	}
+	for _, p := range lexicon.Pleasantries() {
+		if t == p {
+			return true
+		}
+	}
+	return false
+}
+
 // offlineOwnRequest is the shape half of the evidence: the offline token tests,
 // which cost nothing and never depend on the model that produced the routing.
 // It is also the whole answer when there is no route to read — the classifier
@@ -225,6 +247,16 @@ func classifyTurnRole(q *dialogue.PendingQuestion, text string, answer dialogue.
 	// hour, and no route saying "question" changes that. It works because the
 	// extractor no longer reads a day word as the current clock, so a sentence
 	// that names no hour now fills nothing to weigh.
+	// A pleasantry is neither (V-663). "спасибо" and "привет" fell through to
+	// roleAnswer, so a question about a reminder's DAY was re-asked at a man
+	// saying thank you, and the retry it spent was one of the three bounds
+	// meant to end the ride. It is an aside: answered as itself, the question
+	// resumed on the tail, no attempt spent, one ride counted. Placed above the
+	// content gate because "доброе утро" has content and states nothing, so
+	// neither half of the evidence below can reach it.
+	if q != nil && isPleasantry(text) {
+		return roleAside
+	}
 	own := false
 	if len(ownContent(text)) > 0 {
 		own = offlineOwnRequest(text) || (ok && carriesOwnRequest(routed, text))

cmd/mavend/turnrole_test.go

@@ -372,4 +372,69 @@ func TestAnAnsweredGapResetsTheSuspendBudget(t *testing.T) {
 	if q.Suspends != 0 {
 		t.Fatalf("answering a gap must reset the suspend budget: suspends = %d", q.Suspends)
 	}
+	// The ride it already took is carried across the re-park (V-663). Resetting
+	// both counters here is what let one question ride twenty-six replies.
+	if q.Rides != 1 {
+		t.Fatalf("the aside it already took was forgotten: rides = %d", q.Rides)
+	}
+}
+
+// TestTwoBoundsCannotRearmEachOther — V-663.
+//
+// MaxSuspends landed and the measurement did not move: twenty-six of 140 turns
+// carried a tail before it and twenty-six after. This is the shape it misses,
+// taken from the 2026-08-08 run, where one question rode turns 7 to 13.
+//
+// An aside spends no attempt, so MaxAttempts never reaches it. A turn that
+// reads as a failed answer zeroes Suspends, so MaxSuspends never reaches the
+// asides either. Alternating the two rearms each bound with the other's
+// traffic. Rides counts both kinds and is never reset, so it is what ends this.
+func TestTwoBoundsCannotRearmEachOther(t *testing.T) {
+	ctx := context.Background()
+	h, _ := newRoutingClarifyHandler(t)
+	id := dialogueIDFor(sourceText, "web")
+	resumed, _ := clarifyResumedFor(dialogue.SlotTime)
+
+	if reply := h.handleText(ctx, "web", "напомни позвонить маме"); !strings.Contains(reply, "?") {
+		t.Fatalf("expected the time question, got %q", reply)
+	}
+
+	// Two asides. Each one rides and neither spends an attempt.
+	for i := 0; i < 2; i++ {
+		reply := h.handleText(ctx, "web", "какие у меня напоминания?")
+		if !strings.HasSuffix(reply, resumed) {
+			t.Fatalf("aside %d: the question must come back, got %q", i+1, reply)
+		}
+	}
+	q := h.clarifyStore.Get(id, h.now())
+	if q == nil || q.Rides != 2 || q.Suspends != 2 {
+		t.Fatalf("after two asides: %+v", q)
+	}
+
+	// A pleasantry. It used to read as a failed answer, so she re-asked the
+	// question at a man saying thank you and spent an attempt doing it. Now it
+	// is an aside: answered as itself, question on the tail, one more ride.
+	reply := h.handleText(ctx, "web", "спасибо")
+	if !strings.HasSuffix(reply, resumed) {
+		t.Fatalf("a pleasantry lost the parked question: %q", reply)
+	}
+	q = h.clarifyStore.Get(id, h.now())
+	if q == nil || q.Attempts != 1 {
+		t.Fatalf("a pleasantry spent an attempt: %+v", q)
+	}
+	if q.Rides != 3 {
+		t.Fatalf("a pleasantry rode free: %+v", q)
+	}
+
+	// One more ride of any kind and the request goes, out loud.
+	reply = h.handleText(ctx, "web", "какие у меня напоминания?")
+	if !strings.Contains(reply, clarifyDropped) {
+		t.Fatalf("the question rode four asides and was let go in silence: %q", reply)
+	}
+	if strings.HasSuffix(reply, resumed) {
+		t.Fatalf("a question she has let go must not be asked again: %q", reply)
+	}
+	if h.clarifyStore.Get(id, h.now()) != nil {
+		t.Fatal("the question must be gone once she has said she let it go")
+	}
 }

cmd/mavend/voicewire.go

@@ -36,7 +36,9 @@ type voiceWiring struct {
 	sessions  *voice.Sessions
 	voiceSink delivery.Sink
 	embedder  router.Embedder
-	handler   *reactiveHandler // the reactive handler for IPC Chat
+	// heads — the routing heads, nil unless embedder.heads_path is set.
+	heads   *router.RouterHeads
+	handler *reactiveHandler // the reactive handler for IPC Chat
 	// worker clients (set when configured as Remote): closed on shutdown so
 	// mavsttd / mavttsd don't keep a stale conn into a restarting daemon.
 	sttClient *worker.Client
@@ -53,7 +55,11 @@ type voiceWiring struct {
 	// 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
+	// sttPair — CrisperWhisper 2.0 on the workstation with mavsttd as the
+	// floor, nil unless the `workstation.stt` block names an address. Held for
+	// the same reason as pair: to stop its prober on shutdown.
+	sttPair *stt.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.
@@ -72,6 +78,9 @@ func (w *voiceWiring) close() {
 	if w.embedder != nil {
 		_ = w.embedder.Close()
 	}
+	if w.heads != nil {
+		_ = w.heads.Close()
+	}
 	if w.server != nil {
 		_ = w.server.Close()
 	}
@@ -84,6 +93,9 @@ func (w *voiceWiring) close() {
 	if w.pair != nil {
 		w.pair.Stop()
 	}
+	if w.sttPair != nil {
+		w.sttPair.Stop()
+	}
 	w.mcp.close()
 }
 
@@ -112,6 +124,7 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 	} else {
 		transcriber = stt.NewStub()
 	}
+	transcriber, w.sttPair = sttSeam(cfg, transcriber)
 	w.transcriber = transcriber
 
 	// ----- tts (Stub in-process OR Remote) -----
@@ -147,6 +160,24 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 		emb = router.NewHashEmbedder(1024)
 	}
 	w.embedder = emb
+
+	// ----- router: routing heads (only when configured, and never fatal) -----
+	// A missing or broken weights file logs and leaves w.heads nil, which is
+	// byte-for-byte the cascade that shipped before V-664. Refusing to start
+	// over a routing accelerator would trade a working box for a better one.
+	if cfg.Voice.Embedder != nil && cfg.Voice.Embedder.HeadsPath != "" {
+		h, err := router.NewRouterHeads(
+			cfg.Voice.Embedder.HeadsPath,
+			cfg.Voice.Embedder.TokenizerPath,
+		)
+		if err != nil {
+			log.Printf("voice: routing heads unavailable, cascade unchanged: %v", err)
+		} else {
+			log.Printf("voice: routing heads loaded from %s", cfg.Voice.Embedder.HeadsPath)
+			w.heads = h
+		}
+	}
+
 	repairFactVectors(dataStore, emb)
 	checkStoredEmbedder(dataStore, emb)
 	// Retention is enforced on write, which is not enough on its own: a box that
@@ -223,7 +254,8 @@ 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(), hot))
+	rtr := buildRouter(emb, matcher, threshold,
+		pickLLMRouter(cfg.Voice.UseLLMRouter(), hot), w.heads)
 
 	// ----- sessions registry (shared with voicesink) -----
 	sessions := voice.NewSessions()
@@ -354,8 +386,13 @@ func modelSeam(cfg *config.Config, resident *llm.Client) (router.Completer, *llm
 		return resident, nil
 	}
 	ws := cfg.Workstation
+	remote := llm.New(ws.URL, time.Duration(ws.Timeout))
+	remote.SetToken(ws.Token)
+	if ws.Token == "" {
+		log.Printf("voice: no workstation.token — mavgpud refuses an unauthenticated request, so this reads as a card that is always busy")
+	}
 	pair := llm.NewPair(
-		llm.New(ws.URL, time.Duration(ws.Timeout)),
+		remote,
 		resident,
 		ws.Health,
 		time.Duration(ws.Probe),
@@ -366,6 +403,44 @@ func modelSeam(cfg *config.Config, resident *llm.Client) (router.Completer, *llm
 	return pair, pair
 }
 
+// sttSeam builds the transcription seam the voice path and the meeting
+// recorder share. It is modelSeam for audio and follows the same rule.
+//
+// With no `workstation.stt` block it hands back the floor untouched, which is
+// today's deploy exactly. With one, it is an stt.Pair preferring CrisperWhisper
+// 2.0 on workpc, which scores 10.4% WER in Russian against the floor's 27.5%
+// (docs/evals/2026-08-09-crisperwhisper2-russian-wer.md).
+//
+// Only the silent half of the degradation rule applies here. A worse transcript
+// is still a turn, so there is nothing to name a gap about and the fallback is
+// never spoken. That is why stt.Pair has no TranscribeRemote.
+func sttSeam(cfg *config.Config, floor stt.Transcriber) (stt.Transcriber, *stt.Pair) {
+	if cfg.Workstation == nil || cfg.Workstation.Stt == nil {
+		return floor, nil
+	}
+	s := cfg.Workstation.Stt
+	lang := ""
+	if cfg.Voice != nil {
+		lang = cfg.Voice.Lang
+		if cfg.Voice.Stt != nil && cfg.Voice.Stt.Lang != "" {
+			lang = cfg.Voice.Stt.Lang
+		}
+	}
+	pair := stt.NewPair(
+		stt.NewHTTPTranscriber(s.URL, s.Token, lang, time.Duration(s.Timeout)),
+		floor,
+		s.Health,
+		time.Duration(s.Probe),
+	)
+	pair.Start(context.Background())
+	if s.Token == "" {
+		log.Print("voice: the workstation transcriber has no token, so anything on the LAN can post audio to it")
+	}
+	log.Printf("voice: workstation transcriber at %s, probed every %s, mavsttd as the floor",
+		s.URL, time.Duration(s.Probe))
+	return pair, pair
+}
+
 func pickLLMRouter(enabled bool, c router.Completer) *router.LLMRouter {
 	if !enabled {
 		return nil
@@ -390,48 +465,13 @@ func pickLLMRouter(enabled bool, c router.Completer) *router.LLMRouter {
 //     intent from seedDir (models/seeds/<intent>.txt) — see seedClassifier
 //     below for the current intent list and file names.
 //   - Threshold is from voice.router_threshold config (default 0.55).
-func buildRouter(emb router.Embedder, acts router.ActMatcher, threshold float64, llmR *router.LLMRouter) *router.Router {
+func buildRouter(emb router.Embedder, acts router.ActMatcher, threshold float64,
+	llmR *router.LLMRouter, heads *router.RouterHeads) *router.Router {
 	cls := router.NewClassifier(emb)
 	seedClassifier(cls)
-	grammars := router.DefaultGrammars(acts)
-	grammars = append(grammars, router.SystemTimeDateGrammars()...)
-	// After the time/date rules on purpose: "какой сегодня день" is a clock
-	// question and must keep reaching replySystem, while "что у меня сегодня"
-	// is an agenda question and must not.
-	grammars = append(grammars, router.AgendaQueryGrammars()...)
-	// Same reason as the agenda rules, for the feeds: "что нового в лентах?"
-	// routed system and answered "пока не умею" (Vikunja #474).
-	// After the agenda rules, which are the narrower claim, and BEFORE the feed
-	// and list rules, which are not: "что такое лента" is a definition question
-	// and the feed rule would take it on the noun alone (V-655).
-	grammars = append(grammars, router.WorldQueryGrammars()...)
-	grammars = append(grammars, router.FeedQueryGrammar())
-	// The list side of the same exposure: a phrasing with no possessive in it
-	// ("список дел") routed system and never reached queryTasks (Vikunja #467).
-	grammars = append(grammars, router.TaskListGrammar())
-	grammars = append(grammars, router.ListGrammars()...)
-	grammars = append(grammars, router.ReminderGrammar())
-	// Before the capture marker, because "отметь" is a capture verb and "отметь
-	// второй пункт" is not a note. The Praxis rules are the narrower claim — a
-	// lifecycle verb AND an item named — so they get first refusal (Vikunja #516).
-	grammars = append(grammars, router.PraxisGrammars()...)
-	// Last, and it matches any utterance shape — its Build is the filter. An
-	// explicit capture marker beats the model, which called it an act and
-	// rewrote the task text (Vikunja #467). After the rules above because a
-	// marker never collides with a clock or agenda question.
-	// After Praxis, whose bare "закрой" claim this rule cannot reach (it needs the
-	// board noun), and before the capture marker, which would otherwise read
-	// "убери из задач купить молоко" as a new task (Vikunja #512).
-	grammars = append(grammars, router.TaskStatusGrammar())
-	// Before the capture markers, which all need an object. A capture verb
-	// alone is a fact with no key, and the clarify path asks for it rather than
-	// letting the model invent an answer (Vikunja #557).
-	grammars = append(grammars, router.BareCaptureGrammar()...)
-	grammars = append(grammars, router.TaskCaptureGrammar())
-	// After the capture marker, so "запиши" still wins over "расскажи", and
-	// last overall because it matches on the first word alone: "расскажи про
-	// X" is a world question the model called a fact (Vikunja #498).
-	grammars = append(grammars, router.NarrativeQueryGrammars()...)
+	// The stage 0 set, in the router package, so the eval fixture runs the rules
+	// the daemon runs (V-693). Order and reasoning live with the list.
+	grammars := router.StageZeroGrammars(acts)
 	return router.New(router.Config{
 		Grammars:   grammars,
 		Classifier: cls,
@@ -442,6 +482,7 @@ func buildRouter(emb router.Embedder, acts router.ActMatcher, threshold float64,
 		},
 		Threshold: threshold,
 		LLM:       llmR,
+		Heads:     heads,
 	})
 }
 

cmd/mavgpud/auth.go

@@ -0,0 +1,145 @@
+package main
+
+import (
+	"crypto/sha256"
+	"crypto/subtle"
+	"fmt"
+	"net"
+	"net/http"
+	"os"
+	"strings"
+)
+
+// The boundary in front of the card.
+//
+// mavgpud has to listen on the LAN, because homesrv is the client and a
+// loopback default takes the model arm down. That makes this the one hop on the
+// workstation anything on the network could reach, and until 2026-08-11 it
+// reverse-proxied every path to llama-server unauthenticated: any client could
+// spend the card, hold the model resident by touching the idle clock, and read
+// /slots, which carries the prompts of whoever else was using it.
+//
+// So: a bearer token every request must carry, read from a file, and a path
+// allowlist so a token that leaks buys the model API and not the admin one. The
+// CW2 transcriber beside this daemon has worked this way since it shipped; this
+// is the same arrangement, not a new one.
+
+// readToken loads the bearer token. The file holds the token and nothing else,
+// trailing newline allowed. A path that is set and unreadable is fatal to the
+// caller: a supervisor that silently ran without its boundary is the failure
+// this exists to prevent.
+func readToken(path string) (string, error) {
+	b, err := os.ReadFile(path)
+	if err != nil {
+		return "", fmt.Errorf("token_file: %w", err)
+	}
+	tok := strings.TrimSpace(string(b))
+	if tok == "" {
+		return "", fmt.Errorf("token_file %s is empty", path)
+	}
+	return tok, nil
+}
+
+// loopbackListen reports whether addr can only be reached from this machine.
+// An empty or wildcard host is not loopback, which is the case that matters:
+// ":8080" is the shipped default and it answers the whole LAN.
+func loopbackListen(addr string) bool {
+	host, _, err := net.SplitHostPort(addr)
+	if err != nil {
+		host = addr
+	}
+	host = strings.Trim(host, "[]")
+	if host == "" {
+		return false
+	}
+	if host == "localhost" {
+		return true
+	}
+	ip := net.ParseIP(host)
+	return ip != nil && ip.IsLoopback()
+}
+
+// allowed is what a token buys. Everything llama-server exposes beyond this is
+// refused, because the endpoints Maven does not call are the expensive ones to
+// hand out: /slots returns other callers' prompts, and its save/restore actions
+// write files chosen by the request.
+//
+// Adding a caller means adding its path here. That is deliberate — the list is
+// short because Maven's use of the workstation is.
+var allowed = map[string]string{
+	"/v1/chat/completions": http.MethodPost,
+	"/v1/completions":      http.MethodPost,
+	"/v1/embeddings":       http.MethodPost,
+	"/v1/models":           http.MethodGet,
+	"/props":               http.MethodGet,
+}
+
+// requireToken authenticates, then bounds. Order matters: an unauthenticated
+// client must not be able to make this daemon allocate a body buffer.
+//
+// /health is not exempt. It reports whether the card is loaded and free, which
+// is exactly what someone deciding whether to take it from him would ask.
+func requireToken(token string, maxBody int64, next http.Handler) http.Handler {
+	want := sha256.Sum256([]byte(token))
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		got := sha256.Sum256([]byte(bearer(r)))
+		if subtle.ConstantTimeCompare(got[:], want[:]) != 1 {
+			w.Header().Set("WWW-Authenticate", "Bearer")
+			http.Error(w, "unauthorized", http.StatusUnauthorized)
+			return
+		}
+		r.Body = http.MaxBytesReader(w, r.Body, maxBody)
+		next.ServeHTTP(w, r)
+	})
+}
+
+// bearer pulls the credential out of the header. A malformed header yields the
+// empty string, which fails the comparison like any other wrong token — there
+// is no separate error for it, because telling a caller *how* it was wrong is
+// the only thing a probe learns from a 401.
+func bearer(r *http.Request) string {
+	h := r.Header.Get("Authorization")
+	const prefix = "Bearer "
+	if len(h) <= len(prefix) || !strings.EqualFold(h[:len(prefix)], prefix) {
+		return ""
+	}
+	return strings.TrimSpace(h[len(prefix):])
+}
+
+// allowlist refuses a path the model arm does not use. It answers 404 rather
+// than 403 so a scan cannot map llama-server's surface through this hop.
+func allowlist(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		method, ok := allowed[r.URL.Path]
+		if !ok {
+			http.NotFound(w, r)
+			return
+		}
+		if r.Method != method {
+			w.Header().Set("Allow", method)
+			http.Error(w, "method not allowed", http.StatusMethodNotAllowed)
+			return
+		}
+		next.ServeHTTP(w, r)
+	})
+}
+
+// limitInflight caps concurrent proxied requests. A waiter leaves when its own
+// context ends, so a client that gave up does not keep a slot: llama-server
+// runs with -np 1 and queueing here is cheaper than queueing inside the child
+// with a body held in memory on both sides.
+func limitInflight(n int, next http.Handler) http.Handler {
+	if n <= 0 {
+		return next
+	}
+	slots := make(chan struct{}, n)
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		select {
+		case slots <- struct{}{}:
+			defer func() { <-slots }()
+			next.ServeHTTP(w, r)
+		case <-r.Context().Done():
+			http.Error(w, "client went away", http.StatusServiceUnavailable)
+		}
+	})
+}

cmd/mavgpud/auth_test.go

@@ -0,0 +1,180 @@
+package main
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"path/filepath"
+	"strings"
+	"sync"
+	"testing"
+)
+
+// ok is what the boundary is protecting: anything that reaches it has spent
+// the card.
+func ok(w http.ResponseWriter, r *http.Request) { w.WriteHeader(http.StatusTeapot) }
+
+func TestRequireTokenRefusesEveryWrongCredential(t *testing.T) {
+	h := requireToken("s3cret", 1<<20, http.HandlerFunc(ok))
+	cases := []struct {
+		name string
+		auth string
+		want int
+	}{
+		{"no header", "", http.StatusUnauthorized},
+		{"wrong token", "Bearer wrong", http.StatusUnauthorized},
+		{"prefix of the token", "Bearer s3cre", http.StatusUnauthorized},
+		{"token with no scheme", "s3cret", http.StatusUnauthorized},
+		{"basic auth", "Basic czNjcmV0", http.StatusUnauthorized},
+		{"right token", "Bearer s3cret", http.StatusTeapot},
+		{"scheme is case-insensitive", "bearer s3cret", http.StatusTeapot},
+	}
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			r := httptest.NewRequest(http.MethodGet, "/health", nil)
+			if tc.auth != "" {
+				r.Header.Set("Authorization", tc.auth)
+			}
+			w := httptest.NewRecorder()
+			h.ServeHTTP(w, r)
+			if w.Code != tc.want {
+				t.Errorf("status %d, want %d", w.Code, tc.want)
+			}
+		})
+	}
+}
+
+// The 401 must not say which part was wrong. A probe that can tell a malformed
+// header from a wrong token learns the header shape for free.
+func TestUnauthorizedSaysNothingUseful(t *testing.T) {
+	h := requireToken("s3cret", 1<<20, http.HandlerFunc(ok))
+	w := httptest.NewRecorder()
+	h.ServeHTTP(w, httptest.NewRequest(http.MethodGet, "/health", nil))
+	if got := strings.TrimSpace(w.Body.String()); got != "unauthorized" {
+		t.Errorf("body %q, want %q", got, "unauthorized")
+	}
+	if got := w.Header().Get("WWW-Authenticate"); got != "Bearer" {
+		t.Errorf("WWW-Authenticate %q, want Bearer", got)
+	}
+}
+
+// The body cap applies to an authenticated request. An unauthenticated one
+// never gets far enough to allocate anything.
+func TestRequireTokenCapsTheBody(t *testing.T) {
+	var read error
+	h := requireToken("t", 8, http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		buf := make([]byte, 64)
+		for read == nil {
+			if _, read = r.Body.Read(buf); read != nil {
+				break
+			}
+		}
+	}))
+	r := httptest.NewRequest(http.MethodPost, "/v1/chat/completions", strings.NewReader(strings.Repeat("x", 4096)))
+	r.Header.Set("Authorization", "Bearer t")
+	h.ServeHTTP(httptest.NewRecorder(), r)
+	if read == nil || !strings.Contains(read.Error(), "too large") {
+		t.Errorf("read error %v, want the body cap", read)
+	}
+}
+
+func TestAllowlistRefusesWhatMavenDoesNotCall(t *testing.T) {
+	h := allowlist(http.HandlerFunc(ok))
+	cases := []struct {
+		method, path string
+		want         int
+	}{
+		{http.MethodPost, "/v1/chat/completions", http.StatusTeapot},
+		{http.MethodGet, "/v1/models", http.StatusTeapot},
+		// /slots returns the prompts of whoever else is using the card, and
+		// its actions write files the request names.
+		{http.MethodGet, "/slots", http.StatusNotFound},
+		{http.MethodPost, "/slots/0?action=save", http.StatusNotFound},
+		{http.MethodGet, "/", http.StatusNotFound},
+		{http.MethodGet, "/v1/chat/completions", http.StatusMethodNotAllowed},
+	}
+	for _, tc := range cases {
+		t.Run(tc.method+" "+tc.path, func(t *testing.T) {
+			w := httptest.NewRecorder()
+			h.ServeHTTP(w, httptest.NewRequest(tc.method, tc.path, nil))
+			if w.Code != tc.want {
+				t.Errorf("status %d, want %d", w.Code, tc.want)
+			}
+		})
+	}
+}
+
+func TestLimitInflightCapsConcurrency(t *testing.T) {
+	const cap = 2
+	var mu sync.Mutex
+	now, peak := 0, 0
+	release := make(chan struct{})
+	h := limitInflight(cap, http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		mu.Lock()
+		now++
+		if now > peak {
+			peak = now
+		}
+		mu.Unlock()
+		<-release
+		mu.Lock()
+		now--
+		mu.Unlock()
+	}))
+
+	var wg sync.WaitGroup
+	for i := 0; i < 8; i++ {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			h.ServeHTTP(httptest.NewRecorder(), httptest.NewRequest(http.MethodPost, "/v1/chat/completions", nil))
+		}()
+	}
+	// Let the first wave arrive, then drain. The assertion is the peak, and a
+	// peak that never reached the cap still cannot exceed it.
+	close(release)
+	wg.Wait()
+	if peak > cap {
+		t.Errorf("%d requests in flight at once, cap is %d", peak, cap)
+	}
+}
+
+func TestLoopbackListen(t *testing.T) {
+	cases := map[string]bool{
+		":8080":              false, // the shipped default, and the whole LAN
+		"0.0.0.0:8080":       false,
+		"[::]:8080":          false,
+		"192.168.1.105:8080": false,
+		"127.0.0.1:8080":     true,
+		"[::1]:8080":         true,
+		"localhost:8080":     true,
+	}
+	for addr, want := range cases {
+		if got := loopbackListen(addr); got != want {
+			t.Errorf("loopbackListen(%q) = %v, want %v", addr, got, want)
+		}
+	}
+}
+
+func TestReadToken(t *testing.T) {
+	dir := t.TempDir()
+	good := filepath.Join(dir, "tok")
+	if err := os.WriteFile(good, []byte("  abc123\n"), 0o600); err != nil {
+		t.Fatal(err)
+	}
+	got, err := readToken(good)
+	if err != nil || got != "abc123" {
+		t.Errorf("readToken = %q, %v; want abc123", got, err)
+	}
+
+	blank := filepath.Join(dir, "blank")
+	if err := os.WriteFile(blank, []byte("\n\n"), 0o600); err != nil {
+		t.Fatal(err)
+	}
+	if _, err := readToken(blank); err == nil {
+		t.Error("an empty token file is not a token")
+	}
+	if _, err := readToken(filepath.Join(dir, "absent")); err == nil {
+		t.Error("a missing token file is not a token")
+	}
+}

cmd/mavgpud/gpu.go

@@ -34,22 +34,31 @@ type probe struct {
 	drmDev  string
 }
 
-// foreign lists every ROCm process that is not ours. selfPID is the supervisor's
-// llama-server child, or 0 when it is not running.
+// foreign lists every ROCm process that is not ours. self holds the pids of the
+// supervisor's own children, and a child that is not running contributes 0.
+//
+// There is more than one child since 09-08-2026. CW2 registers on the KFD like
+// any ROCm job, so a supervisor that excluded only llama-server would read its
+// own transcriber as a contender, yield the card to it, and never keep a model
+// loaded again.
 //
 // An unreadable kfd tree returns no processes and no error. That is deliberate
 // and it is the safe direction only because startVRAM also has to agree before
 // anything launches: a supervisor that cannot see the KFD never sees free VRAM
 // either, because the CPT run holding the card shows up in the drm totals.
-func (p probe) foreign(selfPID int) []gpuProc {
+func (p probe) foreign(self ...int) []gpuProc {
 	entries, err := os.ReadDir(p.kfdRoot)
 	if err != nil {
 		return nil
 	}
+	mine := make(map[int]bool, len(self))
+	for _, pid := range self {
+		mine[pid] = true
+	}
 	var out []gpuProc
 	for _, e := range entries {
 		pid, err := strconv.Atoi(e.Name())
-		if err != nil || pid == selfPID {
+		if err != nil || mine[pid] {
 			continue
 		}
 		out = append(out, gpuProc{

cmd/mavgpud/gpu_test.go

@@ -1,6 +1,7 @@
 package main
 
 import (
+	"context"
 	"net/http"
 	"net/http/httptest"
 	"net/url"
@@ -8,6 +9,7 @@ import (
 	"path/filepath"
 	"strconv"
 	"testing"
+	"time"
 )
 
 // fakeKFD builds the sysfs shape the workstation actually has: one directory
@@ -47,6 +49,24 @@ func TestForeignExcludesOurChild(t *testing.T) {
 	}
 }
 
+// The transcriber is a ROCm process on the same card, so it registers on the
+// KFD exactly like a contender does. Reading it as one is what happened on
+// 2026-08-09 while CW2 ran under its own systemd unit: mavgpud yielded, waited
+// five polls, loaded the model, yielded again, and never held it for a whole
+// minute. Excluding every child is the fix and this is the test of it.
+func TestForeignExcludesEveryChild(t *testing.T) {
+	p := probe{kfdRoot: fakeKFD(t, map[int]int64{478104: 12791693312, 999: 4096, 1001: 1717986918})}
+
+	ours := p.foreign(999, 1001)
+	if len(ours) != 1 || ours[0].PID != 478104 {
+		t.Fatalf("only the CPT run is a contender, got %+v", ours)
+	}
+	// A child that is not running reports pid 0, which must exclude nothing.
+	if got := p.foreign(999, 0); len(got) != 2 {
+		t.Errorf("a stopped child excludes nobody: got %d contenders, want 2", len(got))
+	}
+}
+
 // An empty KFD tree is the state that permits a start, so it must read as empty
 // rather than as an error the caller has to interpret.
 func TestForeignEmptyAndMissing(t *testing.T) {
@@ -81,12 +101,14 @@ func TestFreeVRAM(t *testing.T) {
 // rather than hanging or proxying into a closed port. Maven reads this endpoint
 // on a timer forever, including while the workstation is busy.
 func TestHealthAndProxyRefuseWhenNotReady(t *testing.T) {
-	s := &supervisor{run: newRunner("/bin/true", nil, "")}
+	s := &supervisor{run: newRunner("fake", "/bin/true", nil, "")}
 	h := s.handler(mustURL(t, "http://127.0.0.1:1"))
 
-	for _, path := range []string{"/health", "/v1/chat/completions"} {
+	// The completion is a POST because the allowlist is in front of the
+	// readiness check now, and it answers 405 to a method it never serves.
+	for path, method := range map[string]string{"/health": http.MethodGet, "/v1/chat/completions": http.MethodPost} {
 		w := httptest.NewRecorder()
-		h.ServeHTTP(w, httptest.NewRequest(http.MethodGet, path, nil))
+		h.ServeHTTP(w, httptest.NewRequest(method, path, nil))
 		if w.Code != http.StatusServiceUnavailable {
 			t.Errorf("%s with no model: got %d, want 503", path, w.Code)
 		}
@@ -101,3 +123,28 @@ func mustURL(t *testing.T, s string) *url.URL {
 	}
 	return u
 }
+
+// Yielding is all or nothing. A CPT run wants the whole card, so handing back
+// the language model while the transcriber keeps 1.6GB mapped would leave the
+// other job failing its allocation, which is the outcome yielding exists to
+// prevent.
+func TestYieldStopsEveryChild(t *testing.T) {
+	idle := "while : ; do sleep 1 ; done"
+	s := &supervisor{
+		cfg:   config{EvictAfter: 1, StopGrace: duration(2 * time.Second)},
+		probe: probe{kfdRoot: fakeKFD(t, map[int]int64{478104: 12791693312})},
+		run:   newRunner("llama-server", fakeServer(t, idle), nil, ""),
+		stt:   newRunner("cw2", fakeServer(t, idle), nil, ""),
+	}
+	for _, r := range s.children() {
+		if err := r.start(); err != nil {
+			t.Fatal(err)
+		}
+	}
+	s.tick(context.Background())
+	for _, r := range s.children() {
+		if r.running() {
+			t.Errorf("%s outlived the yield", r.name)
+		}
+	}
+}

cmd/mavgpud/main.go

@@ -10,6 +10,11 @@
 // the card. Not on demand, because a 7-14B takes tens of seconds to load and a
 // world question would be answered by a gap every time the card had been quiet.
 // Not always on, because that holds 16GB against the owner's own jobs.
+//
+// It supervises a second child since 09-08-2026, the CW2 transcriber, and for
+// one reason only: it is a ROCm process on the same card. Any GPU service the
+// owner leaves running beside this daemon reads as a contender and evicts the
+// model, so the card needs one owner rather than two neighbours.
 package main
 
 import (
@@ -31,11 +36,30 @@ type config struct {
 	Listen    string `json:"listen"`     // what Maven talks to
 	LlamaAddr string `json:"llama_addr"` // where llama-server binds
 	LlamaBin  string `json:"llama_bin"`
+
+	// TokenFile holds the bearer token every request must carry. It is a path
+	// and never the token itself, the rule mavpoll, mavmaild and the CW2
+	// transcriber already follow: a secret in a committed config is a secret
+	// in the history. Empty is allowed only on a loopback Listen, and
+	// requireToken is where that is decided.
+	TokenFile string `json:"token_file,omitempty"`
+
+	// MaxBody bounds a proxied request body. A completion is a prompt, and a
+	// prompt that does not fit here would not fit the context window either.
+	MaxBody int64 `json:"max_body_bytes,omitempty"`
+	// MaxInflight bounds how many proxied requests reach llama-server at once.
+	// It runs with -np 1, so anything above a handful only queues inside the
+	// child while holding a connection and a body in memory here.
+	MaxInflight int `json:"max_inflight,omitempty"`
 	// LlamaArgs must include the flags that bind LlamaAddr. They are passed
 	// through untouched so the model, context size and layer count stay the
 	// owner's business and not this daemon's schema.
 	LlamaArgs []string `json:"llama_args"`
 
+	// Stt is optional. Without it mavgpud supervises llama-server alone, which
+	// is everything it did before 09-08-2026.
+	Stt *sttConfig `json:"stt,omitempty"`
+
 	KFDRoot   string `json:"kfd_root"`
 	DRMDevice string `json:"drm_device"`
 
@@ -51,6 +75,22 @@ type config struct {
 	StartAfter int `json:"start_after_polls"`
 }
 
+// sttConfig is the CW2 transcriber, which mavgpud runs for one reason: it is a
+// ROCm process on this card. Left to its own systemd unit it registers on the
+// KFD, the supervisor reads it as a contender, and llama-server is evicted
+// within two polls and restarted five polls later, forever. That thrash was
+// observed on 2026-08-09 and it is what folded the service in here.
+//
+// Maven talks to it directly, not through this daemon. There is no proxy and no
+// idle timer: at 1.6GB it denies the card to nobody, and unloading it would only
+// send the next voice turn to the homesrv floor for no gain.
+type sttConfig struct {
+	// Addr is where the service binds, and it is read only to probe /health.
+	Addr string   `json:"addr"`
+	Bin  string   `json:"bin"`
+	Args []string `json:"args"`
+}
+
 func defaults() config {
 	return config{
 		Listen:      ":8080",
@@ -63,6 +103,8 @@ func defaults() config {
 		MinFreeVRAM: 15 << 30,
 		EvictAfter:  2,
 		StartAfter:  5,
+		MaxBody:     8 << 20,
+		MaxInflight: 4,
 	}
 }
 
@@ -98,13 +140,33 @@ func main() {
 		log.Fatal("mavgpud: llama_bin is required")
 	}
 
+	// A LAN listener with no token is refused rather than downgraded to
+	// loopback. Downgrading would look like a safe default and would take the
+	// model arm down instead: homesrv is the client and it is on the LAN.
+	var token string
+	if cfg.TokenFile != "" {
+		var err error
+		if token, err = readToken(cfg.TokenFile); err != nil {
+			log.Fatalf("mavgpud: %v", err)
+		}
+	} else if !loopbackListen(cfg.Listen) {
+		log.Fatalf("mavgpud: listen %s is reachable from the network and token_file is unset — "+
+			"set token_file, or listen on 127.0.0.1 and accept that Maven cannot reach it", cfg.Listen)
+	}
+
 	base := "http://" + cfg.LlamaAddr
-	run := newRunner(cfg.LlamaBin, cfg.LlamaArgs, base+"/health")
+	run := newRunner("llama-server", cfg.LlamaBin, cfg.LlamaArgs, base+"/health")
 	sup := &supervisor{
 		cfg:   cfg,
 		probe: probe{kfdRoot: cfg.KFDRoot, drmDev: cfg.DRMDevice},
 		run:   run,
 	}
+	if s := cfg.Stt; s != nil {
+		if s.Bin == "" || s.Addr == "" {
+			log.Fatal("mavgpud: stt needs both bin and addr")
+		}
+		sup.stt = newRunner("cw2", s.Bin, s.Args, "http://"+s.Addr+"/health")
+	}
 	sup.touch()
 
 	ctx, cancel := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
@@ -114,7 +176,20 @@ func main() {
 	if err != nil {
 		log.Fatalf("mavgpud: llama_addr: %v", err)
 	}
-	srv := &http.Server{Addr: cfg.Listen, Handler: sup.handler(target)}
+	var h http.Handler = sup.handler(target)
+	if token != "" {
+		h = requireToken(token, cfg.MaxBody, h)
+	}
+	srv := &http.Server{
+		Addr:    cfg.Listen,
+		Handler: h,
+		// A slow-loris client holds a connection and a header buffer for free
+		// otherwise. No ReadTimeout or WriteTimeout: a completion legitimately
+		// takes minutes on this card, and either one would cut it off.
+		ReadHeaderTimeout: 10 * time.Second,
+		IdleTimeout:       60 * time.Second,
+		MaxHeaderBytes:    1 << 16,
+	}
 	go func() {
 		log.Printf("mavgpud: listening on %s, model %s", cfg.Listen, cfg.LlamaBin)
 		if err := srv.ListenAndServe(); err != nil && err != http.ErrServerClosed {
@@ -129,13 +204,17 @@ func main() {
 	shut, done := context.WithTimeout(context.Background(), 5*time.Second)
 	defer done()
 	_ = srv.Shutdown(shut)
-	run.stop(time.Duration(cfg.StopGrace))
+	for _, r := range sup.children() {
+		r.stop(time.Duration(cfg.StopGrace))
+	}
 }
 
 type supervisor struct {
 	cfg   config
 	probe probe
 	run   *runner
+	// stt is the CW2 transcriber, or nil when the config names none.
+	stt *runner
 
 	lastReq atomic.Int64 // unix nanos of the last request Maven sent
 
@@ -168,14 +247,14 @@ func (s *supervisor) handler(target *url.URL) http.Handler {
 		w.Header().Set("Content-Type", "application/json")
 		_, _ = w.Write([]byte(`{"status":"ok"}`))
 	})
-	mux.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
+	mux.Handle("/", allowlist(limitInflight(s.cfg.MaxInflight, http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
 		if !s.run.isReady() {
 			http.Error(w, "model not loaded", http.StatusServiceUnavailable)
 			return
 		}
 		s.touch()
 		proxy.ServeHTTP(w, r)
-	})
+	}))))
 	return mux
 }
 
@@ -198,7 +277,11 @@ func (s *supervisor) loop(ctx context.Context) {
 // allocates, so we see a contender during its startup rather than after it has
 // already failed to get the memory it wanted.
 func (s *supervisor) tick(ctx context.Context) {
-	others := s.probe.foreign(s.run.pid())
+	var pids []int
+	for _, r := range s.children() {
+		pids = append(pids, r.pid())
+	}
+	others := s.probe.foreign(pids...)
 	if len(others) > 0 {
 		s.foreignStreak++
 		s.clearStreak = 0
@@ -207,31 +290,65 @@ func (s *supervisor) tick(ctx context.Context) {
 		s.clearStreak++
 	}
 
-	if s.run.running() {
-		s.run.refreshReady(ctx)
-		switch {
-		case s.foreignStreak >= s.cfg.EvictAfter:
-			log.Printf("mavgpud: yielding the card to %s", describe(others))
-			s.run.stop(time.Duration(s.cfg.StopGrace))
-		case s.idle() > time.Duration(s.cfg.IdleTimeout):
-			log.Printf("mavgpud: idle for %s, unloading", s.idle().Round(time.Second))
-			s.run.stop(time.Duration(s.cfg.StopGrace))
+	// Yielding is all or nothing. A CPT run wants the whole card, and handing
+	// back 8GB while holding 1.6GB is the shape of a failed allocation.
+	if s.foreignStreak >= s.cfg.EvictAfter && s.anyRunning() {
+		log.Printf("mavgpud: yielding the card to %s", describe(others))
+		for _, r := range s.children() {
+			r.stop(time.Duration(s.cfg.StopGrace))
 		}
 		return
 	}
 
-	if s.clearStreak < s.cfg.StartAfter {
+	clear := s.clearStreak >= s.cfg.StartAfter
+
+	if s.run.running() {
+		s.run.refreshReady(ctx)
+		if s.idle() > time.Duration(s.cfg.IdleTimeout) {
+			log.Printf("mavgpud: idle for %s, unloading", s.idle().Round(time.Second))
+			s.run.stop(time.Duration(s.cfg.StopGrace))
+		}
+	} else if clear && s.probe.freeVRAM() >= s.cfg.MinFreeVRAM {
+		s.touch() // the idle clock starts at load, not at the last request before it
+		if err := s.run.start(); err != nil {
+			log.Printf("mavgpud: start llama-server: %v", err)
+		}
+	}
+
+	if s.stt == nil {
 		return
 	}
-	if free := s.probe.freeVRAM(); free < s.cfg.MinFreeVRAM {
+	if s.stt.running() {
+		s.stt.refreshReady(ctx)
 		return
 	}
-	s.touch() // the idle clock starts at load, not at the last request before it
-	if err := s.run.start(); err != nil {
-		log.Printf("mavgpud: start llama-server: %v", err)
+	// No VRAM precondition here, unlike llama-server. That check exists because
+	// a 12B refuses to load when the card is short, and 1.6GB fits wherever the
+	// KFD is clear. Reading free VRAM would also block the transcriber for good
+	// once the language model was resident, since it holds more than the floor.
+	if clear {
+		if err := s.stt.start(); err != nil {
+			log.Printf("mavgpud: start cw2: %v", err)
+		}
 	}
 }
 
+func (s *supervisor) children() []*runner {
+	if s.stt == nil {
+		return []*runner{s.run}
+	}
+	return []*runner{s.run, s.stt}
+}
+
+func (s *supervisor) anyRunning() bool {
+	for _, r := range s.children() {
+		if r.running() {
+			return true
+		}
+	}
+	return false
+}
+
 // describe names the contenders in the log. This log is the instrument for the
 // open question in #488: whether polling the KFD misses a job that wants the
 // card without registering there.

cmd/mavgpud/runner.go

@@ -10,14 +10,18 @@ import (
 	"time"
 )
 
-// runner owns one llama-server process. Owning it is the point of the daemon:
-// the workstation cannot keep a 7-14B resident, because that holds 16GB against
+// runner owns one GPU process. Owning it is the point of the daemon: the
+// workstation cannot keep a 7-14B resident, because that holds 16GB against
 // the owner's CPT runs, Correx and the manga-recap pipeline. So the thing that
 // stays up is this, which costs no VRAM, and the model comes and goes under it.
+//
+// There are two of them since 09-08-2026: llama-server and the CW2 transcriber.
+// name is what the log calls this one.
 type runner struct {
+	name string
 	bin  string
 	args []string
-	// ready is llama-server's own /health, which answers "is a model loaded".
+	// ready is the child's own /health, which answers "is a model loaded".
 	// Loading a 7-14B takes tens of seconds, so started is not ready.
 	readyURL string
 
@@ -32,9 +36,9 @@ type runner struct {
 	http     *http.Client
 }
 
-func newRunner(bin string, args []string, readyURL string) *runner {
+func newRunner(name, bin string, args []string, readyURL string) *runner {
 	return &runner{
-		bin: bin, args: args, readyURL: readyURL,
+		name: name, bin: bin, args: args, readyURL: readyURL,
 		http: &http.Client{Timeout: 2 * time.Second},
 	}
 }
@@ -60,7 +64,7 @@ func (r *runner) isReady() bool {
 	return r.ready
 }
 
-// start launches llama-server. It returns as soon as the process exists, not
+// start launches the child. It returns as soon as the process exists, not
 // when the model is loaded.
 func (r *runner) start() error {
 	r.mu.Lock()
@@ -76,7 +80,7 @@ func (r *runner) start() error {
 		return err
 	}
 	r.cmd, r.ready, r.yielding = cmd, false, false
-	log.Printf("mavgpud: started llama-server pid=%d", cmd.Process.Pid)
+	log.Printf("mavgpud: started %s pid=%d", r.name, cmd.Process.Pid)
 	go func() {
 		err := cmd.Wait()
 		r.mu.Lock()
@@ -84,15 +88,15 @@ func (r *runner) start() error {
 		r.cmd, r.ready, r.yielding = nil, false, false
 		r.mu.Unlock()
 		if yielded {
-			log.Printf("mavgpud: llama-server stopped, card yielded (%v)", err)
+			log.Printf("mavgpud: %s stopped, card yielded (%v)", r.name, err)
 			return
 		}
-		log.Printf("mavgpud: llama-server exited: %v", err)
+		log.Printf("mavgpud: %s exited: %v", r.name, err)
 	}()
 	return nil
 }
 
-// stop ends llama-server and waits for the VRAM to come back. SIGTERM first so
+// stop ends the child and waits for the VRAM to come back. SIGTERM first so
 // it unmaps cleanly, SIGKILL after the grace window. Returning before the
 // process is gone would let the supervisor report a free card while 14GB is
 // still mapped, which is the one lie that would make yielding useless.
@@ -117,11 +121,11 @@ func (r *runner) stop(grace time.Duration) {
 		}
 		time.Sleep(100 * time.Millisecond)
 	}
-	log.Printf("mavgpud: llama-server did not exit in %s, killing", grace)
+	log.Printf("mavgpud: %s did not exit in %s, killing", r.name, grace)
 	_ = syscall.Kill(pgid, syscall.SIGKILL)
 }
 
-// refreshReady asks llama-server whether the model is loaded. Called once per
+// refreshReady asks the child whether the model is loaded. Called once per
 // supervisor tick, never per request.
 func (r *runner) refreshReady(ctx context.Context) {
 	if !r.running() {
@@ -141,6 +145,6 @@ func (r *runner) refreshReady(ctx context.Context) {
 	r.ready = ok
 	r.mu.Unlock()
 	if ok && !was {
-		log.Printf("mavgpud: model ready")
+		log.Printf("mavgpud: %s ready", r.name)
 	}
 }

cmd/mavgpud/runner_test.go

@@ -25,7 +25,7 @@ func fakeServer(t *testing.T, body string) string {
 // 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, "")
+	r := newRunner("fake", fakeServer(t, "while : ; do sleep 1 ; done"), nil, "")
 	if err := r.start(); err != nil {
 		t.Fatalf("start: %v", err)
 	}
@@ -49,7 +49,7 @@ func TestStopMarksTheExitAsAYield(t *testing.T) {
 // 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 := newRunner("fake", "/nonexistent", nil, "")
 	r.stop(10 * time.Millisecond)
 	r.mu.Lock()
 	defer r.mu.Unlock()

cmd/mavwaked/main.go

@@ -5,12 +5,24 @@
 // is detected sends it as a PushToTalk frame to the voice server. The reply
 // audio is played back through aplay(1).
 //
-// No wake-word model yet (MVP uses voice-activity-only trigger). The
-// SurfaceVoice auth layer caps all commands at L0 (no destructive acts),
-// making accidental triggers safe by design. A proper wake-word engine
-// (openWakeWord / Silero VAD ONNX) is the planned upgrade — the VAD shape
-// (30ms frames, 16kHz PCM) matches silero-vad's input interface exactly, so
-// swapping energy-threshold for ONNX-inference is a local change in vad.go.
+// Voice activity is silero-vad when -vad-model points at the graph, and an
+// energy threshold when it does not. Silero declines noise the threshold
+// accepts: 0 frames against 68 to 99 on the four fixtures, measured in
+// docs/evals/2026-08-09-silero-vad.md. Note that the model window is 512
+// samples and the capture frame is 480, so silero.go re-chunks. This comment
+// used to say the two matched, which was true of silero v4.
+//
+// The keyword is "Мэйвен" and it is required, when -wake-model points at the
+// head (V-487 stage two). Without it anything spoken near the microphone
+// becomes a turn, which the SurfaceVoice auth layer makes safe rather than
+// expensive: it caps all commands at L0, no destructive acts. It does not cap
+// reading, so an open gate still lets the room hear his facts read back.
+// wakeword.go holds the cadence and wakefeatures.go the three models.
+//
+// The conn carries both directions. mavwaked sends utterances and receives
+// proactive nudges on it, and it is opened at startup rather than at the first
+// utterance, because mavend registers a voice session on accept. See nudge.go
+// for why a nudge that is not heard is worse than one that is not delivered.
 //
 // While a reply is playing the capture side is muted (half-duplex): without
 // it, Maven's own voice comes back in through the mic and she answers
@@ -52,6 +64,12 @@ const (
 	defaultAddr     = "127.0.0.1:9100"
 	defaultLang     = "ru"
 	defaultReadSize = 4096 // max PCM bytes per read from arecord (fits multiple frames)
+
+	// defaultWakeWindowMs — how long the keyword stays good for. He says
+	// "Мэйвен" and then a sentence, and the VAD does not close the utterance
+	// until he stops, so this has to outlive the word by the length of what
+	// follows it. It is spent on dispatch: one keyword, one turn.
+	defaultWakeWindowMs = 8000
 )
 
 func main() {
@@ -73,17 +91,38 @@ func run(args []string) error {
 	bargeIn := flag.Bool("barge-in", false, "cut Maven off when he talks over her (needs a room-tuned -barge-in-rms)")
 	bargeRMS := flag.Int("barge-in-rms", defaultBargeRMS, "RMS x10000 a frame must clear to count as barge-in")
 	bargeFrames := flag.Int("barge-in-frames", defaultBargeFrames, "consecutive frames over -barge-in-rms before playback is cut")
+	vadModel := flag.String("vad-model", "", "silero-vad onnx file; empty runs the energy threshold instead")
+	vadThreshold := flag.Float64("vad-threshold", defaultSileroThreshold, "speech probability a frame must clear")
+	onnxLib := flag.String("onnx-lib", os.Getenv("MAVEN_ONNX_LIB"), "libonnxruntime.so, needed with -vad-model")
+	wakeModel := flag.String("wake-model", "", "keyword head onnx; empty ships every utterance, as before V-487")
+	wakeMel := flag.String("wake-mel", "", "melspectrogram.onnx, required with -wake-model")
+	wakeEmbed := flag.String("wake-embed", "", "embedding_model.onnx, required with -wake-model")
+	wakeThreshold := flag.Float64("wake-threshold", defaultWakeThreshold, "score the keyword must clear")
+	wakeWindowMs := flag.Int("wake-window-ms", defaultWakeWindowMs, "ms an utterance may still start after the keyword")
 	flag.CommandLine.Parse(args)
 
 	ctx, stop := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM, syscall.SIGHUP)
 	defer stop()
 
-	// Voice client — reused across utterances; SendRequest reconnects on error.
+	// Voice client — one conn carrying both directions. SendRequest reconnects
+	// on error, and the push receiver redials on its own clock.
 	vc := voice.Dial(*addr)
 	defer vc.Close()
 
-	// VAD engine.
+	// VAD engine. A model that will not load is logged and not fatal: the
+	// energy threshold is worse, and it is a great deal better than a
+	// listening client that refuses to start.
 	vad := NewVAD(*minRMS, *speechMs, *silenceMs, *maxMs)
+	if *vadModel != "" {
+		s, err := newSileroVAD(*vadModel, *onnxLib)
+		if err != nil {
+			log.Printf("mavwaked: silero unavailable, energy threshold unchanged: %v", err)
+		} else {
+			defer s.Close()
+			vad.UseSilero(s, *vadThreshold)
+			log.Printf("mavwaked: silero-vad from %s, threshold %.2f", *vadModel, *vadThreshold)
+		}
+	}
 
 	// Audio source.
 	var src io.ReadCloser
@@ -145,6 +184,29 @@ func run(args []string) error {
 	}
 	sess := newSession(vad, newAplayPlayer(), &voiceSender{vc: vc}, *lang, barge)
 
+	// Keyword gate. A model that will not load is logged and not fatal, for
+	// the same reason silero's is not: an open gate is the daemon he had
+	// yesterday, and a daemon that refuses to start is not.
+	if *wakeModel != "" {
+		w, err := newWakeWord(*wakeMel, *wakeEmbed, *wakeModel, *onnxLib, *wakeThreshold)
+		if err != nil {
+			log.Printf("mavwaked: wake word unavailable, every utterance is a turn: %v", err)
+		} else {
+			defer w.Close()
+			sess.UseWakeWord(w, time.Duration(*wakeWindowMs)*time.Millisecond)
+			log.Printf("mavwaked: wake word from %s, threshold %.3f, window %dms",
+				*wakeModel, *wakeThreshold, *wakeWindowMs)
+		}
+	}
+
+	// Listen for nudges alongside capture. Connect eagerly so mavend has a
+	// voice session before he has said anything: without one, a nudge routed
+	// to voice finds nobody home and goes to the away channels instead.
+	if err := vc.Connect(ctx); err != nil {
+		log.Printf("mavwaked: voice server not reachable yet, retrying in background: %v", err)
+	}
+	go runNudgeReceiver(ctx, vc, sess)
+
 	return captureLoop(ctx, src, sess)
 }
 

cmd/mavwaked/nudge.go

@@ -0,0 +1,79 @@
+package main
+
+// The receiving half of the voice reach (V-671).
+//
+// mavwaked used to send and never listen. It wired no PushHandler, and
+// SendRequest discards a push frame when there is none. The consequence was
+// not a missing feature but a silent one: mavend routes a nudge to the voice
+// session that spoke most recently, and once mavwaked had spoken once it WAS
+// that session. PushToMostRecent succeeded, the dispatcher counted the nudge
+// delivered and stopped rerouting to telegram and ntfy, and mavwaked threw the
+// audio away. He heard nothing, anywhere.
+//
+// So the connection is opened at startup rather than at the first utterance,
+// and it is held open. A client that has never connected has no session, and
+// the dispatcher must be able to tell "he is not at the machine" from "he is,
+// and she has nothing to say".
+
+import (
+	"context"
+	"encoding/json"
+	"log"
+	"time"
+
+	"github.com/kami/maven/internal/voice"
+)
+
+// nudgeRetry is how long to wait before dialling again after the conn ends.
+// mavend restarts on every deploy, and a listener that gives up then is a
+// listener that is deaf until the next reboot.
+const nudgeRetry = 5 * time.Second
+
+// nudgeHandler decodes a push and hands the audio to the session, which
+// speaks it through the same player the reply path uses. It does not play
+// anything itself: the half-duplex gate and barge-in live on the capture
+// loop, and a nudge has to sit under both.
+type nudgeHandler struct{ sess *session }
+
+func (h *nudgeHandler) OnPush(p voice.Push) {
+	if p.Kind != voice.PushKindAudioNudge {
+		log.Printf("mavwaked: ignoring push of unknown kind %q", p.Kind)
+		return
+	}
+	var ap voice.AudioNudgePush
+	if err := json.Unmarshal(p.Params, &ap); err != nil {
+		log.Printf("mavwaked: nudge: decode: %v", err)
+		return
+	}
+	log.Printf("mavwaked: nudge from rule %q (severity %d): %q (%.2fs audio)",
+		ap.RuleName, ap.Severity, ap.Text, ap.Audio.Duration())
+	if len(ap.Audio.Bytes) == 0 {
+		// mavttsd was down or the text was empty. Say so rather than going
+		// quiet: the dispatcher already counted this one as delivered.
+		log.Printf("mavwaked: nudge %q carried no audio, nothing to speak", ap.RuleName)
+		return
+	}
+	h.sess.Nudge(ap.Audio)
+}
+
+// runNudgeReceiver keeps a push handler wired for as long as ctx lives,
+// redialling whenever the conn ends. Returns when ctx is cancelled.
+func runNudgeReceiver(ctx context.Context, vc *voice.Client, sess *session) {
+	h := &nudgeHandler{sess: sess}
+	for {
+		err := vc.RunPushReceiver(ctx, h)
+		if ctx.Err() != nil {
+			return
+		}
+		if err != nil {
+			log.Printf("mavwaked: nudge receiver: %v", err)
+		} else {
+			log.Printf("mavwaked: voice connection ended, reconnecting in %s", nudgeRetry)
+		}
+		select {
+		case <-ctx.Done():
+			return
+		case <-time.After(nudgeRetry):
+		}
+	}
+}

cmd/mavwaked/nudge_test.go

@@ -0,0 +1,170 @@
+package main
+
+// The receiving half: a nudge pushed by mavend has to reach the speaker, and
+// it has to obey the same two gates a reply obeys (V-671).
+
+import (
+	"context"
+	"encoding/json"
+	"testing"
+	"time"
+
+	"github.com/kami/maven/internal/audio"
+	"github.com/kami/maven/internal/voice"
+)
+
+func nudgeAudio() audio.Audio {
+	return audio.Audio{Format: audio.PCM16kMono, Bytes: make([]byte, 8000)}
+}
+
+// pushFrame builds the frame mavend's voicesink sends.
+func pushFrame(t *testing.T, a audio.Audio) voice.Push {
+	t.Helper()
+	body, err := json.Marshal(voice.AudioNudgePush{
+		RuleName: "test-rule",
+		Severity: 3,
+		Audio:    a,
+		Text:     "пора пить воду",
+		Ts:       time.Unix(0, 0),
+	})
+	if err != nil {
+		t.Fatalf("marshal push: %v", err)
+	}
+	return voice.Push{Kind: voice.PushKindAudioNudge, Params: body}
+}
+
+// The defect itself: the push arrived and nothing came out of the speaker.
+func TestNudgeReachesThePlayer(t *testing.T) {
+	sess, p, snd := newTestSession(bargeInConfig{})
+	(&nudgeHandler{sess: sess}).OnPush(pushFrame(t, nudgeAudio()))
+
+	if p.plays != 0 {
+		t.Fatal("nudge played from the push goroutine; it must wait for the capture loop")
+	}
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != 1 {
+		t.Fatalf("plays = %d, want 1", p.plays)
+	}
+	if len(p.last.Bytes) != 8000 {
+		t.Errorf("played %d bytes, want the nudge audio", len(p.last.Bytes))
+	}
+	if sess.nudges != 1 {
+		t.Errorf("nudges = %d, want 1", sess.nudges)
+	}
+	if len(snd.sent) != 0 {
+		t.Errorf("a nudge must not be shipped back to the daemon as an utterance")
+	}
+}
+
+// A push of some other kind, or one carrying no audio, must not reach the
+// player and must not wedge the one that follows.
+func TestNudgeIgnoresUnusablePushes(t *testing.T) {
+	sess, p, _ := newTestSession(bargeInConfig{})
+	h := &nudgeHandler{sess: sess}
+
+	h.OnPush(voice.Push{Kind: "something-else", Params: json.RawMessage(`{}`)})
+	h.OnPush(voice.Push{Kind: voice.PushKindAudioNudge, Params: json.RawMessage(`not json`)})
+	h.OnPush(pushFrame(t, audio.Audio{Format: audio.PCM16kMono}))
+
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != 0 {
+		t.Fatalf("plays = %d, want 0", p.plays)
+	}
+
+	h.OnPush(pushFrame(t, nudgeAudio()))
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != 1 {
+		t.Fatalf("plays after a usable nudge = %d, want 1", p.plays)
+	}
+}
+
+// The half-duplex gate covers a nudge exactly as it covers a reply: she does
+// not start one over herself, and the mic stays muted while it runs.
+func TestNudgeWaitsForTheReplyToFinish(t *testing.T) {
+	sess, p, _ := newTestSession(bargeInConfig{})
+	speakThenPause(t, sess)
+	if !p.Playing() {
+		t.Fatal("expected the reply to be playing")
+	}
+	plays := p.plays
+
+	(&nudgeHandler{sess: sess}).OnPush(pushFrame(t, nudgeAudio()))
+	for i := 0; i < 20; i++ {
+		if err := sess.feed(context.Background(), silentBytes()); err != nil {
+			t.Fatalf("feed: %v", err)
+		}
+	}
+	if p.plays != plays {
+		t.Fatalf("nudge cut across the reply: plays = %d, want %d", p.plays, plays)
+	}
+
+	p.playing = false
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != plays+1 {
+		t.Fatalf("nudge never played after the reply ended: plays = %d", p.plays)
+	}
+}
+
+// Speaking a nudge must not leave half a sentence in the VAD. The frames
+// captured before it are pre-nudge speech, and splicing them onto whatever he
+// says afterwards ships one utterance that is two.
+func TestNudgeResetsTheVAD(t *testing.T) {
+	sess, p, snd := newTestSession(bargeInConfig{})
+	loud := frameAt(0.35)
+	speechFrames := (defaultSpeechMs + defaultFrameMs - 1) / defaultFrameMs
+	for i := 0; i < speechFrames+5; i++ {
+		if err := sess.feed(context.Background(), loud); err != nil {
+			t.Fatalf("feed: %v", err)
+		}
+	}
+
+	(&nudgeHandler{sess: sess}).OnPush(pushFrame(t, nudgeAudio()))
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != 1 {
+		t.Fatalf("nudge did not play: plays = %d", p.plays)
+	}
+
+	// Playback ends, silence follows. The half-formed utterance must be gone
+	// rather than closing on the first quiet frame.
+	p.playing = false
+	silenceFrames := (defaultSilenceMs+defaultFrameMs-1)/defaultFrameMs + 2
+	for i := 0; i < silenceFrames; i++ {
+		if err := sess.feed(context.Background(), silentBytes()); err != nil {
+			t.Fatalf("feed: %v", err)
+		}
+	}
+	if len(snd.sent) != 0 {
+		t.Fatalf("sent %d utterances after a nudge, want 0", len(snd.sent))
+	}
+}
+
+// Two nudges queued back to back: the newer one is what he hears. The
+// PushHandler contract in internal/voice says the next nudge replaces the
+// stale one rather than dogpiling on it.
+func TestNudgeReplacesAnUnspokenOne(t *testing.T) {
+	sess, p, _ := newTestSession(bargeInConfig{})
+	h := &nudgeHandler{sess: sess}
+
+	h.OnPush(pushFrame(t, audio.Audio{Format: audio.PCM16kMono, Bytes: make([]byte, 4000)}))
+	h.OnPush(pushFrame(t, audio.Audio{Format: audio.PCM16kMono, Bytes: make([]byte, 12000)}))
+
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if p.plays != 1 {
+		t.Fatalf("plays = %d, want 1", p.plays)
+	}
+	if len(p.last.Bytes) != 12000 {
+		t.Errorf("played %d bytes, want the newer nudge", len(p.last.Bytes))
+	}
+}

cmd/mavwaked/session.go

@@ -7,6 +7,7 @@ package main
 import (
 	"context"
 	"log"
+	"sync"
 	"time"
 
 	"github.com/kami/maven/internal/audio"
@@ -19,6 +20,15 @@ type utteranceSender interface {
 	Send(ctx context.Context, utt audio.Audio, lang string) (audio.Audio, error)
 }
 
+// keywordGate answers whether the keyword has just been spoken. The
+// production one is wakeWord; tests substitute a recorder, because a gate that
+// can only be exercised with three ONNX files is a gate nobody tests.
+type keywordGate interface {
+	Feed(frame []int16) bool
+	Reset()
+	Score() float64
+}
+
 // bargeInConfig holds the two numbers barge-in needs. Zero Frames disables
 // barge-in entirely — the half-duplex gate still runs.
 type bargeInConfig struct {
@@ -62,11 +72,29 @@ type session struct {
 	// whenever playback ends.
 	loudFrames int
 
+	// wake is the keyword gate, or nil when no model was loaded. wakeUntil is
+	// how long a keyword stays good for: he says "Мэйвен" and then a sentence,
+	// and the VAD does not close the utterance until he stops, so the window
+	// has to outlive the word by the length of what follows it.
+	wake       keywordGate
+	wakeWindow time.Duration
+	wakeUntil  time.Time
+
+	// pending holds a nudge the push receiver handed over, waiting for the
+	// capture loop to speak it. It is the one field written from another
+	// goroutine, hence the mutex; everything else in this struct belongs to
+	// the capture loop alone.
+	nudgeMu sync.Mutex
+	pending *audio.Audio
+
 	// counters, read by tests and logged on the way out.
 	suppressed int // frames dropped because she was speaking
 	dropped    int // frames dropped as round-trip backlog
 	bargeIns   int // times playback was cut because he spoke over her
 	sent       int // utterances shipped to the daemon
+	nudges     int // proactive pushes spoken through the speaker
+	wakes      int // times the keyword opened the gate
+	ignored    int // complete utterances dropped because the keyword was absent
 
 	// loudSum and loudSeen accumulate the energy of suppressed frames, so
 	// the operator can read what the room actually measures and set
@@ -79,6 +107,12 @@ func newSession(vad *VAD, p player, s utteranceSender, lang string, barge bargeI
 	return &session{vad: vad, player: p, sender: s, lang: lang, barge: barge, now: time.Now}
 }
 
+// UseWakeWord puts the keyword gate in front of dispatch. Without it every
+// utterance is shipped, which is what mavwaked did before V-487 stage two.
+func (s *session) UseWakeWord(w keywordGate, window time.Duration) {
+	s.wake, s.wakeWindow = w, window
+}
+
 // frameDuration is the wall time one captured frame represents.
 const frameDuration = defaultFrameMs * time.Millisecond
 
@@ -138,6 +172,7 @@ func (s *session) feed(ctx context.Context, frame []byte) error {
 		s.bargeIns++
 		s.loudFrames = 0
 		s.vad.Reset()
+		s.resetWake()
 		log.Printf("mavwaked: barge-in — stopped playback")
 		s.replayRecent()
 		return nil
@@ -148,15 +183,101 @@ func (s *session) feed(ctx context.Context, frame []byte) error {
 	if s.loudFrames != 0 {
 		s.loudFrames = 0
 		s.vad.Reset()
+		// The wake word saw nothing during playback, so what it holds is from
+		// before she spoke. Judging what he says next on it would score a
+		// sentence that ended a reply ago.
+		s.resetWake()
 	}
 
-	utt, state := s.vad.Feed(PCMToI16(frame))
+	if s.startPendingNudge() {
+		return nil
+	}
+
+	// The keyword is scored on the same frames the VAD sees, and only on the
+	// ones that reach here: every path above returns while she is speaking, so
+	// her own voice saying "Мэйвен" cannot wake her.
+	pcm := PCMToI16(frame)
+	if s.wake != nil && s.wake.Feed(pcm) {
+		s.wakes++
+		s.wakeUntil = s.now().Add(s.wakeWindow)
+		log.Printf("mavwaked: keyword heard (score %.3f), listening for %s",
+			s.wake.Score(), s.wakeWindow)
+	}
+
+	utt, state := s.vad.Feed(pcm)
 	if state == StateSpeech || utt.Bytes == nil {
 		return nil
 	}
 	return s.dispatch(ctx, utt)
 }
 
+// Nudge hands proactive audio to the session, to be spoken as soon as the
+// capture loop finds a quiet moment. Safe to call from the push receiver
+// goroutine; nothing else here is.
+//
+// A nudge arriving while one is already waiting REPLACES it. That is the
+// contract internal/voice states for PushHandler: the next nudge replaces the
+// stale one in his attention rather than dogpiling on it.
+func (s *session) Nudge(a audio.Audio) {
+	if len(a.Bytes) == 0 {
+		return
+	}
+	s.nudgeMu.Lock()
+	if s.pending != nil {
+		log.Printf("mavwaked: nudge replaced one still waiting to be spoken")
+	}
+	s.pending = &a
+	s.nudgeMu.Unlock()
+}
+
+// takeNudge removes and returns the waiting nudge, or nil.
+func (s *session) takeNudge() *audio.Audio {
+	s.nudgeMu.Lock()
+	defer s.nudgeMu.Unlock()
+	a := s.pending
+	s.pending = nil
+	return a
+}
+
+// startPendingNudge speaks a waiting nudge and reports whether it started
+// one. It runs on the capture loop, past the half-duplex gate, so a nudge
+// never cuts across a reply and never plays into a backlog drain.
+//
+// The VAD is reset first. Playback is about to suppress every frame until it
+// ends, and a half-heard sentence left in the VAD would splice onto whatever
+// he says afterwards. Barge-in needs no special case: it reads the player,
+// and the player does not care which audio it is playing.
+func (s *session) startPendingNudge() bool {
+	a := s.takeNudge()
+	if a == nil {
+		return false
+	}
+	s.vad.Reset()
+	s.nudges++
+	log.Printf("mavwaked: speaking nudge (%.2fs audio)", a.Duration())
+	s.player.Play(*a)
+	return true
+}
+
+// awake reports whether an utterance ending now was addressed to her.
+//
+// With no wake word loaded every utterance is, which is exactly what mavwaked
+// did before this gate existed. An operator with no model file gets the old
+// daemon rather than a daemon that refuses to hear anything.
+func (s *session) awake() bool {
+	if s.wake == nil {
+		return true
+	}
+	return s.now().Before(s.wakeUntil)
+}
+
+// resetWake drops the gate's streaming state when there is a gate.
+func (s *session) resetWake() {
+	if s.wake != nil {
+		s.wake.Reset()
+	}
+}
+
 // keepRecent stores a copy of one barge-in trigger frame, keeping at most
 // barge.Frames of them.
 func (s *session) keepRecent(frame []byte) {
@@ -197,6 +318,19 @@ func (s *session) replayRecent() {
 // whole backlog straight into the VAD, and a Send error did the same on every
 // failed turn, so a dead socket drove a retry loop off nothing but backlog.
 func (s *session) dispatch(ctx context.Context, utt audio.Audio) error {
+	if !s.awake() {
+		s.ignored++
+		log.Printf("mavwaked: utterance ignored, keyword not heard (%.2fs, %d ignored so far)",
+			utt.Duration(), s.ignored)
+		s.vad.Reset()
+		s.resetWake()
+		return nil
+	}
+	// One keyword, one turn. A window that renewed itself on every reply would
+	// leave the microphone open for as long as he kept talking, which is the
+	// state this gate exists to end.
+	s.wakeUntil = time.Time{}
+
 	log.Printf("mavwaked: utterance complete (%.2fs, %d bytes), sending...", utt.Duration(), len(utt.Bytes))
 	start := s.now()
 	reply, err := s.sender.Send(ctx, utt, s.lang)
@@ -225,6 +359,7 @@ func (s *session) dispatch(ctx context.Context, utt audio.Audio) error {
 // recorded before she started speaking.
 func (s *session) dropBacklog(start time.Time) {
 	s.vad.Reset()
+	s.resetWake()
 	s.loudFrames = 0
 	s.recent = s.recent[:0]
 	if elapsed := s.now().Sub(start); elapsed > 0 {

cmd/mavwaked/silero.go

@@ -0,0 +1,171 @@
+package main
+
+// silero-vad, the speech detector that replaces the energy threshold (V-487).
+//
+// Why an energy threshold is not a voice activity detector. It answers "is
+// this frame loud", and a fan, a door and a television are all loud. mavwaked
+// sends every utterance it accepts to speech-to-text and then to the daemon,
+// so a false trigger is a turn Maven takes on something nobody said to her.
+// Silero answers "is this frame speech", which is the question.
+//
+// It is 2.3MB of ONNX and runs on one CPU core in real time. That is not an
+// aside: this is the one model in the system that may never be offloaded or
+// gated on GPU admission, because a wake path that waits on a card is not a
+// wake path.
+//
+// Nil is a working value. Without -vad-model the daemon runs the energy VAD
+// exactly as it did before this file existed.
+
+import (
+	"fmt"
+	"sync"
+
+	ort "github.com/yalue/onnxruntime_go"
+)
+
+const (
+	// sileroWindow — samples per inference at 16kHz. The model is fixed at
+	// 512 and does not accept another size, which is why this file
+	// re-chunks rather than reusing the 480-sample capture frame. main.go
+	// used to claim the two matched; that was true of silero v4.
+	sileroWindow = 512
+
+	// sileroContext — samples of the previous window prepended to each
+	// inference, as the reference implementation does. Without it the first
+	// milliseconds of every window are judged with no history and speech
+	// onsets score low.
+	sileroContext = 64
+
+	// sileroState — the LSTM state carried between windows, [2][1][128].
+	sileroStateDim = 128
+
+	// defaultSileroThreshold — probability above which a window is speech.
+	// 0.5 is the reference default. Raising it costs speech onsets, which
+	// are the quietest part of an utterance.
+	defaultSileroThreshold = 0.5
+)
+
+// sileroVAD holds one ONNX session and the streaming state around it. It is
+// fed 30ms capture frames and answers per frame, buffering across calls
+// because 480 samples never line up with a 512-sample window.
+type sileroVAD struct {
+	mu      sync.Mutex
+	session *ort.DynamicAdvancedSession
+
+	pending []float32              // samples not yet part of a full window
+	context [sileroContext]float32 // tail of the previous window
+	state   []float32              // [2][1][128], carried between windows
+	last    float64                // most recent probability, held between windows
+	sr      []int64
+}
+
+// newSileroVAD loads the graph. The ONNX environment is initialised here when
+// nothing else has done it, because mavwaked has no embedder to do it first.
+func newSileroVAD(modelPath, libPath string) (*sileroVAD, error) {
+	if !ort.IsInitialized() {
+		if libPath != "" {
+			ort.SetSharedLibraryPath(libPath)
+		}
+		if err := ort.InitializeEnvironment(); err != nil {
+			return nil, fmt.Errorf("silero: onnx runtime: %w", err)
+		}
+	}
+	s, err := ort.NewDynamicAdvancedSession(modelPath,
+		[]string{"input", "state", "sr"}, []string{"output", "stateN"}, nil)
+	if err != nil {
+		return nil, fmt.Errorf("silero: load %s: %w", modelPath, err)
+	}
+	return &sileroVAD{
+		session: s,
+		state:   make([]float32, 2*sileroStateDim),
+		sr:      []int64{16000},
+	}, nil
+}
+
+// Speech reports whether the frame carries speech, and the probability behind
+// that answer. A frame that completes no window inherits the previous
+// probability, so the caller sees one answer per frame either way.
+func (s *sileroVAD) Speech(frame []int16, threshold float64) (bool, float64) {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	for _, v := range frame {
+		s.pending = append(s.pending, float32(v)/32768.0)
+	}
+	for len(s.pending) >= sileroWindow {
+		p, err := s.infer(s.pending[:sileroWindow])
+		if err != nil {
+			// A failed inference must not silence the microphone. Hold the
+			// last answer and let the next window try again.
+			break
+		}
+		s.last = p
+		s.pending = s.pending[sileroWindow:]
+	}
+	return s.last >= threshold, s.last
+}
+
+// infer runs one window and rolls the state and the context forward.
+func (s *sileroVAD) infer(window []float32) (float64, error) {
+	in := make([]float32, sileroContext+sileroWindow)
+	copy(in, s.context[:])
+	copy(in[sileroContext:], window)
+
+	inT, err := ort.NewTensor(ort.NewShape(1, int64(len(in))), in)
+	if err != nil {
+		return 0, err
+	}
+	defer inT.Destroy()
+	stT, err := ort.NewTensor(ort.NewShape(2, 1, sileroStateDim), s.state)
+	if err != nil {
+		return 0, err
+	}
+	defer stT.Destroy()
+	srT, err := ort.NewTensor(ort.NewShape(1), s.sr)
+	if err != nil {
+		return 0, err
+	}
+	defer srT.Destroy()
+
+	out, err := ort.NewEmptyTensor[float32](ort.NewShape(1, 1))
+	if err != nil {
+		return 0, err
+	}
+	defer out.Destroy()
+	next, err := ort.NewEmptyTensor[float32](ort.NewShape(2, 1, sileroStateDim))
+	if err != nil {
+		return 0, err
+	}
+	defer next.Destroy()
+
+	if err := s.session.Run(
+		[]ort.Value{inT, stT, srT},
+		[]ort.Value{out, next},
+	); err != nil {
+		return 0, err
+	}
+	copy(s.state, next.GetData())
+	copy(s.context[:], in[len(in)-sileroContext:])
+	return float64(out.GetData()[0]), nil
+}
+
+// Reset drops the streaming state. Called at every utterance boundary and
+// after barge-in, so echo-era history never scores the next sentence.
+func (s *sileroVAD) Reset() {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	s.pending = s.pending[:0]
+	s.context = [sileroContext]float32{}
+	for i := range s.state {
+		s.state[i] = 0
+	}
+	s.last = 0
+}
+
+// Close releases the session.
+func (s *sileroVAD) Close() error {
+	if s == nil || s.session == nil {
+		return nil
+	}
+	return s.session.Destroy()
+}

cmd/mavwaked/silero_test.go

@@ -0,0 +1,149 @@
+package main
+
+// What this measures. The energy threshold cannot tell a voice from a
+// television, and every utterance it accepts becomes a turn. So the test that
+// matters is not "does silero find speech" — it is "does it decline what the
+// energy threshold accepts".
+//
+// Speech is the four piper fixtures mavsttd already scores against. They are
+// synthesised, so nothing of the owner's voice is committed. Non-speech is
+// white noise at the same loudness, which is the cheapest thing that fools an
+// energy floor and the honest floor for this claim.
+//
+// Both halves skip without models/vad/silero_vad.onnx and MAVEN_ONNX_LIB,
+// like the TestONNX measurements in internal/router/eval.
+
+import (
+	"math"
+	"math/rand"
+	"os"
+	"path/filepath"
+	"testing"
+)
+
+const wavHeader = 44 // 16kHz mono s16le, written by piper
+
+func loadSilero(t *testing.T) *sileroVAD {
+	t.Helper()
+	model := filepath.Join("..", "..", "models", "vad", "silero_vad.onnx")
+	lib := os.Getenv("MAVEN_ONNX_LIB")
+	if _, err := os.Stat(model); err != nil {
+		t.Skipf("missing %s: %v", model, err)
+	}
+	if lib == "" {
+		t.Skip("MAVEN_ONNX_LIB unset")
+	}
+	s, err := newSileroVAD(model, lib)
+	if err != nil {
+		t.Skipf("silero unavailable: %v", err)
+	}
+	return s
+}
+
+// feedAll runs a whole clip through a VAD and reports how many utterances it
+// produced and how many frames it called speech.
+func feedAll(v *VAD, pcm []int16) (utterances, speechFrames int) {
+	for i := 0; i+frameSamples <= len(pcm); i += frameSamples {
+		frame := pcm[i : i+frameSamples]
+		utt, state := v.Feed(frame)
+		if state == StateSpeech {
+			speechFrames++
+		}
+		if utt.Bytes != nil {
+			utterances++
+		}
+	}
+	return utterances, speechFrames
+}
+
+func readFixture(t *testing.T, name string) []int16 {
+	t.Helper()
+	raw, err := os.ReadFile(filepath.Join("..", "mavsttd", "testdata", name))
+	if err != nil {
+		t.Skipf("missing fixture %s: %v", name, err)
+	}
+	if len(raw) <= wavHeader {
+		t.Fatalf("%s: %d bytes, no audio", name, len(raw))
+	}
+	return PCMToI16(raw[wavHeader:])
+}
+
+// noise returns white noise scaled to the same RMS as ref. Same loudness,
+// nothing said.
+func noise(ref []int16, seed int64) []int16 {
+	target := frameRMS(ref)
+	r := rand.New(rand.NewSource(seed))
+	out := make([]int16, len(ref))
+	for i := range out {
+		out[i] = int16(r.NormFloat64() * target * 32768.0)
+	}
+	return out
+}
+
+func TestSileroHearsSpeechAndDeclinesNoise(t *testing.T) {
+	s := loadSilero(t)
+	defer s.Close()
+
+	for _, name := range []string{"ru_fact.wav", "ru_query.wav", "ru_reminder.wav", "en_act.wav"} {
+		pcm := readFixture(t, name)
+
+		v := NewVAD(0, 0, 0, 0)
+		v.UseSilero(s, defaultSileroThreshold)
+		_, spoke := feedAll(v, pcm)
+		if spoke == 0 {
+			t.Errorf("%s: silero heard no speech in a spoken clip", name)
+		}
+
+		s.Reset()
+		v2 := NewVAD(0, 0, 0, 0)
+		v2.UseSilero(s, defaultSileroThreshold)
+		_, heard := feedAll(v2, noise(pcm, 7))
+
+		energy := NewVAD(0, 0, 0, 0)
+		_, energyHeard := feedAll(energy, noise(pcm, 7))
+
+		t.Logf("%s: speech frames — silero on speech %d, silero on noise %d, energy on noise %d",
+			name, spoke, heard, energyHeard)
+		if heard >= energyHeard {
+			t.Errorf("%s: silero called %d noise frames speech, energy called %d — no improvement",
+				name, heard, energyHeard)
+		}
+		s.Reset()
+	}
+}
+
+// BenchmarkSileroFrame answers the only performance question that matters
+// here: one 30ms frame must cost far less than 30ms on one core, or the
+// detector cannot run always-on beside everything else on that machine.
+func BenchmarkSileroFrame(b *testing.B) {
+	s := loadSilero(&testing.T{})
+	if s == nil {
+		b.Skip("silero unavailable")
+	}
+	defer s.Close()
+	frame := make([]int16, frameSamples)
+	for i := range frame {
+		frame[i] = int16(i%400 - 200)
+	}
+	for i := 0; i < b.N; i++ {
+		s.Speech(frame, defaultSileroThreshold)
+	}
+}
+
+// TestSileroRechunksAcrossFrames pins the reason this file exists. The capture
+// frame is 480 samples and the model window is 512, so a detector that ran one
+// inference per frame would be feeding the model a shape it does not accept.
+func TestSileroRechunksAcrossFrames(t *testing.T) {
+	s := loadSilero(t)
+	defer s.Close()
+
+	silence := make([]int16, frameSamples)
+	for i := 0; i < 20; i++ {
+		if _, p := s.Speech(silence, defaultSileroThreshold); math.IsNaN(p) {
+			t.Fatalf("frame %d: probability is NaN", i)
+		}
+	}
+	if len(s.pending) >= sileroWindow {
+		t.Errorf("pending grew to %d samples, so windows are not being consumed", len(s.pending))
+	}
+}

cmd/mavwaked/vad.go

@@ -68,6 +68,37 @@ type VAD struct {
 	// follows the room's ambient level. Initialised to minRMS; updated
 	// on each silence frame.
 	floorRMS float64
+
+	// speech is silero-vad, or nil. When it is set the energy floor decides
+	// nothing: the question becomes "is this speech" rather than "is this
+	// loud", and the noise floor is not even tracked. Everything after that
+	// answer — the speech hold, the silence hold, the length cap, the
+	// buffer — is the same state machine either way, which is why the
+	// detector goes here and not around this type.
+	speech    *sileroVAD
+	speechMin float64
+}
+
+// UseSilero swaps the energy threshold for the model. Passing nil is a
+// no-op, so a caller that could not load the graph keeps a working VAD.
+func (v *VAD) UseSilero(s *sileroVAD, threshold float64) {
+	if s == nil {
+		return
+	}
+	if threshold <= 0 {
+		threshold = defaultSileroThreshold
+	}
+	v.speech = s
+	v.speechMin = threshold
+}
+
+// isSpeech answers the one question the state machine asks of a frame.
+func (v *VAD) isSpeech(frame []int16, rms float64) bool {
+	if v.speech != nil {
+		ok, _ := v.speech.Speech(frame, v.speechMin)
+		return ok
+	}
+	return rms >= v.floorRMS
 }
 
 // NewVAD creates a VAD with the given thresholds. Zero values use defaults.
@@ -110,7 +141,7 @@ func (v *VAD) State() SpeechState { return v.state }
 // should send the audio to the voice server before feeding more frames.
 func (v *VAD) Feed(frame []int16) (_ audio.Audio, state SpeechState) {
 	rms := frameRMS(frame)
-	isSpeech := rms >= v.floorRMS
+	isSpeech := v.isSpeech(frame, rms)
 
 	switch v.state {
 	case StateSilence:
@@ -175,6 +206,9 @@ func (v *VAD) Feed(frame []int16) (_ audio.Audio, state SpeechState) {
 func (v *VAD) Reset() { v.reset() }
 
 func (v *VAD) reset() {
+	if v.speech != nil {
+		v.speech.Reset()
+	}
 	v.state = StateSilence
 	v.speechFrames = 0
 	v.silenceFrames = 0

cmd/mavwaked/wakefeatures.go

@@ -0,0 +1,202 @@
+package main
+
+// The three models behind the wake word (V-487 stage two).
+//
+// openWakeWord's pipeline, run in a row:
+//
+//	audio      -> melspectrogram.onnx -> 32-bin mel frames, one per 10ms
+//	76 frames  -> embedding_model.onnx -> one 96-dim embedding per 80ms
+//	16 embeds  -> maven_wakeword.onnx  -> one score
+//
+// The first two are frozen and pretrained. Only the last was trained here,
+// which is why it is 100KB and the other two are megabytes. The shapes are
+// not guesses: 2.0s of 16kHz audio measures 197 mel frames, and 76-frame
+// windows at stride 8 give exactly the 16 embeddings the head was fitted on.
+//
+// This file knows ONNX and nothing about the 80ms cadence. wakeword.go knows
+// the cadence and nothing about tensors.
+
+import (
+	"fmt"
+
+	ort "github.com/yalue/onnxruntime_go"
+)
+
+const (
+	// melHop — samples per mel frame. 10ms at 16kHz.
+	melHop = 160
+	// melBins — mel bins per frame, fixed by melspectrogram.onnx.
+	melBins = 32
+	// embedFrames — mel frames one embedding is computed over, 760ms.
+	embedFrames = 76
+	// embedStride — mel frames between embeddings, 80ms.
+	embedStride = 8
+	// embedDim — the embedding width.
+	embedDim = 96
+	// headWindow — embeddings the head scores at once, 1.28s of audio.
+	headWindow = 16
+
+	// melContext — samples of history prepended to each incremental mel
+	// call, chosen so the eight frames this call yields continue exactly
+	// where the previous call's eight stopped.
+	//
+	// melspectrogram.onnx returns N/160-3 frames for N samples, and frame i
+	// covers [i*160, i*160+400). With 480 samples of history the buffer is
+	// 1760 samples, which is 8 frames, and the oldest of them starts one hop
+	// after the newest of the previous call. Less history leaves a gap: the
+	// first frames of a bare chunk would be computed against silence.
+	melContext = 480
+
+	// chunkSamples — audio per embedding step, 80ms.
+	chunkSamples = embedStride * melHop
+)
+
+// wakeModels holds the three ONNX sessions. It runs on CPU threads beside
+// silero and never touches the GPU. That is a rule, not a result: a wake word
+// that waits on card admission is not a wake word.
+type wakeModels struct {
+	mel  *ort.DynamicAdvancedSession
+	emb  *ort.DynamicAdvancedSession
+	head *ort.DynamicAdvancedSession
+}
+
+// newWakeModels loads all three. melPath and embedPath are openWakeWord's
+// frozen feature models; headPath is the keyword head trained for "Мэйвен".
+func newWakeModels(melPath, embedPath, headPath, libPath string) (*wakeModels, error) {
+	if !ort.IsInitialized() {
+		if libPath != "" {
+			ort.SetSharedLibraryPath(libPath)
+		}
+		if err := ort.InitializeEnvironment(); err != nil {
+			return nil, fmt.Errorf("wake word: onnx runtime: %w", err)
+		}
+	}
+	// One thread per session, not the default of every core. Measured on
+	// workpc: the default took mavwaked from 68% of one core to 335% of
+	// three, for three graphs that each run in well under 80ms single
+	// threaded. An always-on gate that eats a quarter of the workstation is
+	// not a gate he will leave running.
+	opts, err := ort.NewSessionOptions()
+	if err != nil {
+		return nil, fmt.Errorf("wake word: session options: %w", err)
+	}
+	defer opts.Destroy()
+	if err := opts.SetIntraOpNumThreads(1); err != nil {
+		return nil, fmt.Errorf("wake word: intra-op threads: %w", err)
+	}
+	if err := opts.SetInterOpNumThreads(1); err != nil {
+		return nil, fmt.Errorf("wake word: inter-op threads: %w", err)
+	}
+	open := func(p string, in, out []string) (*ort.DynamicAdvancedSession, error) {
+		s, err := ort.NewDynamicAdvancedSession(p, in, out, opts)
+		if err != nil {
+			return nil, fmt.Errorf("wake word: load %s: %w", p, err)
+		}
+		return s, nil
+	}
+	m := &wakeModels{}
+	if m.mel, err = open(melPath, []string{"input"}, []string{"output"}); err != nil {
+		return nil, err
+	}
+	if m.emb, err = open(embedPath, []string{"input_1"}, []string{"conv2d_19"}); err != nil {
+		m.Close()
+		return nil, err
+	}
+	if m.head, err = open(headPath, []string{"embeddings"}, []string{"score"}); err != nil {
+		m.Close()
+		return nil, err
+	}
+	return m, nil
+}
+
+// Close releases the three sessions.
+func (m *wakeModels) Close() {
+	if m == nil {
+		return
+	}
+	for _, s := range []*ort.DynamicAdvancedSession{m.mel, m.emb, m.head} {
+		if s != nil {
+			s.Destroy()
+		}
+	}
+	m.mel, m.emb, m.head = nil, nil, nil
+}
+
+// melFrames runs one buffer of samples and returns the mel frames it yielded.
+func (m *wakeModels) melFrames(buf []float32) ([][melBins]float32, error) {
+	in, err := ort.NewTensor(ort.NewShape(1, int64(len(buf))), buf)
+	if err != nil {
+		return nil, err
+	}
+	defer in.Destroy()
+
+	n := int64(len(buf)/melHop - 3)
+	if n < 1 {
+		return nil, fmt.Errorf("wake word: %d samples yield no mel frames", len(buf))
+	}
+	out, err := ort.NewEmptyTensor[float32](ort.NewShape(1, 1, n, melBins))
+	if err != nil {
+		return nil, err
+	}
+	defer out.Destroy()
+
+	if err := m.mel.Run([]ort.Value{in}, []ort.Value{out}); err != nil {
+		return nil, err
+	}
+	data := out.GetData()
+	frames := make([][melBins]float32, n)
+	for i := range frames {
+		for j := 0; j < melBins; j++ {
+			// The scaling openWakeWord applies between the two feature
+			// models, and the head was fitted on its output.
+			frames[i][j] = data[i*melBins+j]/10.0 + 2.0
+		}
+	}
+	return frames, nil
+}
+
+// embedding runs embedFrames mel frames through the frozen embedder.
+func (m *wakeModels) embedding(mels [][melBins]float32) ([embedDim]float32, error) {
+	var e [embedDim]float32
+	flat := make([]float32, 0, embedFrames*melBins)
+	for _, f := range mels {
+		flat = append(flat, f[:]...)
+	}
+	in, err := ort.NewTensor(ort.NewShape(1, embedFrames, melBins, 1), flat)
+	if err != nil {
+		return e, err
+	}
+	defer in.Destroy()
+	out, err := ort.NewEmptyTensor[float32](ort.NewShape(1, 1, 1, embedDim))
+	if err != nil {
+		return e, err
+	}
+	defer out.Destroy()
+	if err := m.emb.Run([]ort.Value{in}, []ort.Value{out}); err != nil {
+		return e, err
+	}
+	copy(e[:], out.GetData())
+	return e, nil
+}
+
+// score runs the trained head over headWindow embeddings.
+func (m *wakeModels) score(embeds [][embedDim]float32) (float64, error) {
+	flat := make([]float32, 0, headWindow*embedDim)
+	for _, e := range embeds {
+		flat = append(flat, e[:]...)
+	}
+	in, err := ort.NewTensor(ort.NewShape(1, headWindow, embedDim), flat)
+	if err != nil {
+		return 0, err
+	}
+	defer in.Destroy()
+	out, err := ort.NewEmptyTensor[float32](ort.NewShape(1, 1))
+	if err != nil {
+		return 0, err
+	}
+	defer out.Destroy()
+	if err := m.head.Run([]ort.Value{in}, []ort.Value{out}); err != nil {
+		return 0, err
+	}
+	return float64(out.GetData()[0]), nil
+}

cmd/mavwaked/wakeword.go

@@ -0,0 +1,195 @@
+package main
+
+// The wake word, "Мэйвен" (V-487 stage two).
+//
+// Silero answers "is this frame speech". It does not answer "was this said to
+// her", and until this file existed nothing did: every utterance near the
+// microphone became a turn. What made that safe rather than expensive was
+// SurfaceVoice capping acts at L0, and L0 does not cap reading, so the room
+// could still hear his facts read back.
+//
+// This file owns the 80ms cadence and the three rings of state between the
+// models. wakefeatures.go owns the tensors.
+//
+// Nil is a working value, and it is the CLOSED gate rather than the open one.
+// Feed on a nil receiver reports no keyword; session.go asks separately
+// whether a gate exists at all. That split is deliberate: a nil that answers
+// "yes, keyword" reads as a working wake word in every log line it produces.
+
+import (
+	"log"
+	"sync"
+)
+
+// defaultWakeThreshold — score above which the keyword was said.
+//
+// Picked from the false-accept rate on held-out Russian speech, not from
+// accuracy: a miss costs him a repeat, a false accept costs a turn nobody
+// asked for. Over 65 minutes of Common Voice, 0.99 woke her three times and
+// 0.999 once, and the difference in recall was one render out of 126. So the
+// default is the strict one. `docs/evals/2026-08-09-wake-word.md` has both
+// tables.
+const defaultWakeThreshold = 0.999
+
+// wakeWord is the streaming state around wakeModels. It is fed the same
+// capture frames the VAD sees and answers whether the keyword has just been
+// spoken.
+type wakeWord struct {
+	mu sync.Mutex
+	m  *wakeModels
+
+	threshold float64
+
+	// pending holds captured samples not yet part of a full 80ms chunk, and
+	// history holds the melContext samples before them.
+	pending []float32
+	history []float32
+
+	// mels is the newest embedFrames mel frames, oldest first.
+	mels [][melBins]float32
+	// embeds is the newest headWindow embeddings, oldest first.
+	embeds [][embedDim]float32
+
+	last float64 // most recent score, held between chunks
+}
+
+// newWakeWord loads the models and wraps them in the streaming gate.
+func newWakeWord(melPath, embedPath, headPath, libPath string, threshold float64) (*wakeWord, error) {
+	m, err := newWakeModels(melPath, embedPath, headPath, libPath)
+	if err != nil {
+		return nil, err
+	}
+	if threshold <= 0 {
+		threshold = defaultWakeThreshold
+	}
+	return &wakeWord{m: m, threshold: threshold}, nil
+}
+
+// Close releases the models.
+func (w *wakeWord) Close() {
+	if w == nil {
+		return
+	}
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	w.m.Close()
+	w.m = nil
+}
+
+// Feed takes one capture frame and reports whether the keyword was heard on
+// it. A nil wakeWord hears nothing.
+func (w *wakeWord) Feed(frame []int16) bool {
+	if w == nil {
+		return false
+	}
+	w.mu.Lock()
+	defer w.mu.Unlock()
+
+	for _, v := range frame {
+		w.pending = append(w.pending, float32(v)/32768.0)
+	}
+	fired := false
+	for len(w.pending) >= chunkSamples {
+		chunk := w.pending[:chunkSamples]
+		if w.step(chunk) {
+			fired = true
+		}
+		w.history = append(w.history[:0], tailFloat32(append(w.history, chunk...), melContext)...)
+		// Slide the remainder to the front rather than reslicing. This runs
+		// every 80ms for as long as the daemon lives.
+		w.pending = append(w.pending[:0], w.pending[chunkSamples:]...)
+	}
+	return fired
+}
+
+// Reset drops the streaming state, so a fresh utterance is not judged on audio
+// from before it. Called after every dispatch and after barge-in, for the same
+// reason silero is: echo-era history must not score the next sentence, and her
+// own voice saying the keyword must not wake her.
+func (w *wakeWord) Reset() {
+	if w == nil {
+		return
+	}
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	w.pending, w.history = w.pending[:0], w.history[:0]
+	w.mels, w.embeds = nil, nil
+	w.last = 0
+}
+
+// Score returns the most recent score, for the operator to read out of the
+// journal when picking a threshold for his room.
+func (w *wakeWord) Score() float64 {
+	if w == nil {
+		return 0
+	}
+	w.mu.Lock()
+	defer w.mu.Unlock()
+	return w.last
+}
+
+// step runs one 80ms chunk through all three models. It returns true when the
+// score crosses the threshold on this chunk.
+func (w *wakeWord) step(chunk []float32) bool {
+	buf := make([]float32, 0, melContext+len(chunk))
+	if pad := melContext - len(w.history); pad > 0 {
+		buf = append(buf, make([]float32, pad)...)
+	}
+	buf = append(buf, tailFloat32(w.history, melContext)...)
+	buf = append(buf, chunk...)
+
+	frames, err := w.m.melFrames(buf)
+	if err != nil {
+		// A failed inference must not silence the microphone. Hold the last
+		// score and let the next chunk try again.
+		log.Printf("mavwaked: wake word: mel: %v", err)
+		return false
+	}
+	w.mels = tailMel(append(w.mels, frames...), embedFrames)
+	if len(w.mels) < embedFrames {
+		return false
+	}
+	e, err := w.m.embedding(w.mels)
+	if err != nil {
+		log.Printf("mavwaked: wake word: embedding: %v", err)
+		return false
+	}
+	w.embeds = tailEmbed(append(w.embeds, e), headWindow)
+	if len(w.embeds) < headWindow {
+		return false
+	}
+	score, err := w.m.score(w.embeds)
+	if err != nil {
+		log.Printf("mavwaked: wake word: head: %v", err)
+		return false
+	}
+	// Report the crossing, not the state. A keyword held above the threshold
+	// for a second is one wake, and firing on every chunk of it would make the
+	// gate look open when it is merely slow to fall.
+	crossed := score >= w.threshold && w.last < w.threshold
+	w.last = score
+	return crossed
+}
+
+// The three rings. Each keeps the newest n entries and nothing older.
+
+func tailFloat32(s []float32, n int) []float32 {
+	if len(s) <= n {
+		return s
+	}
+	return s[len(s)-n:]
+}
+
+func tailMel(s [][melBins]float32, n int) [][melBins]float32 {
+	if len(s) <= n {
+		return s
+	}
+	return append(s[:0], s[len(s)-n:]...)
+}
+
+func tailEmbed(s [][embedDim]float32, n int) [][embedDim]float32 {
+	if len(s) <= n {
+		return s
+	}
+	return append(s[:0], s[len(s)-n:]...)
+}

cmd/mavwaked/wakeword_test.go

@@ -0,0 +1,174 @@
+package main
+
+import (
+	"context"
+	"testing"
+	"time"
+)
+
+// fakeGate fires on demand instead of running three ONNX models. The gate's
+// own arithmetic is measured on real audio in docs/evals; what these tests
+// cover is the thing that decides whether an utterance is shipped.
+type fakeGate struct {
+	fireOn int // fire when this many frames have been fed, 0 never fires
+	fed    int
+	resets int
+}
+
+func (g *fakeGate) Feed(_ []int16) bool {
+	g.fed++
+	return g.fireOn > 0 && g.fed == g.fireOn
+}
+func (g *fakeGate) Reset()         { g.resets++ }
+func (g *fakeGate) Score() float64 { return 1 }
+
+// wakingSession wires a session whose gate fires on the first frame it sees.
+func wakingSession(fireOn int, window time.Duration) (*session, *fakePlayer, *fakeSender, *fakeGate) {
+	sess, p, snd := newTestSession(bargeInConfig{})
+	g := &fakeGate{fireOn: fireOn}
+	sess.UseWakeWord(g, window)
+	return sess, p, snd, g
+}
+
+func TestKeywordlessSpeechNeverReachesSTT(t *testing.T) {
+	sess, p, snd, g := wakingSession(0, 8*time.Second)
+	speakThenPause(t, sess)
+
+	if len(snd.sent) != 0 {
+		t.Fatalf("sent %d utterances, want 0 — this is the whole point of V-487", len(snd.sent))
+	}
+	if sess.ignored != 1 {
+		t.Errorf("ignored = %d, want 1", sess.ignored)
+	}
+	if p.plays != 0 {
+		t.Errorf("plays = %d, want 0", p.plays)
+	}
+	if g.fed == 0 {
+		t.Error("the gate was never fed a frame")
+	}
+}
+
+func TestKeywordOpensTheGate(t *testing.T) {
+	sess, p, snd, _ := wakingSession(1, 8*time.Second)
+	speakThenPause(t, sess)
+
+	if len(snd.sent) != 1 {
+		t.Fatalf("sent %d utterances, want 1", len(snd.sent))
+	}
+	if sess.wakes != 1 {
+		t.Errorf("wakes = %d, want 1", sess.wakes)
+	}
+	if sess.ignored != 0 {
+		t.Errorf("ignored = %d, want 0", sess.ignored)
+	}
+	if p.plays != 1 {
+		t.Errorf("plays = %d, want 1", p.plays)
+	}
+}
+
+// One keyword buys one turn. Without this the microphone stays open for as
+// long as he keeps talking, which is the state the gate exists to end.
+func TestOneKeywordBuysOneTurn(t *testing.T) {
+	sess, p, snd, _ := wakingSession(1, 8*time.Second)
+	speakThenPause(t, sess)
+	p.Stop()         // she finished her reply
+	sess.discard = 0 // the backlog drain is not what this measures
+	speakThenPause(t, sess)
+
+	if len(snd.sent) != 1 {
+		t.Fatalf("sent %d utterances, want 1: the second had no keyword", len(snd.sent))
+	}
+	if sess.ignored != 1 {
+		t.Errorf("ignored = %d, want 1", sess.ignored)
+	}
+}
+
+// The keyword is heard, then he says nothing for longer than the window. What
+// he says after that is not addressed to her.
+func TestTheKeywordExpires(t *testing.T) {
+	sess, _, snd, _ := wakingSession(1, 500*time.Millisecond)
+	now := time.Unix(1750000000, 0)
+	sess.now = func() time.Time { return now }
+
+	if err := sess.feed(context.Background(), silentBytes()); err != nil {
+		t.Fatalf("feed: %v", err)
+	}
+	if sess.wakes != 1 {
+		t.Fatalf("wakes = %d, want 1", sess.wakes)
+	}
+	now = now.Add(2 * time.Second)
+	speakThenPause(t, sess)
+
+	if len(snd.sent) != 0 {
+		t.Fatalf("sent %d utterances, want 0 — the keyword had expired", len(snd.sent))
+	}
+}
+
+// Barge-in cuts her off whether or not the keyword was heard. What he says
+// after cutting her off still has to carry it.
+func TestBargeInStillInterruptsHer(t *testing.T) {
+	sess, p, _, g := wakingSession(0, 8*time.Second)
+	sess.barge = bargeInConfig{RMS: 0.2, Frames: 3}
+	p.playing = true
+	loud := frameAt(0.35)
+	for i := 0; i < 4; i++ {
+		if err := sess.feed(context.Background(), loud); err != nil {
+			t.Fatalf("feed %d: %v", i, err)
+		}
+	}
+	if sess.bargeIns != 1 {
+		t.Fatalf("bargeIns = %d, want 1", sess.bargeIns)
+	}
+	if p.stops != 1 {
+		t.Errorf("stops = %d, want 1", p.stops)
+	}
+	if g.resets == 0 {
+		t.Error("barge-in left pre-playback audio in the gate")
+	}
+}
+
+// Her own reply must not wake her. Frames captured while the player runs never
+// reach the gate, and the gate is cleared when playback ends.
+func TestHerOwnVoiceNeverReachesTheGate(t *testing.T) {
+	sess, p, _, g := wakingSession(1, 8*time.Second)
+	p.playing = true
+	for i := 0; i < 10; i++ {
+		if err := sess.feed(context.Background(), frameAt(0.35)); err != nil {
+			t.Fatalf("feed: %v", err)
+		}
+	}
+	if g.fed != 0 {
+		t.Fatalf("gate was fed %d frames while she was speaking, want 0", g.fed)
+	}
+	if sess.wakes != 0 {
+		t.Errorf("wakes = %d, want 0", sess.wakes)
+	}
+}
+
+// No model, no gate: the daemon behaves exactly as it did before V-487 stage
+// two. An operator with a missing file gets yesterday's mavwaked, not one that
+// refuses to hear anything.
+func TestNoGateShipsEveryUtterance(t *testing.T) {
+	sess, _, snd := newTestSession(bargeInConfig{})
+	speakThenPause(t, sess)
+
+	if len(snd.sent) != 1 {
+		t.Fatalf("sent %d utterances, want 1", len(snd.sent))
+	}
+	if sess.ignored != 0 {
+		t.Errorf("ignored = %d, want 0", sess.ignored)
+	}
+}
+
+// A nil *wakeWord is the closed gate, not a crash and not an open one.
+func TestNilWakeWordHearsNothing(t *testing.T) {
+	var w *wakeWord
+	if w.Feed([]int16{0, 0, 0}) {
+		t.Error("a nil wake word reported the keyword")
+	}
+	if w.Score() != 0 {
+		t.Error("a nil wake word reported a score")
+	}
+	w.Reset()
+	w.Close()
+}

deploy/cw2/serve.py

@@ -0,0 +1,156 @@
+"""CrisperWhisper 2.0 turbo as an HTTP service, for Maven's stt.Pair.
+
+Two endpoints and no framework.
+
+  GET  /health      200 once the model is loaded, 503 while it is loading.
+  POST /transcribe  raw 16kHz mono PCM in, {"text","confidence"} out.
+
+The body is the PCM itself rather than JSON. A minute of 16kHz mono is under
+2MB raw and about 2.6MB base64, and the format is fixed at the Maven seam, so
+headers carry it more cheaply than an envelope.
+
+Why this exists at all: whisper.cpp cannot load CW2. It derives its language
+count from the vocabulary size, and CW2's 51897 tokens shift seven special
+token ids. So mavsttd stays whisper.cpp on homesrv and this runs beside the
+model on workpc, where it scores 10.4% WER in Russian against the floor's 27.5%
+(docs/evals/2026-08-09-crisperwhisper2-russian-wer.md in the Maven repo).
+
+Intended mode, not verbatim. The owner asked for what he meant to say, not
+every stutter on the way there.
+"""
+
+import hmac
+import json
+import logging
+import os
+import sys
+import threading
+import time
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+
+import numpy as np
+
+HOST = os.environ.get("CW2_HOST", "0.0.0.0")
+PORT = int(os.environ.get("CW2_PORT", "8081"))
+SIZE = os.environ.get("CW2_SIZE", "turbo")
+MODE = os.environ.get("CW2_MODE", "intended")
+TOKEN = os.environ.get("CW2_TOKEN", "")
+# 25MB is about thirteen minutes of 16kHz mono. Longer than any utterance and
+# short enough that a wrong caller cannot exhaust memory.
+MAX_BODY = int(os.environ.get("CW2_MAX_BODY", str(25 * 1024 * 1024)))
+
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s cw2: %(message)s", stream=sys.stderr
+)
+log = logging.getLogger("cw2")
+
+_model = None
+# The card holds one model and transcribes one utterance at a time. The lock is
+# what makes a second caller wait rather than corrupt the first.
+_lock = threading.Lock()
+
+
+def load_model():
+    global _model
+    from crisperwhisper import CrisperWhisperModel
+
+    t0 = time.perf_counter()
+    # backend is forced. With ctranslate2 importable, "auto" picks ct2, which is
+    # CUDA-only and this card is AMD.
+    m = CrisperWhisperModel(
+        SIZE, backend="transformers", compute_type="float16", device="cuda"
+    )
+    _model = m
+    log.info("loaded %s in %.1fs, mode=%s", SIZE, time.perf_counter() - t0, MODE)
+
+
+def authorised(headers):
+    if not TOKEN:
+        return True
+    got = headers.get("Authorization", "")
+    return hmac.compare_digest(got, "Bearer " + TOKEN)
+
+
+class Handler(BaseHTTPRequestHandler):
+    protocol_version = "HTTP/1.1"
+
+    def log_message(self, fmt, *args):
+        log.info(fmt, *args)
+
+    def _send(self, code, payload):
+        body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+        self.send_response(code)
+        self.send_header("Content-Type", "application/json; charset=utf-8")
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        self.wfile.write(body)
+
+    def do_GET(self):
+        if self.path.rstrip("/") != "/health":
+            self._send(404, {"error": "not found"})
+            return
+        if _model is None:
+            self._send(503, {"status": "loading"})
+            return
+        self._send(200, {"status": "ok", "model": SIZE, "mode": MODE})
+
+    def do_POST(self):
+        if self.path.rstrip("/") != "/transcribe":
+            self._send(404, {"error": "not found"})
+            return
+        if not authorised(self.headers):
+            self._send(401, {"error": "unauthorised"})
+            return
+        if _model is None:
+            self._send(503, {"error": "loading"})
+            return
+
+        length = int(self.headers.get("Content-Length", "0"))
+        if length <= 0 or length > MAX_BODY:
+            self._send(413, {"error": "bad body length"})
+            return
+        raw = self.rfile.read(length)
+
+        rate = int(self.headers.get("X-Sample-Rate", "16000"))
+        channels = int(self.headers.get("X-Channels", "1"))
+        bits = int(self.headers.get("X-Sample-Bits", "16"))
+        lang = self.headers.get("X-Language", "ru") or "ru"
+        if channels != 1 or bits != 16:
+            self._send(400, {"error": "want 16-bit mono pcm"})
+            return
+
+        # int16 little-endian to the float32 the encoder wants.
+        wav = np.frombuffer(raw, dtype="<i2").astype(np.float32) / 32768.0
+        if wav.size == 0:
+            self._send(200, {"text": "", "confidence": 0.0})
+            return
+
+        t0 = time.perf_counter()
+        try:
+            with _lock:
+                res = _model.transcribe(wav, sr=rate, language=lang, mode=MODE)
+        except Exception as exc:  # noqa: BLE001 - the caller falls back to mavsttd
+            log.exception("transcribe failed")
+            self._send(500, {"error": str(exc)})
+            return
+        elapsed = time.perf_counter() - t0
+        text = (res.text or "").strip()
+        log.info("%.2fs audio in %.2fs: %r", wav.size / rate, elapsed, text[:60])
+        # The model reports no calibrated score. 1.0 would be a claim, and the
+        # Maven side reads confidence only to log it.
+        self._send(200, {"text": text, "confidence": 0.0})
+
+
+def main():
+    if not TOKEN:
+        log.warning("no CW2_TOKEN set: anything on the LAN can post audio here")
+    # Bind before loading, so a restart answers 503 rather than refusing the
+    # connection. Both make Maven fall back, but only one of them says why.
+    srv = ThreadingHTTPServer((HOST, PORT), Handler)
+    threading.Thread(target=load_model, daemon=True).start()
+    log.info("listening on %s:%d", HOST, PORT)
+    srv.serve_forever()
+
+
+if __name__ == "__main__":
+    main()

deploy/maven-voice-tunnel.service

@@ -0,0 +1,40 @@
+# maven-voice-tunnel — the ssh leg that carries the voice wire to homesrv.
+#
+# Runs on workpc, as a user unit (`systemctl --user`), beside mavgpud.service.
+#
+# WHY THIS EXISTS AT ALL. internal/voice is plaintext and unauthenticated.
+# Its own server doc says production binds inside the wg tunnel, because "the
+# wg layer IS the L0 floor". workpc is not a wg peer, it sits on wlan0. So ssh
+# is the substitute floor: it authenticates with his key and encrypts the leg,
+# and mavend's published port stays on homesrv loopback (127.0.0.1:9110).
+# Nothing about this puts a Maven port on the LAN.
+#
+# Do not replace this with a LAN bind. SurfaceVoice caps acts at L0, so an
+# unauthorized speaker could not run a destructive tool. It would still hear
+# his facts, his notes and his calendar read back, and L0 does not cap reading.
+#
+#   install: cp to ~/.config/systemd/user/ on workpc
+#            systemctl --user enable --now maven-voice-tunnel.service
+
+[Unit]
+Description=SSH tunnel to mavend's voice wire on homesrv
+After=network-online.target
+Wants=network-online.target
+
+[Service]
+# -N: no remote command, forwarding only.
+# ExitOnForwardFailure: fail loudly rather than sit up with a dead forward,
+# which is what makes Restart meaningful.
+# ServerAlive*: a laptop that suspends drops the tunnel silently otherwise.
+ExecStart=/usr/bin/ssh -N \
+  -o ExitOnForwardFailure=yes \
+  -o ServerAliveInterval=30 \
+  -o ServerAliveCountMax=3 \
+  -o BatchMode=yes \
+  -L 127.0.0.1:9100:127.0.0.1:9110 \
+  kami@192.168.1.104
+Restart=always
+RestartSec=5
+
+[Install]
+WantedBy=default.target

deploy/mavend.json

@@ -78,10 +78,31 @@
     "Addressed by LAN address, not container name: mavgpud runs on another",
     "machine and there is no shared docker network to name it on."
   ],
+  "//workstation.stt": [
+    "CrisperWhisper 2.0 turbo on the same machine, a second service on port",
+    "8081 and not a second endpoint on mavgpud. whisper.cpp cannot load CW2 at",
+    "all: it derives its language count from the vocabulary size, and CW2's",
+    "51897 tokens shift seven special token ids. So it runs under transformers",
+    "there and mavsttd stays whisper.cpp here.",
+    "Worth the second service: CW2 turbo scores 10.4% WER in Russian against",
+    "27.5% for the ggml-small.bin mavsttd loads, measured on 200 Golos clips",
+    "in docs/evals/2026-08-09-crisperwhisper2-russian-wer.md.",
+    "Deleting this block sends every utterance to mavsttd, which is what the",
+    "box did before it existed. A worse transcript is still a turn, so the",
+    "fallback is silent and Kami is never told which machine heard him.",
+    "The token is what stops anything on the LAN posting audio to that port."
+  ],
   "workstation": {
     "url": "http://192.168.1.105:8080",
+    "token": "${MAVEN_GPU_TOKEN}",
     "probe": "15s",
-    "timeout": "90s"
+    "timeout": "90s",
+    "stt": {
+      "url": "http://192.168.1.105:8081/transcribe",
+      "token": "${MAVEN_STT_TOKEN}",
+      "probe": "15s",
+      "timeout": "10s"
+    }
   },
 
   "//search": [
@@ -231,7 +252,8 @@
     "embedder": {
       "model_path": "/opt/maven/models/embedder/multilingual-e5-small/model_quantized.onnx",
       "tokenizer_path": "/opt/maven/models/embedder/multilingual-e5-small/tokenizer.json",
-      "lib_path": "/opt/maven/lib/libonnxruntime.so"
+      "lib_path": "/opt/maven/lib/libonnxruntime.so",
+      "heads_path": "/opt/maven/models/embedder/router-heads/router_heads.onnx"
     },
     "llm_router": true,
     "query_min_score": 0.55,

deploy/mavgpud.json

@@ -1,10 +1,24 @@
 {
   "listen": ":8080",
+  "//token_file": [
+    "The bearer token every request must carry. homesrv is the client and it",
+    "is on the LAN, so this port cannot be loopback and the token is what",
+    "stops anything else on the network spending the card or reading /slots.",
+    "A path, never the token: mavgpud refuses to start when listen is",
+    "reachable from the network and this is unset.",
+    "Same value as MAVEN_GPU_TOKEN in homesrv's deploy/telegram.env."
+  ],
+  "token_file": "/home/kami/.config/mavgpud.token",
   "llama_addr": "127.0.0.1:10000",
   "llama_bin": "llama-server",
+  "//llama_args": [
+    "E4B carries no MTP tensors, so the speculative flags are gone with the 12B.",
+    "MTP on this box is a separate gguf of architecture gemma4-assistant with",
+    "nextn_predict_layers=4, and mtp-gemma-4-12B-it-BF16 is the only one there is.",
+    "Its head is trained against the 12B's hidden states, so it cannot drive E4B."
+  ],
   "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",
+    "-m", "/mnt/D/AI/gemma4/gemma-4-E4B-it-qat-UD-Q4_K_XL.gguf",
     "-ngl", "99",
     "-fa", "on",
     "-np", "1",
@@ -15,11 +29,22 @@
     "--batch-size", "2048",
     "--ubatch-size", "512",
     "--jinja",
-    "--chat-template-kwargs", "{\"enable_thinking\":false}",
-    "--spec-type", "draft-mtp",
-    "--spec-draft-n-max", "2"
+    "--chat-template-kwargs", "{\"enable_thinking\":false}"
   ],
 
+  "//stt": [
+    "CrisperWhisper 2.0 turbo, which Maven reaches directly on port 8081.",
+    "mavgpud runs it because it is a ROCm process on this card: under its own",
+    "systemd unit it registered on the KFD and the supervisor evicted",
+    "llama-server every few seconds. CW2_TOKEN comes from the unit's",
+    "EnvironmentFile and is never a flag value."
+  ],
+  "stt": {
+    "addr": "127.0.0.1:8081",
+    "bin": "/home/kami/Programs/cw2-eval/.venv/bin/python",
+    "args": ["/home/kami/Programs/cw2-service/serve.py"]
+  },
+
   "kfd_root": "/sys/class/kfd/kfd/proc",
   "drm_device": "/sys/class/drm/card1/device",
 

deploy/mavgpud.service

@@ -1,5 +1,5 @@
 [Unit]
-# Runs on the workstation (bugmachine), not on homesrv. Install as a systemd
+# Runs on the workstation (workpc), 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:
 #
@@ -8,10 +8,15 @@
 #   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)
+Description=Maven GPU supervisor (holds llama-server and CW2 while the card is free)
 After=network.target
 
 [Service]
+# CW2_TOKEN for the transcriber child, which inherits this environment. The
+# token is read from a file and never appears as a flag value, the rule
+# mavpoll and mavmaild follow. Missing file, no transcriber auth, so keep the
+# dash off: a mavgpud that cannot read it must fail loudly.
+EnvironmentFile=%h/Programs/cw2-service/cw2.env
 ExecStart=%h/.local/bin/mavgpud -config %h/.config/mavgpud.json
 Restart=always
 RestartSec=5

deploy/mavwaked.service

@@ -0,0 +1,76 @@
+# mavwaked — always-on listening, on workpc where the microphone is.
+#
+# User unit, beside mavgpud.service and maven-voice-tunnel.service. It is a
+# user unit because it needs his ALSA session and his ssh agent, and because
+# it should stop when he logs out.
+#
+# The keyword is "Мэйвен" and the three -wake- flags are what require it
+# (V-487 stage two). Without them anything spoken near the fifine becomes a
+# turn, which voiceSender makes safe rather than expensive: it sends
+# Surface=SurfaceVoice, capping every command at L0. That does not stop her
+# answering out loud, which is the whole reason the keyword exists.
+#
+# The threshold is 0.999 and it is the binary's default, so it is not passed.
+# It came from 65 minutes of held-out Russian speech through this same binary:
+# 0.9 false wakes an hour against 2.8 at 0.99, for one lost render out of 126
+# (docs/evals/2026-08-09-wake-word.md). If the room proves noisier than the
+# corpus, read the scores out of this unit's journal and pass -wake-threshold.
+# Do not lower it by guessing.
+#
+# A keyword shorter than 1.32s can be heard too late to be used, because the
+# head scores 1.28s of audio and the VAD has closed the utterance by then.
+# "Мэйвен, <request>" is unaffected. A bare "Мэйвен" is the case that fails.
+#
+# -vad-model is passed on purpose. Silero answers "is this frame speech" where
+# the energy floor answers "is this frame loud". It declines white noise at
+# the same RMS 0 frames to 68-99, and still hears all four spoken fixtures
+# (docs/evals/2026-08-09-silero-vad.md). It costs 509us a frame, 1.7% of one
+# core, and never touches the GPU. Drop the flag and the energy floor is back.
+#
+# -barge-in is NOT passed. The threshold is room-specific and this room has no
+# number yet. Turn it on only after reading the "suppressed while speaking"
+# means out of this unit's own journal, never by guessing.
+#
+#   install: cp to ~/.config/systemd/user/ on workpc
+#            systemctl --user enable --now mavwaked.service
+
+[Unit]
+Description=Maven always-on listening (silero VAD, "Мэйвен" keyword)
+# The tunnel is the only path to mavend and the only thing authenticating it.
+Requires=maven-voice-tunnel.service
+After=maven-voice-tunnel.service
+
+[Service]
+# The Scarlett Solo 4th Gen, and not the fifine. The fifine was the device
+# here for three days and mavwaked never logged one utterance in them, because
+# it returns RMS 0.00004 with its capture switch on and its ALSA volume at the
+# full 496 of 496. That silence is in the hardware, so no flag reaches it.
+#
+# Named CARD=Gen and not card 4, because a USB card number moves when
+# something else is replugged and this daemon must not change ears quietly.
+# Not "default" either: that follows whatever pipewire last decided.
+#
+# plughw and not hw. mavwaked asks arecord for 16kHz mono, which is what the
+# whole pipeline is canonical in. Neither microphone offers it, so bare hw
+# dies on "Channels count non available" before a frame is read. plughw puts
+# ALSA's downmix and resampler in front. Any replacement wants the same.
+#
+# The Scarlett measured RMS 0.003 against 0.14 on the onboard input, so its
+# front-panel gain is the thing to raise if she mishears. That is a knob, not
+# a control ALSA exposes. The two loud devices, the onboard ALC897 and the
+# camera, both clip at peak 1.0 and are worse candidates, not better ones.
+Environment=LD_LIBRARY_PATH=%h/.local/lib
+ExecStart=%h/.local/bin/mavwaked \
+  -device plughw:CARD=Gen,DEV=0 \
+  -addr 127.0.0.1:9100 \
+  -lang ru \
+  -vad-model %h/.local/share/maven/models/silero_vad.onnx \
+  -wake-model %h/.local/share/maven/models/maven_wakeword.onnx \
+  -wake-mel %h/.local/share/maven/models/melspectrogram.onnx \
+  -wake-embed %h/.local/share/maven/models/embedding_model.onnx \
+  -onnx-lib %h/.local/lib/libonnxruntime.so
+Restart=on-failure
+RestartSec=5
+
+[Install]
+WantedBy=default.target

deploy/telegram.env.example

@@ -10,3 +10,12 @@ TELEGRAM_CHAT_ID=
 #   ntfy token add --expires=never maven
 # Read access is not needed — mavend publishes and never subscribes.
 NTFY_TOKEN=
+
+# Bearer token for mavgpud, the workstation's GPU supervisor (V-673). It fronts
+# the big model on a LAN port, so the token is the whole boundary in front of
+# the card. Any long random string; mint one with:
+#   openssl rand -hex 32
+# The same value goes in a file on workpc, named by token_file in
+# deploy/mavgpud.json. Unset here and every workstation turn falls back to the
+# resident model, because mavgpud answers 401 and Maven reads that as down.
+MAVEN_GPU_TOKEN=

docker-compose.yml

@@ -53,6 +53,19 @@ services:
     # the decrypted working copy lives in RAM (see db_tmpfs in mavend.json).
     tmpfs:
       - /dev/shm
+    # the voice wire, for mavwaked and mavenclient on workpc (V-515).
+    #
+    # LOOPBACK ONLY, and that is the whole security argument. internal/voice
+    # is plaintext with no auth: its own server doc says production binds
+    # inside the wg tunnel, "the wg layer IS the L0 floor". workpc is not a wg
+    # peer, it is on wlan0. So the tunnel is ssh instead, terminated on this
+    # loopback address, and nothing new is on the LAN. Anyone who could reach
+    # a LAN-bound port here could push audio and hear his facts read back.
+    # SurfaceVoice caps acts at L0; it does not cap reading.
+    #
+    # Host 9100 is Vikunja's MCP, hence 9110. The container side stays 9100
+    # so mavweb keeps reaching mavend:9100 by name.
+    ports: ["127.0.0.1:9110:9100"]
 
   mavsttd:
     <<: *image

docs/CLAUDE.md

@@ -0,0 +1,30 @@
+# docs/
+
+Everything an agent needs that is not a rule and not code. `CLAUDE.md` at the
+root carries the rules and points here. Nothing here restates a rule.
+
+The tier is the path, so staleness is visible from the filename.
+
+| path | holds | lifetime |
+| --- | --- | --- |
+| `docs/*.md` | living. One file per subsystem: the reasoning, corrected in place. Each carries `Last verified: <date> @ <sha>`. | until it is wrong |
+| `docs/evals/` | dated measurements, one file per measurement. **Never edited after the day.** A newer number is a new file. | forever |
+| `docs/caveats/` | known limits, one entry per limit, each with a task id and a revisit trigger. Indexed in `docs/caveats/CLAUDE.md`. | until fixed, then deleted |
+| `docs/plans/` | the plan for one piece of work, frozen once it starts | until the work lands |
+| `docs/archive/` | dead. Read by nobody by default. | forever |
+
+## Rules for this directory
+
+* One fact, one home. A measurement is cited from a living doc, never copied
+  into it. The two drift the moment they are both edited.
+* A living doc is corrected in place and its `Last verified` line moves with the
+  correction. Do not append a changelog to it.
+* A number in prose with no `docs/evals/` file behind it is an opinion.
+* Fixing something deletes its caveat. It does not edit the eval that found it.
+
+## Where a subsystem's reasoning lives
+
+`routing.md`, `language.md`, `world.md`, `offload.md`, `deployment.md`,
+`ecosystem.md`, `workflow.md`, `design.md`, `rearchitecture.md`,
+`determinism.md`, `protocol.md`, `handler-wiring.md`, `operations.md`, `qa.md`.
+The root `CLAUDE.md` says which one to read before touching what.

docs/caveats/CLAUDE.md

@@ -0,0 +1,50 @@
+# docs/caveats/
+
+One entry per known limit: something broken, deferred or unsafe that a session
+will otherwise walk into. An entry names what fails, who it costs, and the
+condition that makes it worth fixing.
+
+Two things do not belong here. The evidence is a dated file under `docs/evals/`.
+The reasoning behind a subsystem is its living doc directly under `docs/`. A
+caveat is the pointer between them plus the trigger.
+
+## Rules for this directory
+
+* One file per area, one `##` section per limit, each carrying its task id.
+* **A caveat with no revisit trigger is a complaint.** Give it one or delete it.
+* Closing a limit deletes its entry. It does not edit it to say "fixed", and it
+  never edits the frozen measurement it came from. The durable record of a fix
+  is the commit and the subsystem's living doc.
+* An entry whose task is closed but whose limit is still live is the failure
+  mode to watch for. The id joins the two directions, so check both.
+
+## Index
+
+Every entry below came from the 2026-08-10 deep audit
+(`docs/evals/2026-08-10-repo-audit.md`), except the last, which came from wiring
+the gate the audit asked for. Five of the twenty findings are fixed and have no
+entry. The unauthenticated mavgpud proxy was V-673. The 20 reachable advisories
+in the toolchain and `x/text` were V-682. The missing analyzers were V-694, and
+what they now report is the baseline entry under V-701. The two unguarded
+invariants were V-692 and V-693, and their guards are described in
+`docs/routing.md`.
+
+| limit | severity |
+| --- | --- |
+| [Anyone past the proxy can enroll a passkey](security.md#enrollment) | high |
+| [Passkey credentials are rewritten in place](security.md#credentials) | medium |
+| [An empty STT transcript reads as a successful one](external-inputs.md#stt) | medium |
+| [Open-Meteo's empty body becomes 0°C](external-inputs.md#weather) | medium |
+| [Dialogue persistence errors are swallowed](storage.md#dialogue) | medium |
+| [The reminder transition is a lost update](storage.md#reminders) | medium |
+| [A recall miss scans two whole tables](storage.md#recall) | medium |
+| [Closing a TCP listener can strand Accept](transport.md#accept) | medium |
+| [PTT reads an unbounded body](transport.md#ptt) | medium |
+| [mavweb errors cannot be traced](transport.md#errors) | medium |
+| [Fact enrichment is a 20-call serial waterfall](workers.md#enrichment) | medium |
+| [A suppressed nudge is phrased anyway](workers.md#nudges) | medium |
+| [Committed absolute paths pin the build to this box](config.md#paths) | medium |
+| [The env example omits deployed variables](config.md#secrets) | medium |
+| [The analyzers pass against a baseline, not zero](dependencies.md#baseline) | medium |
+| [Domain packages depend on store and IPC types](layering.md#dtos) | low |
+| [Eleven symbols are unreachable](layering.md#deadcode) | low |

docs/caveats/config.md

@@ -0,0 +1,21 @@
+# Configuration and environment
+
+## Committed absolute paths pin the build to this box [#690] {#paths}
+
+Costs: `go.mod` replaces Hexis with `/home/kami/apps/hexis`, `start-maven.sh`
+hardcodes the checkout and the data directory, and `deploy/mavgpud.json` holds
+workstation model and Python paths. Vendoring hides the `go.mod` problem for an
+ordinary build. `-mod=mod`, `go mod tidy` and a fresh checkout all fail.
+Revisit when: anyone clones this repo elsewhere, or a `tidy` is needed.
+Workaround: build only from this checkout, with the vendor directory.
+
+## The env example omits deployed variables [#691] {#secrets}
+
+Costs: a fresh deploy can lose remote speech-to-text or ambient authentication
+and run on fallback behaviour with an apparently valid config. Three variables
+are referenced and undocumented: `MAVEN_STT_TOKEN`, `MAVEN_AMBIENT_TOKEN` and
+`CW2_TOKEN`. V-673 added `MAVEN_GPU_TOKEN` to the example.
+It is not silent. The loader logs which variables were unset and says whatever
+they configure is off. What is missing is a startup failure.
+Revisit when: the box is redeployed from scratch, or a new secret is added.
+Workaround: read that log line at startup.

docs/caveats/dependencies.md

@@ -0,0 +1,13 @@
+# Dependencies
+
+## The analyzers pass against a baseline, not against zero [#701] {#baseline}
+
+Costs: `make lint` and `make deadcode` are wired and green (V-694), but green
+means "nothing new since 2026-08-11". The accepted set is 19 staticcheck
+findings and 13 unreachable symbols, listed with a reason each in
+`scripts/analyzers/*.baseline`. Three of the unreachable symbols must stay:
+[layering.md](layering.md#deadcode). One accepted staticcheck finding is V-687.
+Revisit when: V-701 sweeps the baseline, or a fix deletes an entry. The gate
+fails on an entry whose finding is gone, so the deletion is not optional.
+Workaround: none needed. Reachability claims are checkable now. Read the
+baseline before trusting that a target reporting clean means the tree is clean.

docs/caveats/external-inputs.md

@@ -0,0 +1,20 @@
+# External inputs
+
+What arrives from a service Maven does not run, and what happens when it
+arrives malformed. The shared shape: a JSON decode into value fields cannot
+tell "absent" from "zero", so a degraded response becomes a confident answer.
+
+## An empty STT transcript reads as a successful one [#675] {#stt}
+
+Costs: one dropped voice turn per malformed 200 from workpc. The mavsttd floor
+is never asked, because `stt.Pair` falls back on a non-nil error alone.
+Revisit when: CW2 returns a 200 with no text. Sooner if a proxy is put between
+homesrv and port 8081.
+Workaround: none. It is silent by design and the fallback is never spoken.
+
+## Open-Meteo's empty body becomes 0°C [#676] {#weather}
+
+Costs: he is told the weather is clear and 0°C when the service answered
+nothing. Distinct from V-589, which covered the HTTP status and not the body.
+Revisit when: a weather answer is reported as wrong, or the geocoder changes.
+Workaround: none.

docs/caveats/layering.md

@@ -0,0 +1,25 @@
+# Layering and dead surface
+
+Neither entry breaks anything today. Both make a later change cost more than it
+should, which is why they are low and not medium.
+
+## Domain packages depend on store and IPC types [#685] {#dtos}
+
+Costs: dialogue exposes `store.DialogueSessionRow` in its port, the pure
+morning planner takes a `store.Fact`, and auth policy imports IPC method and
+caller types. There is no Go import cycle. A schema change reaches further than
+it should.
+Revisit when: the dialogue or fact schema changes, or a second transport
+appears beside IPC.
+Workaround: none needed. It compiles and it is correct.
+
+## Eleven symbols are unreachable [#686] {#deadcode}
+
+Costs: extra API and test surface, and comments that claim callers which no
+longer exist. Three of the eleven must not be deleted. `HisGender` is a
+documented seam tied to V-399. `AudioDuration` duplicates `internal/audio` and
+should call it. `CountWord` is a one-line alias nobody uses and can go.
+Revisit when: `deadcode` is wired into the audit gate, which needs the
+allowlist this entry describes. **An unannotated list invites deleting the
+three above.**
+Workaround: none needed.

docs/caveats/security.md

@@ -0,0 +1,26 @@
+# Security
+
+Both entries are mavweb's passkey seam. Assertion itself is sound and is not
+the problem: `userVerification` is required, and a sign count that does not
+increase is rejected.
+
+## Anyone past the proxy can enroll a passkey [#683] {#enrollment}
+
+Costs: registration is gated on nothing, so any client that reaches mavweb can
+enroll its own key and become him. Step-up is worse than per-client: one
+process-global `assertedAt` means every client inherits the same five-minute
+window after any successful assertion. The voice WebSocket accepts every
+origin, which makes cross-site use easier. V-317 covers which routes are gated
+and V-605 covers challenge-map growth. Neither covers this.
+Revisit when: mavweb is reachable from anything but the tunnel, and before any
+new credential is enrolled.
+Workaround: the reverse proxy is the only boundary today. That is the finding.
+
+## Passkey credentials are rewritten in place [#684] {#credentials}
+
+Costs: `os.WriteFile` over the live file. A crash, a full disk or an
+interrupted write corrupts every enrolled credential at once, and mavweb will
+not start afterwards.
+Revisit when: a second credential is enrolled, since the blast radius grows
+with the count. Sooner if the box loses power unexpectedly.
+Workaround: back the file up before enrolling.

docs/caveats/storage.md

@@ -0,0 +1,30 @@
+# Storage
+
+The DB seam: what it loses quietly, and what it reads more of than it needs.
+
+## Dialogue persistence errors are swallowed [#677] {#dialogue}
+
+Costs: restart continuity can vanish with nothing in the log, and a failed
+delete can bring stale conversation state back. Current-turn dialogue is
+unaffected, which is why this has never been noticed.
+Revisit when: a restart is reported as losing context. Sooner if a turn starts
+reading dialogue rows back to him.
+Workaround: none. The failure is invisible from outside.
+
+## The reminder transition is a lost update [#678] {#reminders}
+
+Costs: a concurrent fire and cancel both succeed and the last writer wins.
+Medium today because cancellation has no surface. High the moment V-622 adds
+one, and V-622 does not describe this invariant.
+Revisit when: V-622 starts, whichever comes first.
+Workaround: none, but the window is small while nothing can cancel.
+
+## A recall miss scans two whole tables [#681] {#recall}
+
+Costs: every missed recall reads all of `memory_vectors` and then decodes and
+sorts every note vector. Not an N+1, and the memory scan is cheap per losing
+row on purpose. The duplicated decode is the legacy notes path alone.
+Revisit when: the note count makes a miss measurably slow. Also when the two
+exclusion filters are proven to agree. `QueryNotes` uses `notHisWordsSQL` and
+`Search` uses `memory.NonRecallPrefix`. The fallback cannot go until they match.
+Workaround: none needed at today's row counts.

docs/caveats/transport.md

@@ -0,0 +1,34 @@
+# Transport
+
+The HTTP and socket seams. What a client can do to them, and what a shutdown
+can do to us.
+
+## Closing a TCP listener can strand Accept [#679] {#accept}
+
+Costs: during close, both `errc` and `done` are ready in `acceptLoop`'s select.
+Go picks uniformly. So roughly one close in two leaves a waiting `Accept`
+blocked forever on a TCP seam. Unix sockets are unaffected.
+Revisit when: a daemon is seen hanging on shutdown, or before any new TCP
+listener is added.
+Workaround: the process usually exits anyway, which hides it.
+
+## PTT reads an unbounded body [#688] {#ptt}
+
+Costs: `handlePTT` does an unlimited `io.ReadAll`, and mavweb sets no header or
+idle timeouts. A client can force unbounded allocation or hold a connection
+open. mavgpud's half of this was fixed in V-673.
+Revisit when: mavweb is reachable from anything but the tunnel.
+Workaround: mavweb is not LAN-exposed today.
+
+`/ws` rides along with this entry. It never calls `SetReadLimit`, so the
+dependency default of 32,768 bytes applies, about a second of audio. Nothing
+reaches it: the browser posts PCM to `/api/ptt`, and only `handlers_test.go`
+opens `/ws`. It gets a caller and a real limit, or it gets deleted.
+
+## mavweb errors cannot be traced [#689] {#errors}
+
+Costs: some handlers return the raw internal error, which discloses internals.
+Others return a generic one with no identifier, which cannot be joined to its
+log line. There is no request-id middleware to join them.
+Revisit when: a reported UI failure cannot be found in the log.
+Workaround: read the log by timestamp.

docs/caveats/workers.md

@@ -0,0 +1,23 @@
+# Background workers
+
+Both entries are a tick doing expensive work it did not need to do.
+
+## Fact enrichment is a 20-call serial waterfall [#680] {#enrichment}
+
+Costs: each fact is resolved in turn and each ecosystem call can spend ten
+seconds. A slow but reachable Nexus holds one tick for minutes, so the worker
+stops observing its configured interval. V-647 covered the duplicate queue
+scan, not this.
+Revisit when: Nexus gets slow, or when a batch-resolution endpoint exists.
+Workaround: an unreachable Nexus is fine. It is the slow-but-answering case
+that hurts.
+
+## A suppressed nudge is phrased anyway [#687] {#nudges}
+
+Costs: `PhraseNudge` runs before the dedupe is known. The `continue` meant to
+skip it is the last statement in the loop body. Every tick that
+keeps suppressing the same rule pays the resident model again. The comment
+above it claims the opposite.
+Revisit when: digestion ticks show up in the model's load, or when nudge rules
+grow past a handful.
+Workaround: none.

docs/deployment.md

@@ -0,0 +1,220 @@
+# Deployment: the boxes, the models, the daemons
+
+*Last verified: 2026-08-09 @ a9b480a*
+
+What runs where, and why each choice was made. `CLAUDE.md` carries only the
+rules. This file carries the reasoning.
+
+## The two boxes
+
+**homesrv** is a Ryzen 5 5600U laptop and the deploy target. It offloads to the
+Vega iGPU over Vulkan (`n_gpu_layers: 99`). Compose passes `/dev/dri` and the
+render gid (993), and without both Vulkan enumerates zero devices and
+llama-server falls back to CPU silently.
+
+**workpc** is the workstation, 16GB of VRAM, reached as `kami@workpc` at
+192.168.1.105. Model work moved there on 2026-08-02 by the owner's call, because
+homesrv cannot grow a GPU.
+
+Three rules govern the seam:
+
+- **The workstation is never assumed up.**
+- **Fall back silently** when it would only do the job better.
+- **Name the gap** when the resident model cannot do the job at all.
+
+A world question goes through `LLMPhraser.PhraseWorld` and returns `worldGap`
+(`cmd/mavend/worldmodel.go`) rather than an invented answer. A box with no
+`workstation` block behaves exactly as it did before the seam. `docs/offload.md`
+says which caller is which.
+
+## The resident model
+
+**Qwen3-1.7B** (`UD-Q4_K_XL`), stock, not yet the CPT'd one. It is a Thinking
+variant, so `n_ctx` is 4096. Reasoning tokens need the room, and 4096 is what
+every score was measured at.
+
+The target is the locally CPT'd Qwen3-1.7B (V-122, training in flight). Stock
+already speaks good Russian. What it gets wrong is the persona. It writes `я рад`
+where Maven needs `рада`.
+
+**Do not bother with sub-500M models.** LFM2.5-230M and 350M were measured on
+2026-07-31 and both are unusable in Russian
+(`docs/evals/2026-07-31-model-bakeoff.md`). Their published IFEval and BFCL
+numbers are English-only.
+
+Model files live in `/mnt/hdd1/llms`, bind-mounted to `/opt/maven/models/llm`.
+That **shadows** the repo's `models/llm/`, so a gguf sitting there is not loaded
+by anything. Swapping the resident model is a one-line change to
+`phraser.model_path` in `deploy/mavend.json`.
+
+The workstation model is gemma-4-E4B as of 2026-08-09, replacing the 12B by the
+owner's call. Keep the 12B gguf. It is the better teacher for label runs, at
+72.7% destination against E4B's 57.6%.
+
+## The embedder
+
+**It stays on homesrv permanently**, because it backs the floor. It is
+multilingual-e5-small, quantized and asymmetric. `EmbedQuery` and `EmbedPassage`
+apply the `query:` and `passage:` prefixes it was trained with. Calling plain
+`Embed` on a note is a bug. See `docs/evals/2026-08-04-recall-e5-small.md`.
+
+The vendored onnxruntime under `deps/` has two copies, and the stale one is
+1.17.1. The live runtime is 1.26.0, and the Go binding asks for API 26. Anything
+shipped to another box needs `deps/onnxruntime-linux-x64-1.26.0`.
+
+## Speech-to-text
+
+`sttSeam` in `cmd/mavend/voicewire.go` builds an `stt.Pair` beside `modelSeam`.
+It prefers CrisperWhisper 2.0 turbo on workpc with mavsttd as the floor. It takes
+only the silent half of the rule, because a worse transcript is still a turn. So
+`stt.Pair` has no `TranscribeRemote` and the fallback is never spoken.
+
+CW2 turbo scores 10.4% WER in Russian against 27.5% for the `ggml-small.bin`
+mavsttd loads, over 200 Golos clips
+(`docs/evals/2026-08-09-crisperwhisper2-russian-wer.md`).
+
+**whisper.cpp cannot load CW2 at all.** It reads its language count off the
+vocabulary size. CW2's 51897 tokens shift seven special token ids. So CW2 is its
+own transformers service on port 8081 (`deploy/cw2/serve.py`).
+`stt.HTTPTranscriber` posts raw PCM to it with a bearer token, because audio is
+the most sensitive thing that crosses this seam. The switch is `workstation.stt`
+in `deploy/mavend.json`, and deleting the block sends every utterance to mavsttd.
+
+**mavgpud runs that service as a second child.** This is not an optimisation.
+CW2 is a ROCm process on the same card, so it registers on the KFD like any
+contender. Under its own systemd unit it made mavgpud evict llama-server every
+few seconds. That took the model arm down for eight minutes on 2026-08-09. The
+card needs one owner. CW2 is on the yield clock and not the idle one. At 1.6GB
+it denies the card to nobody.
+
+Text-to-speech has not moved. piper on homesrv is the only synthesizer.
+
+## The daemons
+
+| Binary | Role |
+|---|---|
+| `mavend` | **Core.** Router, phraser, memory, reminders, digestion tick. Owns the DB and IPC socket. |
+| `mavweb` | HTTP UI and PWA (`/dash`, `/history`, `/trace`, `/notifications`, `/tools`), WebAuthn auth. |
+| `mavsttd` | Speech-to-text (whisper.cpp, CGO). |
+| `mavttsd` | Text-to-speech (piper subprocess). |
+| `mavwaked` | Wake-word and VAD gate. Runs on workpc. |
+| `mavenclient` | Voice loop client (mic, stt, core, tts). Not deployed. |
+| `mavpoll` | Environment poller: netdata alarms, uptime-kuma, zenmoney, wireguard presence. Writes facts, sends nothing. Telegram is `internal/delivery/telegramsink`. |
+| `mavcaldav` | CalDAV calendar sync. |
+| `mavmaild` | Mail reader (IMAP, read-only). Holds the IMAP password, core never sees it. |
+| `mavgpud` | GPU supervisor. **Runs on workpc**, own unit `deploy/mavgpud.service`. Keeps llama-server loaded while the card is free (V-488). Maven never asks it for anything and reads `/health` through `llm.Pair`. Its LAN port needs a bearer token in the file named by `token_file`, matching `MAVEN_GPU_TOKEN` on homesrv, or it refuses to start (V-673). |
+| `mavupdate` | Not a daemon. Operator CLI a human runs on the box to deploy a new build. |
+
+Two binaries have no Makefile target and neither is deployed. `mavseal` encrypts
+a live tmpfs working copy back to the ciphertext file when mavend was killed
+before `defer st.Close()` sealed it. `labelgen` runs the stage 0 grammars over
+utterances and prints JSONL, the training data for the routing heads.
+
+Daemons are wired socket-to-socket, not linked. `internal/ipc` is the wire
+protocol. `deploy/mavend.json` sets socket paths, model paths and the phraser and
+embedder blocks, with `${VAR}` expansion from gitignored `deploy/telegram.env`.
+
+### Who is in compose, and who is not
+
+**`docker-compose.yml` runs five**: `mavend`, `mavsttd`, `mavttsd`, `mavweb`,
+`mavpoll`. Count against compose, not against the table above.
+
+`mavmaild` and `mavcaldav` are commented out, each with the reason beside it. The
+first needs a mail account and the second a CalDAV account, and this box has
+neither. Two things ride on the CalDAV absence (V-644). Agenda questions route to
+`IntentQuery` at stage 0, and the `calendar` query source then reads a table
+nobody writes. And `loop.State.CalendarBusy` is fed by the same facts, so the
+gate's "do not nag mid-meeting" is permanently false.
+
+`mavenclient` is still absent. `mavwaked` moved to workpc on 2026-08-09 (V-515).
+
+### The voice wire
+
+`internal/voice` is plaintext with no auth. Its own server doc says production
+binds inside the wg tunnel, because the wg layer is the L0 floor. workpc is not
+a wg peer, it sits on wlan0. So the tunnel is ssh instead.
+
+mavend publishes the voice port to homesrv loopback only, `127.0.0.1:9110`.
+Host 9100 is Vikunja's MCP, hence 9110. The container side stays 9100 so mavweb
+keeps reaching `mavend:9100` by name. `deploy/maven-voice-tunnel.service` on
+workpc forwards it over his key.
+
+**Do not replace this with a LAN bind.** `SurfaceVoice` caps acts at L0, so an
+unauthorized speaker could not run a destructive tool. L0 does not cap reading,
+so they would still hear his facts, notes and calendar read back.
+
+Both `mavwaked` and `mavenclient` speak `voice.Dial`, not `ipc.Dial`. The
+`netaddr` token guards the daemon-to-daemon IPC seam and never touches this one.
+`ipc.Dial` does take `tcp://host:port?token=...`, which is why V-515 was filed
+as a config change. That premise was wrong, and the ssh leg is the correction.
+
+The voice loop belongs on a client machine where the owner is standing, and that
+machine is workpc (V-463, `docs/plans/17-where-the-voice-loop-runs.md`). homesrv
+has a microphone, because it is a laptop, but it is in the wrong room.
+
+### mavwaked on workpc
+
+`deploy/mavwaked.service`, a user unit beside `mavgpud.service`. Two flags are
+deliberate.
+
+`-vad-model` is passed. Silero answers "is this frame speech" where the energy
+floor answers "is this frame loud". It declines white noise at the same RMS, 0
+frames against 68 to 99, and still hears all four spoken fixtures
+(`docs/evals/2026-08-09-silero-vad.md`). It costs 509µs a frame and never touches
+the GPU. A model that will not load is logged and not fatal.
+
+`-barge-in` is not passed. The threshold is room-specific and this room has no
+number yet. Read the "suppressed while speaking" means out of the journal first.
+
+The device is `plughw:CARD=Gen,DEV=0`, the Scarlett Solo. It is `plughw` and
+not `hw` because mavwaked asks arecord for 16kHz mono. No microphone here
+offers that, so bare `hw` dies on "Channels count non available" before a
+frame is read. It is named `CARD=Gen` and not card 4 because a USB card number
+moves when something else is replugged.
+
+It used to be the fifine on card 0, and that cost three days. mavwaked logged
+zero completed utterances across them, before the wake word existed and after.
+The fifine returns RMS 0.00004 with its capture switch on and its ALSA volume
+at the full 496 of 496. That silence is in the hardware and no flag reaches
+it. Over the same eight seconds of speech the onboard ALC897 read 0.142 and
+the camera 0.289, both clipping at peak 1.0. The Scarlett read 0.003 clean.
+
+Check the level before blaming the gate. Stop the unit, run `arecord` against
+the device for five seconds, and measure. A live room floor reads near 0.001.
+
+The three `-wake-` flags require the keyword "Мэйвен" (V-487 stage two). Drop
+them and the loop runs open, which is what it did before. The threshold is the
+binary's default of 0.999 and is not passed. Over 65 minutes of held-out
+Russian speech it woke her 0.9 times an hour against 2.8 at 0.99. That cost one
+lost render out of 126 (`docs/evals/2026-08-09-wake-word.md`).
+
+The three sessions are pinned to one thread each. onnxruntime otherwise sizes
+its pool to every core and spins between runs, which took mavwaked from 68% of
+one core to 335%. With the cap it sits at 81%, so the gate costs about 13%.
+
+A keyword shorter than 1.32s can be heard too late to be used. The head scores
+1.28s of audio, and the VAD has closed the utterance by then.
+"Мэйвен, <request>" is unaffected. A bare "Мэйвен" is the case that fails.
+
+mavwaked connects at startup and holds the conn, so a nudge routed to voice
+reaches the speaker before he has said anything (V-671). It used to connect
+lazily, which made the failure silent rather than absent: after one utterance
+the session existed, `PushToMostRecent` succeeded, the dispatcher stopped
+rerouting to telegram and ntfy, and mavwaked discarded the audio.
+
+**Passwords are read from files, never taken as flag values.** `mavcaldav` uses
+`-pass-file` and `-render-pass-file`. `mavpoll` and `mavmaild` follow the same
+rule.
+
+## Web UI conventions
+
+Server-rendered pages share `cmd/mavweb/static/ui.css` (served at `/ui.css`) and
+the shell partial in `cmd/mavweb/shell.html`. A page opens with
+`{{template "shellTop" "<page-key>"}}` and closes with `{{template "shellBottom"}}`,
+and the key marks the active sidebar link.
+
+Every page is its own embedded `.html` file next to `main.go`. No page markup
+lives in Go, and the sidebar is data (`sidebarSections`, `pageIcon`) the template
+renders. No per-page `<style>` beyond true one-offs. Wrap every table in
+`<div class=scroll>` so wide data pans on a phone. Local preview and headless
+screenshot recipes are in `AGENTS.md`.

docs/design.md

@@ -308,6 +308,29 @@ The count is of CONSECUTIVE step-asides. It resets the moment he answers, in
 too. "Позвонить маме" against a question about the time is still him in the
 exchange. The retry it costs is bound enough on its own.
 
+#### And it may ride four turns in all
+
+Decided 2026-08-08 (V-663), because the bound above did not move the number it
+was written for. Twenty-six of 140 turns carried a tail before it landed and
+twenty-six carried one after.
+
+Two bounds rearm each other. An aside spends no attempt, so `MaxAttempts` never
+reaches it. A turn that reads as a failed answer zeroes `Suspends`, so
+`MaxSuspends` never reaches the asides. Alternating them, each bound is restored
+by the other's traffic. Measured on 2026-08-08: one question about a reminder's
+day rode turns 7 to 13. It ended only because turn 14 was a new request.
+
+`PendingQuestion.Rides` counts the same event as `Suspends` with the resets
+taken out. It is set once, incremented only in `noteSuspended`, carried across
+the re-park in `askRemainingGap`, and read by nothing that could lower it.
+`MaxRides` is 4, one looser than `MaxSuspends` so that the tighter statement
+about a run stays reachable.
+
+This is a bound, not a cure. It ends the measured ride one turn early. Most of
+that ride's length is attempts, spent because `classifyTurnRole` reads "спасибо"
+and "привет" as failed answers to a question about a day. That is the next
+thing to fix and it is not a bound.
+
 The re-ask is also two sentences rather than one. It used to be spliced onto the
 answer with a comma. On a real answer that buries the question in the tail of
 one run-on thought:

docs/ecosystem.md

@@ -30,6 +30,12 @@ Maven is the user-facing control center, but not the source of truth for identit
 
 Maven provides the human interface over the other systems.
 
+| Service | Owns | Maven's client | Configured at |
+|---|---|---|---|
+| **Nexus** | Canonical entity ids, names, aliases, relationships. Projects, services, devices, people, pets, places. | `nexusClient` in `cmd/mavend/ecosystem.go`, `POST /api/v1/resolve` | `nexus.url` (`http://nexus:9740`) |
+| **Praxis** | Operational attention and item lifecycle. What needs looking at, what changed, what is unresolved. | `praxisClient`, the HTTP tools API under `/api/v1/tools/` | `praxis.url` (`http://praxis:8989`) |
+| **Hexis** | The capability registry and the only path to executing anything. | vendored `github.com/kami/hexis/pkg/client` | `hexis.url` (`http://hexis:9741`) |
+
 It is responsible for:
 
 - interpreting Russian and English utterances

docs/evals/2026-08-08-clarify-head-four-head.md

@@ -0,0 +1,123 @@
+# A clarify head, and a confidence that is not a hardcode
+
+Measured 2026-08-08 on workpc, the same day and the same fixtures as
+`2026-08-08-routing-heads-two-head.md` and `2026-08-08-slot-head-three-head.md`.
+New script `gen_clarify.py`, new fixture `eval_fixture_clarify.jsonl`.
+
+## A softmax has no clarify class
+
+That sentence closed the two-head measurement. It is why the head's fixture was
+88 cases and not 96. The eight `want_clarify` cases sat outside every number
+measured, and the head had no way to produce the answer they wanted.
+
+A fourth head is the answer. Clarify is not a value of intent. It is a second
+question asked of the same pooled vector: can Maven act on this at all.
+
+## The corpus had one class
+
+Every row in `train_heads_slots.jsonl` was generated FOR an intent or a
+destination. So every row is answerable by construction. A head trained on that
+alone sees one class and learns to say yes.
+
+`gen_clarify.py` makes the other class. Five shapes, ten topics. The shapes are
+the gate's own reasons in `gateLLMDecision` plus the two the fixture carries:
+bare noun, bare verb, demonstrative, deictic time, dangling reference.
+
+**The agreement filter that worked for destination cannot work here.**
+`routeGrammar` has no clarify value. So the router always names an intent, and
+any generated line always agrees with itself. The second pass is a judge
+instead. Gemma is asked, without seeing the label, whether Maven would have to
+ask a question back.
+
+## The first judge was worthless and the second was measured
+
+The first judge said "needs clarify" on 24 of 40 plainly answerable corpus
+rows. It flagged `запиши что я пообедал` and `Покажи расписание поездов на
+вечер`. It was judging against a generic assistant, one that asks "where?"
+about lunch. Maven writes that note.
+
+Rewriting it to state what she can already do took false positives to 16 of 60.
+It also catches all eight fixture clarifies. So the judge discriminates.
+
+On the generated pile it removed 7 of 306, a 97.7% keep rate. That is not the
+judge failing. The generator is aimed at underspecified lines, so there is
+little for a filter to catch. The 27% false-positive rate is the number to
+quote, and it is label noise on the positive class.
+
+**Both passes are gemma-4-12b.** Generation and judging. So the corpus is
+gemma's opinion of what is underspecified, and the head distills that opinion.
+What keeps it honest is the fixture. Those eight cases were written by the owner
+and gemma never saw them.
+
+299 rows kept, against 3604 answerable. The positive class carries `intent:
+null`, so it costs the intent head nothing.
+
+## Result
+
+Three seeds, 24 epochs, epoch still chosen on the intent dev slice.
+
+| | two heads | three heads | four heads |
+|---|---|---|---|
+| intent mean | 93.6% | 92.8% | 91.7% |
+| destination mean | 80.8% | 82.8% | 79.8% |
+| slot span F1 mean | — | 72.4% | 68.3% |
+| clarify caught | — | — | 7.0 of 8 |
+| false clarifies | — | — | 2.3 of 88 |
+
+**The fourth head is not free the way the third was.** Intent, destination and
+slot F1 all move down. The drop is one to four points, and the seed spread is
+wide enough to contain it. Seed 2 scores intent 94.3% and destination 84.8%, both above
+every three-head seed. Read the drop as unproven rather than as absent.
+
+Accuracy is the wrong number for this head and is reported for completeness at
+95.8% to 96.9%. Eight of ninety-six cases are positive, so a head that never
+asks scores 91.7%. Recall on those eight is the number.
+
+Compare it to what ships. The cascade today misses 1 clarify and produces 2
+false ones. The head catches 7 of 8 and produces 2.3 false ones. That is
+parity, from a 118M encoder with no rules in front of it.
+
+The saved checkpoint is seed 2 at epoch 10. Intent 94.3%, destination 28/33,
+slot F1 73.6%, clarify 7 of 8 with 3 false. `heads.pt` carries four state dicts.
+
+## What it gets wrong is consistent across seeds
+
+`поужинал` is a false clarify on all three seeds. That utterance is already
+recorded as a real defect. `thinSingleToken` was narrowed on 2026-08-01 to spare
+a token carrying a Russian verb ending. One word is routinely a whole sentence
+in Russian. The head relearned the mistake the rule was narrowed to
+fix.
+
+`что дальше?` is a false clarify on two seeds. That one is a disagreement rather
+than an error. The utterance is underspecified, and V-498 decided stage 0 claims
+it for the calendar on purpose.
+
+`ну это` is missed on two seeds. `amb-003` is the shortest case in the fixture
+and the generated demonstratives are longer.
+
+## Confidence
+
+`Confidence: 1.0` was a hardcode in `llmrouter.go`, so a correct low confidence
+could not exist. Max softmax over the intent head is the replacement. It is only worth reading
+if it is lower where the head is wrong.
+
+It is. Mean 0.851 where the head is right against 0.604 where it is wrong. It
+ranks a right case above a wrong one in 83.4% of pairs.
+
+So there are two signals now and they are not the same signal. Confidence says
+the head is unsure which intent this is. The clarify head says the utterance
+does not carry enough to act on. A confident wrong route and an honest "I cannot
+tell" are different failures, and one number cannot report both.
+
+## What this does not measure
+
+The same gap as every head run. **Nothing of this runs in Go.** Four heads
+instead of three does not change that.
+
+There is no threshold. Both signals are reported as raw numbers. Turning either
+into a gate needs a decision about where to cut, and that trades false clarifies
+against wrong acts. The fixture has 8 positives, which is too few to fit a
+threshold on.
+
+The 299 generated rows have no held-out slice of their own. Clarify is scored on
+the fixture alone.

docs/evals/2026-08-08-destination-fixture.md

@@ -46,9 +46,9 @@ possible. A clarify names nothing, so it would satisfy an empty label for free.
 `Score` requires the route to land the case's intent before it credits a
 destination hit, or the floor label would score itself.
 
-## Seven cases assert the floor, and six of them cluster
+## Seven cases assert the floor, and five of them cluster
 
-The six are homelab operations. `SourceRecall`, `SourceNetwork` and
+The five are homelab operations. `SourceRecall`, `SourceNetwork` and
 `SourceAttention` overlap on every question about the box, because `mavpoll`
 writes its netdata and uptime-kuma observations into the fact store recall
 reads. "почему сервер тормозит" is answerable from all three. Naming one takes
@@ -57,6 +57,8 @@ the other two off the turn.
 That is a finding about the enum rather than a gap in the labelling. The floor
 is the right answer there and the fixture now says so out loud.
 
+All seven were written by an agent and confirmed by the owner on 08-08-2026.
+
 ## A drift the labelling found
 
 `WorldQueryGrammars` went into `buildRouter` with V-655 and never into

docs/evals/2026-08-08-parked-clarify-ride.md

@@ -0,0 +1,88 @@
+# The parked clarify ride, bounded and re-measured
+
+Date: 2026-08-08, V-663. Same 140 turns, same driver, third and fourth runs of
+the day. Before is `d6f3914`, after is that plus two changes.
+
+## What was measured before
+
+One question about a reminder's day rode turns 6 to 13. It ended only because
+turn 14 was a new request. Two of those turns are the worst replies in the
+corpus:
+
+```text
+спасибо   -> Сейчас 21:25. В какой день?
+привет    -> Сейчас 21:25. В какой день?
+```
+
+V-654 had already added `MaxSuspends` and the tail count had not moved.
+
+## Why three bounds let it happen
+
+The TTL, `MaxAttempts` and `MaxSuspends` all exist and all were rearmed.
+
+An aside spends no attempt, so `MaxAttempts` never reaches it. A turn reading as
+a failed answer zeroes `Suspends`, so `MaxSuspends` never reaches the asides.
+`спасибо` read as a failed answer. So the pleasantry both produced the reply
+above and paid for it out of the budget meant to end the ride.
+
+## The two changes
+
+`PendingQuestion.Rides` counts the same event as `Suspends` with the resets
+taken out. Set once, incremented only in `noteSuspended`, carried across the
+re-park in `askRemainingGap`. `MaxRides` is 4, one looser than `MaxSuspends` so
+that the tighter statement about a run stays reachable.
+
+A pleasantry is an aside, not a failed answer. The set is a new closed lexicon
+entry, matched as whole utterances rather than tokens. `вечер` answers `это утра
+или вечера?` and `нет` answers a confirm. Anything that could fill a slot stays
+out.
+
+## Result
+
+| | before | after |
+|---|---|---|
+| turns carrying a clarify tail | 21 | 17 |
+| turns carrying any failure string | 31 | 29 |
+| the longest ride | 8 turns | 4 turns |
+
+The turns carrying a tail, by number:
+
+```text
+before  6 7 8 9 10 11 12 13   53 54 55 56 57   99 102 103   116   136 137 138 139
+after   6 7 8 9                53 54 55 56 57   99 100 101   116   136 137 138 139
+```
+
+Turn 10 is the change. It now reads:
+
+```text
+спасибо -> Прошлую просьбу отпускаю. Пожалуйста, я всегда готова помочь тебе.
+```
+
+She lets the request go, says so, and answers the man. Turns 11 to 13 are clean.
+
+**`MaxRides` is not what fired.** The pleasantry is an aside now, so it no
+longer breaks the run. `MaxSuspends` reached three on turn 10 and ended it.
+`Rides` is the backstop for the shape where an answer really does break the run.
+No turn in this corpus reaches it.
+
+## What did not move
+
+Four rides are untouched. Turns 53 to 57 are five consecutive asides against a
+reminder missing its day. Turn 58 is a new request that drops it. Nothing
+pleasant appears in that run, so neither change applies. Turns 99 to 101 shifted
+by one, and 116 and 136 to 139 are unchanged.
+
+So the fix is worth four turns of twenty-one. What is left is asides against a
+question the owner never answers. `MaxSuspends` was written for that shape and
+does bound it, at four turns each.
+
+## Not attributable
+
+Latency moved p50 1.1s to 1.5s and p95 2.8s to 3.0s, and the 34.3s outlier in
+the earlier run is gone. Both runs had the workstation up. Read none of it as
+caused by this change.
+
+One unrelated defect appeared in the after run and is recorded here because it
+is visible in the transcript. Turn 4 answered `Я записала твою привычкуRegarding
+coffee without sugar.` That is English leaking into a Russian reply with no
+space in front of it. It is a phrasing defect and it has no task yet.

docs/evals/2026-08-08-routing-heads-in-go.md

@@ -0,0 +1,146 @@
+# The routing heads, running in Go
+
+Date: 2026-08-08. Vikunja V-664.
+Weights: `router_heads.onnx`, fp32, exported from `heads.pt` on workpc.
+Fixture: `internal/router/eval/ru_routing_v1.json`, 96 cases, 33 carrying a destination.
+Runner: `make t PKG=./internal/router/eval/ RUN=TestONNXRoutingHeads`.
+
+The four heads of V-661 ran nowhere. This is the number they score through the
+Go cascade. Same fixture and same grammars as `TestONNXBaseline`, and only the
+middle stage varies.
+
+## Headline
+
+| | classifier + ONNX | heads + classifier | gemma-4-12b cascade |
+|---|---|---|---|
+| intent | 75.0% (72/96) | **96.9% (93/96)** | 84.4% |
+| destination | 33.3% (11/33) | **75.8% (25/33)** | 72.7% |
+| false clarify | 0 | 1 | 2 |
+| missed clarify | 8 | 1 | 1 |
+| p50 | 24.5ms | 27.9ms | 329ms |
+
+A 118M encoder beats the 12B teacher it was distilled from. It wins on both
+halves of the route, at a twelfth of the latency. The workstation stays the
+better phraser and is no longer the better router.
+
+The p50 is not the heads. Most of it is the classifier's own embedder pass on
+the turns the heads decline, plus process warm-up on the first case. The heads'
+own forward pass measures 7.3ms on workpc.
+
+## Two defects were in the way, and the first was not in the heads
+
+**The tokenizer read every long word backwards.** `encodeWord` backtracks the
+Viterbi path from the end of a word and prepends each piece. That puts them back
+in reading order, and a second reverse after the loop undid it. So
+`query: вода` tokenized to `[0 12 1294 41 12489 2]` where the reference
+tokenizer gives `[0 41 1294 12 12489 2]`.
+
+It was found here and only here. The heads were trained through transformers and
+are read through the hand-written tokenizer. So a mismatch shows up as a score
+far below what Python measured on the same weights. Nothing else in the suite
+compares the two.
+
+Measured on the recall fixture, same 27 cases either way:
+
+| | reversed | fixed |
+|---|---|---|
+| recall@1 | 70.4% (19/27) | **77.8% (21/27)** |
+| recall@3 | 85.2% (23/27) | **96.3% (26/27)** |
+| answered after gate | 63.0% | 66.7% |
+| wrong note on top | 8 | 6 |
+| false recall | 0/5 | 1/5 |
+
+The classifier barely moved, 76.0% to 75.0%, and destination 36.4% to 33.3%.
+Both are one case on 96 and neither is a finding. Seeds and queries were mangled
+the same way, so cosine survived it. Recall is where it cost, because a stored
+passage and a live query are different lengths and break differently.
+
+The one new false recall is the honest cost and it is not being hidden. A
+sharper embedder scores every candidate higher, including the ones that should
+have stayed under the gate. That is the same trade `2026-08-04-recall-e5-small.md`
+recorded when e5-small replaced MiniLM.
+
+The embedder id now carries a tokenizer revision, `model_quantized@384/tok2`.
+Stored vectors were written under rev 1 and no longer sit in the same space as a
+query embedded now. The model file's name never moved, so nothing would have
+triggered `ReembedAll`. On the box the marker fired on start, and the re-embed
+rewrote 65 notes and 19 facts in 5 seconds.
+
+**The clarify head was being thrown away.** It was read only when the intent head
+cleared its own threshold. That cost 6 of the 8 ambiguous cases. `вода` reads as intent
+`act` at 0.233 and clarify at 0.983. Burying that handed the turn to the
+classifier, which routed it confidently and never asked. The clarify head answers
+a different question, which is whether there is enough here to act on at all. So
+it decides on its own and decides first.
+
+| | intent-gated | clarify decides first |
+|---|---|---|
+| intent | 90.6% | 96.9% |
+| missed clarify | 7 | 1 |
+| false clarify | 0 | 1 |
+
+## The threshold is measured, not chosen
+
+Max softmax over the intent head, on the 88 cases carrying an intent:
+
+| threshold | kept | accuracy kept | wrong kept | right dropped |
+|---|---|---|---|---|
+| 0.5 | 84 | 96.4% | 3 | 2 |
+| **0.6** | **81** | **97.5%** | **2** | **4** |
+| 0.7 | 75 | 97.3% | 2 | 10 |
+| 0.8 | 64 | 96.9% | 2 | 21 |
+| 0.9 | 46 | 100.0% | 0 | 37 |
+
+0.6 is the knee. Every value from 0.7 to 0.85 drops right answers and keeps the
+same two wrong ones. 0.9 is the only value that clears them, and it costs 37
+correct routes to do it.
+
+## Quantization was measured and rejected
+
+| build | size | intent | destination | p50 |
+|---|---|---|---|---|
+| fp32 | 470MB | 83/88 (94.3%) | 28/33 (84.8%) | 7.3ms |
+| int8 | 118MB | 79/88 (89.8%) | 26/33 (78.8%) | 4.0ms |
+| fp16 | 235MB | will not load | — | — |
+
+Python numbers, on the heads alone rather than through the cascade. int8 costs
+4.5 points of intent and 6 of destination to save 3ms. The cascade around it has
+a p50 over a second when the resident model answers. The fp16 graph is broken:
+`convert_float_to_float16` leaves a Cast node emitting float16 where the graph
+expects float, and onnxruntime refuses the session. It was not worth fixing.
+
+The exporter also had to be told to write one file. It splits weights into a
+`.onnx.data` sidecar by default. This onnxruntime resolves that path against the
+process working directory rather than the model. A split graph loads from one
+directory only.
+
+## What is still wrong
+
+**Four of the eight destination misses are calendar.** Training cannot move them.
+The possessive agenda rules claim those cases at stage 0 and name nothing on
+purpose. That caution was free while nothing downstream could name anything
+either. It has now cost four points in three separate measurements. The call is
+the owner's and it is still open.
+
+**The slot head is exported and not read.** Slots come from the stage-2
+extractor. Mapping BIO tags back to text needs character offsets the unigram
+tokenizer does not keep, which is its own piece of work.
+
+**`поужинал` is a false clarify**, which is the same defect `thinSingleToken`
+was narrowed for on 2026-08-01, arriving now from a different direction.
+
+## On the box
+
+Deployed to homesrv the same day. `voice: routing heads loaded` on start, and
+`/trace` shows `routing-heads` winning or thinning every turn. The resident model
+and the classifier are both marked never asked. Live probes:
+
+```text
+что такое TCP?                 -> kiwix     a real definition
+кто такой Линус Торвальдс?     -> kiwix     a real answer
+во сколько я лёг вчера         -> personal  не нашла у тебя такой записи
+вода                           -> thinned to clarify at 0.233 / 0.983
+```
+
+A missing or broken weights file logs and leaves the heads nil, which is
+byte-for-byte the cascade that shipped before this.

docs/evals/2026-08-08-routing-heads-two-head.md

@@ -0,0 +1,192 @@
+# Two heads on e5-small, and the first destination the router did not need a model for
+
+Measured 2026-08-08 on workpc (Radeon RX 7900 GRE, ROCm). Covers V-661, step 3 of
+`docs/plans/18-routing-heads-on-e5-small.md`. Workspace is `~/Programs/embed-training`,
+scripts `gen_query_source.py`, `label_source.py`, `build_heads_corpus.py`,
+`train_heads.py`, `score_confidence.py`.
+
+## Two heads, not four
+
+Intent is 7 classes and destination is 12 plus the `SourceUnknown` floor, sharing
+one masked mean pool over one forward pass. The plan asked for four. Two of them
+have no labels and neither is a GPU problem.
+
+**Mood is cut, not deferred.** Maven's enum is `neutral, happy, thinking, tired,
+confused` and it describes her own reply state, not the speaker's emotion.
+`psytechlab/EmpatheticIntents-ru` was the only candidate and its 32 emotion
+labels do not map onto it. There is nothing to train against.
+
+**BIO slot tags stay in the MASSIVE body.** No Maven-domain span corpus exists.
+`2026-08-08-massive-warm-start.md` records that no second Russian slot-filling
+corpus is reachable at all.
+
+The destination loss is masked with `ignore_index`. Only a query turn reaches
+`queryWalk`, so a reminder contributes nothing to it.
+
+## Where the destination labels came from
+
+V-660 taught the router prompt to name a destination. That made gemma-4-12b a
+teacher, and this distils it.
+
+Labelling the 300 query rows already in `train_v5.jsonl` gave 277 destinations.
+The shape was unusable: the floor 101, calendar 62, recall 47, and `feeds` and
+`attention` at zero. A 13-way softmax cannot learn a class with no examples.
+
+`gen_query_source.py` is the destination half of `gen_corpus.py` and runs the
+same two passes. Gemma writes questions whose answer lives in one named place.
+The daemon's own `routeSystem` prompt then routes each one back. A line survives
+only when the intent is `query` **and** the source is the destination it was
+generated for. The glosses are copied verbatim out of `route_system.txt`, so the
+generator and the labeller work from one definition.
+
+The prompt and the GBNF are dumped from `internal/router/llmrouter.go` by
+`TestDumpPrompt`, never retyped. The workspace held its own copies and V-660
+changed both.
+
+1229 kept of 2373 generated, 51.8%. Merged corpus is 3664 rows carrying 1727
+destinations:
+
+| | rows | | rows |
+|---|---|---|---|
+| the floor | 220 | world | 132 |
+| calendar | 180 | money | 126 |
+| tasks | 169 | list | 124 |
+| recall | 167 | self, feeds, attention | 120 each |
+| weather | 136 | network | 103 |
+| | | home | 50 |
+
+`home` is thin because the agreement filter rejected most of what was generated
+for it. A question about the house routes `act` more often than `query`. That is
+the filter working, and 50 is the finding rather than a shortfall.
+
+**The floor was regenerated once.** The first 120 rows carried one sentence
+shape across eight topics. That shape was "что там с X" and its two synonyms.
+Every named destination varied and only the floor collapsed. The reason is that
+the generator varies a topic, and ambiguity is not a topic.
+
+`gen_query_source.py` now rotates six floor shapes. A `почему` question, a yes
+or no question, and a question carried by intonation alone. Then a
+better-or-worse question, a status question, and an existence question. That is
+a fix to degenerate generation. It is not fitting to the fixture, whose floor
+cases are homelab operations and match none of the six.
+
+The 1229 generated rows carry `intent: null`. Every one is a query by
+construction. There are five times as many as the corpus has query rows, so
+including them would make query half the intent corpus.
+
+## Result
+
+Three seeds, two bodies, epoch chosen on the intent dev slice and never on a
+destination number.
+
+| body | floor corpus | intent mean | destination mean |
+|---|---|---|---|
+| warm-started `out/body_massive` | one shape | 93.6% | 75.8% |
+| stock `multilingual-e5-small` | one shape | 93.6% | 76.8% |
+| warm-started `out/body_massive` | six shapes | 93.6% | **80.8%** |
+
+Best single run is destination **29/33 (87.9%)**, seed 0 on the rotated floor.
+Peak 1.68GB of 17.2GB, under four minutes end to end.
+
+Against the two arms already measured on the same 33 labelled cases:
+
+| | destination |
+|---|---|
+| classifier cascade (V-659) | 12/33 (36.4%) |
+| cascade + gemma-4-12b (V-660) | 24/33 (72.7%) |
+| two heads on e5-small | 29/33 (87.9%) |
+
+A 118M encoder beats the 12B teacher it was distilled from, on the fixture. The
+per-destination split is where it happens: **recall 15/15** and **world 5/5**.
+Recall was 0/15 on the cascade and 14/15 through gemma.
+
+Read 87.9% as one seed of a mean of 80.8%, not as a headline. Three seeds score
+29, 25 and 26 of 33. One case is 3 points on a fixture this small.
+
+Intent is **not** comparable to the 73/96 and 81/96 figures those two arms
+scored. A softmax has no clarify class. The head's fixture is the 88 cases that
+carry an intent, and the 8 `want_clarify` cases are scored separately below.
+
+## The MASSIVE warm-start is worth nothing here either
+
+Step 2 measured it at +0.4 points of intent accuracy and called that inside seed
+noise. Destination was the open question, because MASSIVE has a
+`definition_word` slot that looked like a `SourceWorld` signal sitting in a head
+already trained.
+
+It is not. The two bodies score the same intent mean to one decimal. Stock is
+one point ahead on destination, which is a third of one case. Nothing here argues
+for keeping the warm-start step. Dropping it removes a dependency on a corpus
+pull that `datasets` 5.0 cannot do.
+
+## The floor moved, calendar did not
+
+Before the rotation, all seven misses at seed 0 were the floor and calendar. The
+head named a destination where the fixture says walk the chain, and it was
+confident doing it. `"почему сервер тормозит"` read `world` at 0.80.
+`"хватает ли места под новые бэкапы"` read `network` at 0.82. Those are the five
+homelab cases V-659 flagged, where `SourceRecall`, `SourceNetwork` and
+`SourceAttention` all overlap. `mavpoll` writes its observations into the fact
+store recall reads.
+
+| | one shape | six shapes |
+|---|---|---|
+| the floor | 3/7 | 6/7, 6/7, 5/7 |
+| calendar | 3/6 | 3/6, 3/6, 3/6 |
+| recall | 15/15 | 15/15 at seed 0 |
+| world | 5/5 | 5/5 |
+
+The floor was a corpus defect and it cost 3 cases. Sentence variety carried it,
+not homelab vocabulary, which the training rows still do not contain.
+
+**Calendar is 3/6 at every seed and is a different problem.** It is the shape
+V-660 named. The possessive agenda rules claim those cases at stage 0 and
+deliberately name nothing, so no destination label reaches the head. Training
+cannot move a case the head never sees. That one wants the owner's call.
+
+## Max softmax separates, weakly, and the gate stays
+
+The plan argues max softmax is a calibratable confidence where `Confidence: 1.0`
+was a hardcode. Measured on the intent head:
+
+| | n | mean confidence |
+|---|---|---|
+| correct | 80 | 0.897 |
+| wrong | 8 | 0.705 |
+| `want_clarify` | 8 | 0.685 |
+
+The softest correct answer is 0.66 and 3 of 8 clarify cases sit below it. So a
+single cut buys three clarifies at no false-clarify cost, and no more.
+
+The other five explain themselves. `"напомни"` scores 0.94 as `reminder` and
+`"сделай это"` scores 0.80 as `act`. Both are intent-certain and slot-empty, and
+confidence was never the signal there. `gateLLMDecision` already catches exactly
+that shape, an act with no allowlisted fn or a keyless fact, and it keeps doing
+so. The head replaces the hardcode. It does not replace the gate.
+
+## An incident worth recording
+
+The first generation run produced zero rows for eight destinations. `mavgpud`
+yields the card when another process wants it (V-488) and llama-server answers
+503 until the model is back. Every generate call inside that window burned one of
+the destination's batches. The run walked its own cap without a single successful
+call. The log said `503` 260 times, and the summary line said 64.1% keep rate,
+which read as success.
+
+`call()` now retries a 503 with backoff. A generator that treats an unloaded
+model as a bad generation is a silent-corpus bug, not a slow one.
+
+## What this does not measure
+
+**Nothing here runs in Go.** The heads are a `heads.pt` and an
+`out/body_heads/` on workpc. Reaching the daemon needs an ONNX export and a
+caller. The resident e5-small must not be replaced by this copy: recall depends
+on that file, and `EmbedQuery`/`EmbedPassage` are its contract.
+
+The destination fixture carries 4 of 13 classes: recall 15, floor 7, calendar 6,
+world 5. `tasks`, `money`, `list`, `home`, `network`, `feeds`, `attention` and
+`self` have no gold case. So 78.8% is silent on eight destinations that together
+hold 800 training rows.
+
+Latency was not measured. A forward pass of a 118M encoder should beat a 1.7B
+decoder on a query turn. That is arithmetic, not a number from this box.

docs/evals/2026-08-08-slot-head-three-head.md

@@ -0,0 +1,106 @@
+# A slot head, and the corpus that did not exist this morning
+
+Measured 2026-08-08 on workpc, the same day as `2026-08-08-routing-heads-two-head.md`
+and against the same fixtures. Workspace is `~/Programs/embed-training`, new
+scripts `slot_grammar.gbnf`, `slot_system.txt`, `label_slots.py`, `merge_slots.py`.
+
+## The corpus was the whole problem
+
+The two-head measurement said BIO slot tags stay in the MASSIVE body, because no
+Maven-domain span corpus exists. That was true of found corpora and false of
+made ones. Destination had the same shape at breakfast. V-660 gave gemma-4-12b a
+string to write and the label problem became a generation problem.
+
+The same trick applies to spans. `slot_grammar.gbnf` emits a list of
+`{"slot": ..., "text": ...}` and the enum closes over Maven's own five: `time`,
+`text`, `key`, `value`, `fn`. `slot_system.txt` demands each span be an exact
+substring of the utterance.
+
+**The agreement filter is free here.** Destination needed a second pass. The
+daemon's own router prompt had to route each generated line back. A span needs
+no second call. It either occurs in the utterance or it does not, and
+`label_slots.py` drops it with `find()`.
+
+1702 rows labelled from `train_v5.jsonl`, the reminder, fact, note, act and
+query intents. Chat and system carry no slot and were never asked.
+
+| slot | spans |
+|---|---|
+| text | 1175 |
+| time | 485 |
+| fn | 381 |
+| key | 72 |
+| value | 65 |
+
+37 spans dropped as not-a-substring, 2.2% of the pile. Nothing failed to parse,
+which is the grammar doing its job. 409 rows came back with no span at all.
+Those are kept and tagged all `O`. An utterance carrying no slot teaches the
+head not to invent one. An empty list is a label and not a miss.
+
+`key` and `value` are thin because they come from facts alone. That is the
+shape of the corpus, not a labeller failure.
+
+## Three heads on one forward pass
+
+Intent and destination were already two linear heads over one masked mean pool.
+Slots is a third head over the per-token states of the same pass, so the marginal
+cost is one `Linear(384, 11)`.
+
+The tag set is `O` plus `B-` and `I-` for each of the five. A softmax cannot
+emit a tag that does not exist. That is the structural guarantee the GBNF buys
+for the teacher, and the head gets it for free.
+
+Two masking rules, both `ignore_index`. A row with no `spans` key contributes
+nothing, which covers the 1900 generated destination rows and every chat and
+system turn. A padding or special-token position contributes nothing either.
+
+Scoring is exact-match span F1, not token accuracy. Most tokens are `O`, so a
+tagger that predicts nothing anywhere scores above 90% on tokens.
+
+## Result
+
+Three seeds, 24 epochs, epoch chosen on the intent dev slice alone.
+
+| | two heads | three heads |
+|---|---|---|
+| intent mean | 93.6% | 92.8% |
+| destination mean | 80.8% | 82.8% |
+| destination best | 29/33 (87.9%) | 29/33 (87.9%) |
+| slot span F1 mean | — | 72.4% |
+
+**The slot head costs nothing and adds a third decision.** Intent moves 0.8
+points down and destination 2 points up. Both sit inside the seed spread those
+two numbers already had. Read this as unchanged, not as a trade.
+
+The saved checkpoint is seed 1 at epoch 10: intent 93.2%, destination 81.8%,
+slot F1 75.8%. `heads.pt` now carries three state dicts and the `bio` list
+beside the intent and source enums.
+
+## Epoch selection is now wrong for one of the three heads
+
+Slot F1 was still climbing when the intent-selected epoch stopped it. Seed 0
+selects epoch 13 at 70.9% and reaches 76.1% at epoch 20. Seed 1 selects epoch 10
+at 75.8% and reaches 80.0% at epoch 24.
+
+So the three tasks want different epochs and the harness picks one. Two ways
+out, and neither was taken here. Select on a joint score, which needs an
+argument about weights. Or give the slot head its own dev slice and its own
+early stop, which means the heads stop being one checkpoint.
+
+Leaving it costs about 4 points of slot F1 and nothing else. The router
+contract is graded on intent. Selecting on it is the conservative choice rather
+than an oversight.
+
+## What this does not measure
+
+The same gap as the two-head run. **Nothing of this runs in Go.** Three heads
+instead of two does not change that. It does change the export. The slot head
+reads the per-token states, so an ONNX graph exposing only the pooled vector is
+not enough.
+
+There is no held-out span fixture. Span F1 is measured on the dev slice of the
+same distilled corpus, so it scores agreement with gemma rather than
+correctness. The V-572 defect utterance, `"напомни в 11:00 позвонить маме"`, is
+labelled correctly by the teacher, but one utterance is not a fixture.
+
+`key` at 72 spans and `value` at 65 are too thin to report a per-slot number.

docs/evals/2026-08-08-two-weeks-after-merge-transcript.md

@@ -0,0 +1,631 @@
+# Raw transcript, two weeks of usage
+
+## Day 1
+
+```
+YOU:   доброе утро
+MAVEN: Я не говорю так сейчас, ведь на улице уже вечер.
+       [2.8s t=218]
+
+YOU:   какой сегодня день?
+MAVEN: сегодня суббота, 8 августа 2026 года
+       [0.0s t=219]
+
+YOU:   сколько времени?
+MAVEN: сейчас 21 час 9 минут
+       [0.0s t=220]
+
+YOU:   запиши что я пью кофе без сахара
+MAVEN: Я записала твою заметку про кофе。
+       [1.5s t=221]
+
+YOU:   мой любимый язык программирования go
+MAVEN: Я записала твою заметку про Go.
+       [1.5s t=222]
+
+YOU:   напомни в 11:00 позвонить маме
+MAVEN: Сейчас 21:10. В какой день?
+       [0.4s t=223]
+
+YOU:   что у меня сегодня?
+MAVEN: на 08.08.2026 ничего нет. На какое время поставить напоминание?
+       [2.5s t=224]
+
+YOU:   что такое TCP?
+MAVEN: вот что я нашла: TCP (Transmission Control Protocol) — это транспортный протокол для передачи данных. Он обеспечивает доставку всех пакетов и позволяет собрать их в правильном порядке. Также TCP является частью сетевой модели TCP/IP, которая описывает способ передачи цифровых данных от источника к получателю. На какое время поставить напоминание?
+       [2.8s t=225]
+
+YOU:   сколько будет 17 на 23?
+MAVEN: вот что я нашла: в источниках нет результата умножения 17 на 23. На какое время поставить напоминание?
+       [1.6s t=226]
+
+YOU:   спасибо
+MAVEN: Сейчас 21:10. В какой день?
+       [1.1s t=227]
+
+```
+
+## Day 2
+
+```
+YOU:   привет
+MAVEN: Сейчас 21:10. В какой день?
+       [0.9s t=228]
+
+YOU:   что нового?
+MAVEN: не знаю — не нашла у тебя такой записи. На какое время поставить напоминание?
+       [1.3s t=229]
+
+YOU:   какая погода?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [0.7s t=230]
+
+YOU:   запиши что пароль от вайфая лежит в ящике стола
+MAVEN: Прошлую просьбу отпускаю. Я записала, что пароль от вай-фай лежит в ящике стола.
+       [1.5s t=231]
+
+YOU:   где лежит вайфай пароль?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.5s t=232]
+
+YOU:   добавь молоко в список покупок
+MAVEN: Такую команду я не знаю.
+       [1.0s t=233]
+
+YOU:   что у меня в списке покупок?
+MAVEN: в списке покупок пусто.
+       [0.0s t=234]
+
+YOU:   кто такой Линус Торвальдс?
+MAVEN: вот что я нашла: Линус Торвальдс — финско-американский программист из Хельсинки, создатель ядра Linux. Он родился 28 декабря 1969 года и окончил факультет компьютерных наук Хельсинкского университета.
+       [2.6s t=235]
+
+YOU:   какой у меня любимый язык?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.6s t=236]
+
+YOU:   сколько у меня задач?
+MAVEN: сначала: проверить интейк форму на боксе (завтра); купить хлеб; купить молоко; полить цветы; починить кран
+       [0.0s t=237]
+
+```
+
+## Day 3
+
+```
+YOU:   как дела?
+MAVEN: У меня всё хорошо, я готова помогать тебе. Чем могу быть полезна?
+       [1.2s t=238]
+
+YOU:   напомни завтра в 9 утра купить хлеб
+MAVEN: хорошо, напомню завтра в 09:00.
+       [0.3s t=239]
+
+YOU:   что у меня завтра?
+MAVEN: на 09.08.2026 ничего нет.
+       [0.0s t=240]
+
+YOU:   отмени напоминание про хлеб
+MAVEN: Такую команду я не знаю.
+       [0.5s t=241]
+
+YOU:   какие у меня напоминания?
+MAVEN: вот что я нашла: в источниках нет информации о твоих напоминаниях.
+       [0.7s t=242]
+
+YOU:   сохрани мне адрес гостиницы в Сочи
+MAVEN: Я записала твой адрес отеля в Сочи.
+       [2.2s t=243]
+
+YOU:   что я сохранил про Сочи?
+MAVEN: вот что я нашла: в источниках нет информации о том, что ты сохранил про Сочи.
+       [1.5s t=244]
+
+YOU:   почему сервер тормозит?
+MAVEN: вот что я нашла: сервер может тормозить из-за неудачной конфигурации или неоптимальных настроек, отсутствия кэширования, неоптимизированного программного обеспечения, которое нужно обновлять, а также из-за режима энергосбережения процессора, когда он не успевает повысить частоту при нагрузке.
+       [3.7s t=245]
+
+YOU:   хватает ли места под новые бэкапы?
+MAVEN: нового вот что: [$] Changes in shadow-utils password-expiration features; The Software Stewardship Lab launches; LightDM lives: version 1.33.0 released
+       [0.7s t=246]
+
+YOU:   выключи свет в спальне
+MAVEN: Такую команду я не знаю.
+       [0.4s t=247]
+
+```
+
+## Day 4
+
+```
+YOU:   доброе утро
+MAVEN: Сейчас уже вечер, поэтому я не могу сказать тебе «доброе утро».
+       [1.4s t=248]
+
+YOU:   что я пропустил?
+MAVEN: вот что я нашла: в источниках нет информации о том, что ты пропустил.
+       [1.3s t=249]
+
+YOU:   о чём мы вчера говорили?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [1.0s t=250]
+
+YOU:   запиши что я записался к врачу на четверг
+MAVEN: Я записала твою запись к врачу.
+       [1.2s t=251]
+
+YOU:   когда я иду к врачу?
+MAVEN: вот что я нашла: в источниках нет информации о том, когда именно ты идешь к врачу.
+       [2.1s t=252]
+
+YOU:   что такое ZFS?
+MAVEN: вот что я нашла: ZFS — это современная и мощная файловая система, в которой сочетаются безопасность, масштабируемость и удобство управления данными. Она была разработана корпорацией Sun Microsystems и объединяет в себе файловую систему с менеджером томов. Изначально она создавалась для того, чтобы преодолеть недостатки традиционных RAID-массивов, такие как проблема «дыры записи».
+       [2.1s t=253]
+
+YOU:   столица Франции?
+MAVEN: вот что я нашла: столица Франции — Париж.
+       [3.7s t=254]
+
+YOU:   переведи слово ремонт на английский
+MAVEN: вот что я нашла: слово «ремонт» на английский можно перевести как repair (например, ремонт дороги или крыши) или fix (общее слово для починки техники и мебели).
+       [2.7s t=255]
+
+YOU:   сколько я потратил в этом месяце?
+MAVEN: я не отслеживаю траты — не подключено.
+       [0.7s t=256]
+
+YOU:   спокойной ночи
+MAVEN: Спокойной ночи. Я уже подготовила всё к твоему отдыху.
+       [1.2s t=257]
+
+```
+
+## Day 5
+
+```
+YOU:   привет
+MAVEN: Привет! Я готова помогать тебе сегодня.
+       [1.0s t=258]
+
+YOU:   какая погода в Москве?
+MAVEN: погода не настроена.
+       [0.9s t=259]
+
+YOU:   что там с бэкапами?
+MAVEN: вот что я нашла: бэкап — это резервная копия данных, созданная в определенный момент времени и сохраненная на локальном или облачном носителе. Его делают, чтобы быстро восстановить информацию в случае необходимости.
+       [2.3s t=260]
+
+YOU:   покажи что требует внимания
+MAVEN: мне пока нечего смотреть — у Praxis нет источников.
+       [0.5s t=261]
+
+YOU:   отметь это как сделанное
+MAVEN: команды «отметь» пока нет в списке — она уже предложена, включи через клиент.
+       [0.0s t=262]
+
+YOU:   запиши что я купил новые наушники
+MAVEN: Я записала твою покупку новых наушников.
+       [1.8s t=263]
+
+YOU:   какие у меня заметки за неделю?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.0s t=264]
+
+YOU:   расскажи про Kubernetes
+MAVEN: вот что я нашла: Kubernetes — это портативная расширяемая платформа с открытым исходным кодом для управления сервисами и контейнеризованными рабочими нагрузками. Управлять им можно через интерфейс в командной строке kubectl. Я нашла эти данные в предоставленных источниках.
+       [1.8s t=265]
+
+YOU:   кто я?
+MAVEN: вот что я нашла: в источниках нет информации о том, кто ты.
+       [1.4s t=266]
+
+YOU:   пока
+MAVEN: До свидания, пока!
+       [1.4s t=267]
+
+```
+
+## Day 6
+
+```
+YOU:   доброе утро
+MAVEN: Сейчас уже вечер, так что доброе утро я тебе сказать не могу. Я готова помочь тебе с чем-нибудь другим.
+       [1.4s t=268]
+
+YOU:   сколько времени?
+MAVEN: сейчас 21 час 11 минут
+       [0.0s t=269]
+
+YOU:   напомни в 18:30 позвонить в банк
+MAVEN: Сейчас 21:11. В какой день?
+       [0.3s t=270]
+
+YOU:   поставь чайник
+MAVEN: Сейчас 21:11. В какой день?
+       [1.1s t=271]
+
+YOU:   включи музыку
+MAVEN: Сейчас 21:11. В какой день?
+       [0.9s t=272]
+
+YOU:   что у меня в календаре на пятницу?
+MAVEN: вот что я нашла: в источниках нет информации о твоих личных событиях в календаре на пятницу. В одном из текстов упоминается пятница 24 июля 2026 года, а в другом — что сегодня пятница 7 августа 2026 года. На какое время поставить напоминание?
+       [2.6s t=273]
+
+YOU:   во сколько у меня встреча?
+MAVEN: не знаю — не нашла у тебя такой записи. На какое время поставить напоминание?
+       [0.3s t=274]
+
+YOU:   запиши что дедлайн по проекту в понедельник
+MAVEN: Прошлую просьбу отпускаю. Я записала твой дедлайн на понедельник.
+       [1.7s t=275]
+
+YOU:   успею ли я до дедлайна?
+MAVEN: вот что я нашла: в источниках нет информации о твоем дедлайне, поэтому я не могу ответить, успеешь ты или нет.
+       [2.0s t=276]
+
+YOU:   спасибо
+MAVEN: Пожалуйста, я всегда готова помочь тебе.
+       [1.4s t=277]
+
+```
+
+## Day 7
+
+```
+YOU:   привет
+MAVEN: Привет! Чем я могу тебе помочь?
+       [1.3s t=278]
+
+YOU:   как ты?
+MAVEN: Я в порядке, спасибо, что спросил. Готова помогать тебе!
+       [1.4s t=279]
+
+YOU:   расскажи анекдот
+MAVEN: вот что я нашла: в источниках есть ссылки на сборники и сайты с анекдотами, но самих текстов анекдотов там нет, кроме одного: ударилась лягушка о землю головой и стала девицей прекрасной, а второй раз — месивом кровавым.
+       [3.9s t=280]
+
+YOU:   что ты умеешь?
+MAVEN: Я Мэйвен, твоя помощница. Я живу на твоём сервере и отвечаю на твои вопросы. Я запоминаю то, что ты мне говоришь, веду заметки, ставлю напоминания, а также читаю твой календарь и задачи. Я отвечаю на вопросы о мире, используя поиск или офлайн-энциклопедию. Если ты разрешишь, я могу управлять домом, локальной сетью, лентами, списком покупок и погодой, а также писать в телеграм. Я говорю по-русски и по-английски.
+       [2.9s t=281]
+
+YOU:   запиши что я начал бегать по утрам
+MAVEN: Я записала твою новую привычку
+       [1.5s t=282]
+
+YOU:   я бегаю по утрам уже неделю
+MAVEN: записала: я бегаю по утрам уже неделю
+       [0.6s t=283]
+
+YOU:   как часто я бегаю?
+MAVEN: вот что я нашла: в источниках нет информации о том, как часто бегаешь ты. Там указано, что для похудения частоту тренировок можно доводить до 5–6 раз в неделю, а новичкам ежедневные тренировки без подготовки могут привести к травмам.
+       [3.3s t=284]
+
+YOU:   сколько стоит биткоин?
+MAVEN: вот что я нашла: в источниках нет информации о стоимости биткоина.
+       [1.2s t=285]
+
+YOU:   какие новости?
+MAVEN: вот что нового: [$] Changes in shadow-utils password-expiration features; The Software Stewardship Lab launches; LightDM lives: version 1.33.0 released
+       [0.8s t=286]
+
+YOU:   хорошего дня
+MAVEN: Я не могу сказать «хорошего дня», так как сейчас уже вечер. Хорошего отдыха!
+       [1.3s t=287]
+
+```
+
+## Day 8
+
+```
+YOU:   доброе утро
+MAVEN: Сейчас вечер, так что доброе утро я не скажу. Чем я могу тебе помочь?
+       [1.2s t=288]
+
+YOU:   что у меня сегодня?
+MAVEN: на 08.08.2026 ничего нет.
+       [0.0s t=289]
+
+YOU:   напомни через час выпить воды
+MAVEN: хорошо, напомню сегодня в 22:11.
+       [0.3s t=290]
+
+YOU:   я выпил воды
+MAVEN: записала: я выпил воды
+       [0.9s t=291]
+
+YOU:   запиши что кот ест только сухой корм
+MAVEN: Я записала твою заметку про кота.
+       [1.2s t=292]
+
+YOU:   чем питается кот?
+MAVEN: вот что я нашла: в природе кошки едят мышей, птиц и ящериц. Дома их правильно кормить мясом, мясными субпродуктами, яйцами и нежирной морской рыбой. Из-за особенностей пищеварения им не подходят каши, макароны и мучные изделия.
+       [3.4s t=293]
+
+YOU:   что такое DNS?
+MAVEN: вот что я нашла: DNS (Domain Name System) — это распределенная база данных, которая сопоставляет домены с IP-адресами. Это не единая база на одном сервере, а иерархическая система серверов, работающих совместно. В ней хранятся записи, которые обеспечивают маршрутизацию трафика и управление электронной почтой.
+       [2.0s t=294]
+
+YOU:   проверь статус uptime kuma
+MAVEN: uptime kuma — Проверяю статус Uptime Kuma...
+       [1.1s t=295]
+
+YOU:   всё ли в порядке с сервером?
+MAVEN: мне пока нечего смотреть — у Praxis нет источников.
+       [0.5s t=296]
+
+YOU:   спасибо
+MAVEN: Пожалуйста, я всегда рада помочь тебе.
+       [2.7s t=297]
+
+```
+
+## Day 9
+
+```
+YOU:   привет
+MAVEN: Привет! Чем я могу тебе помочь сегодня?
+       [1.9s t=298]
+
+YOU:   какой сегодня день недели?
+MAVEN: сегодня суббота, 8 августа 2026 года
+       [0.0s t=299]
+
+YOU:   добавь хлеб и сыр в список покупок
+MAVEN: Такую команду я не знаю.
+       [0.6s t=300]
+
+YOU:   что в списке покупок?
+MAVEN: в списке покупок пусто.
+       [0.0s t=301]
+
+YOU:   удали молоко из списка
+MAVEN: Такую команду я не знаю.
+       [0.4s t=302]
+
+YOU:   напомни завтра утром вынести мусор
+MAVEN: Сейчас 21:11. Когда?
+       [0.3s t=303]
+
+YOU:   запиши что я поменял масло в машине
+MAVEN: Прошлую просьбу отпускаю. записала: запиши что я поменял масло в машине
+       [0.7s t=304]
+
+YOU:   когда я менял масло?
+MAVEN: масло в машине — поменял
+       [0.4s t=305]
+
+YOU:   сколько будет 144 делить на 12?
+MAVEN: вот что я нашла: в источниках указано решение деления 144 на 12 в столбик, но итоговый результат деления не прописан.
+       [2.3s t=306]
+
+YOU:   пока
+MAVEN: До свидания! Если понадобится что-то записать или проверить, я всегда на связи.
+       [1.2s t=307]
+
+```
+
+## Day 10
+
+```
+YOU:   доброе утро
+MAVEN: Я не могу сказать «доброе утро», так как сейчас уже вечер. Я записала твои предыдущие вопросы.
+       [1.8s t=308]
+
+YOU:   что нового за ночь?
+MAVEN: вот что я нашла: в ночь на 8 августа вооруженные силы РФ нанесли второй за последнее время массированный удар, при этом киевская ПВО пропустила все российские ракеты.
+       [3.3s t=309]
+
+YOU:   почему интернет медленный?
+MAVEN: вот что я нашла: интернет может работать медленно из-за плохого сигнала из-за удаленности вышки, исчерпанного лимита трафика по тарифу, нахождения вне зоны покрытия 4G/LTE, устаревшей или неисправной SIM-карты, а также из-за сбоя в настройках роутера. Чтобы исправить ситуацию, можно попробовать перезагрузить роутер или перенастроить устройство.
+       [3.0s t=310]
+
+YOU:   какая скорость у меня сейчас?
+MAVEN: для какого города?
+       [0.6s t=311]
+
+YOU:   запиши что новый роутер стоит 8000 рублей
+MAVEN: Я записала стоимость нового роутера.
+       [1.2s t=312]
+
+YOU:   сколько стоил роутер?
+MAVEN: вот что я нашла: цена зависит от модели. Роутер от Ростелеком стоит 5 900 рублей, на OZON есть варианты от 4 052 до 3 251 рублей, а в Билайне цены начинаются от 2 149 рублей.
+       [2.6s t=313]
+
+YOU:   что такое NAT?
+MAVEN: вот что я нашла: NAT (Network Address Translation) — это технология в сетях TCP/IP, которая позволяет изменять IP-адрес в заголовке пакета при его прохождении через устройство. Она перенаправляет трафик между локальной и глобальной сетями, позволяет объединять устройства в локальной сети под одним публичным IP-адресом, а также скрывает сеть и экономит IP.
+       [2.1s t=314]
+
+YOU:   напомни в субботу позвонить бабушке
+MAVEN: Сейчас 21:12. Когда?
+       [0.3s t=315]
+
+YOU:   покажи мои напоминания
+MAVEN: вот что я нашла: в источниках нет информации о твоих конкретных напоминаниях, там только описано, как их создавать и где искать в приложениях. На какое время поставить напоминание?
+       [3.0s t=316]
+
+YOU:   спасибо
+MAVEN: Сейчас 21:12. Во сколько напомнить?
+       [1.0s t=317]
+
+```
+
+## Day 11
+
+```
+YOU:   привет
+MAVEN: Сейчас 21:12. Скажи время — например, «в семь вечера» или «через час».
+       [0.8s t=318]
+
+YOU:   как погода на выходных?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [0.7s t=319]
+
+YOU:   что у меня на этой неделе?
+MAVEN: вот что я нашла: в источниках нет информации о том, что у тебя на этой неделе. На какое время поставить напоминание?
+       [1.0s t=320]
+
+YOU:   запиши что я хочу прочитать книгу про Go
+MAVEN: Прошлую просьбу отпускаю. Я записала твою заметку о книге по Go.
+       [1.6s t=321]
+
+YOU:   что я хотел прочитать?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.9s t=322]
+
+YOU:   объясни что такое горутина
+MAVEN: вот что я нашла: я проверила источники и узнала, что горутины — это легковесные потоки выполнения или функции, которые позволяют параллельно выполнять код внутри программы. Они являются легкими нитями, которые создаются и планируются для независимого выполнения планировщиком Go.
+       [2.5s t=323]
+
+YOU:   кто написал Войну и мир?
+MAVEN: вот что я нашла: роман «Война и мир» написал Лев Николаевич Толстой.
+       [2.2s t=324]
+
+YOU:   включи свет на кухне
+MAVEN: Такую команду я не знаю.
+       [0.5s t=325]
+
+YOU:   закрой шторы в комнате
+MAVEN: Такую команду я не знаю.
+       [0.4s t=326]
+
+YOU:   спокойной ночи
+MAVEN: Спокойной ночи! Я уже подготовила всё к твоему отдыху.
+       [1.1s t=327]
+
+```
+
+## Day 12
+
+```
+YOU:   доброе утро
+MAVEN: Сейчас уже вечер, поэтому я не могу сказать тебе «доброе утро».
+       [1.1s t=328]
+
+YOU:   сколько сейчас времени?
+MAVEN: сейчас 21 час 12 минут
+       [0.0s t=329]
+
+YOU:   я не то имел в виду
+MAVEN: Я поняла тебя, исправь, пожалуйста, свою мысль. Что именно ты хотел сказать?
+       [1.8s t=330]
+
+YOU:   о чём мы говорили?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.6s t=331]
+
+YOU:   напомни
+MAVEN: Сейчас 21:12. Когда?
+       [0.5s t=332]
+
+YOU:   сделай это
+MAVEN: пока не умею отвечать на этот вопрос. На какое время поставить напоминание?
+       [0.5s t=333]
+
+YOU:   запиши что я перешёл на новый тариф
+MAVEN: Прошлую просьбу отпускаю. Я записала твою новую информацию о тарифе.
+       [1.5s t=334]
+
+YOU:   какой у меня тариф?
+MAVEN: вот что я нашла: в источниках нет информации о твоем тарифе.
+       [1.2s t=335]
+
+YOU:   сколько я плачу за интернет?
+MAVEN: вот что я нашла: в источниках нет информации о том, сколько именно ты платишь за интернет. Я проверила документы, но точной суммы там нет.
+       [2.6s t=336]
+
+YOU:   спасибо
+MAVEN: Я поняла тебя. Что именно ты хочешь, чтобы я сделала сейчас?
+       [1.5s t=337]
+
+```
+
+## Day 13
+
+```
+YOU:   привет
+MAVEN: Привет! Я готова помогать тебе. Чем займемся сегодня?
+       [1.3s t=338]
+
+YOU:   что там с задачами?
+MAVEN: вот что я нашла: я отметила задачу про бэкапы как сделанную.
+       [1.5s t=339]
+
+YOU:   закрывай
+MAVEN: какой пункт отметить сделанным?
+       [0.0s t=340]
+
+YOU:   отметь задачу про бэкапы как сделанную
+MAVEN: Такую команду я не знаю.
+       [0.6s t=341]
+
+YOU:   что осталось нерешённым?
+MAVEN: вот что я нашла: в твоей жизни всё, что остается нерешенным или незавершенным, будет продолжать существовать и требовать внимания. Также в источниках упоминается пример с скобками, который может остаться нерешенным, и вопрос о том, что остается нерешенным в краснодарских перевозках.
+       [4.2s t=342]
+
+YOU:   запиши что я договорился о встрече в среду
+MAVEN: Я записала твою договоренность о встрече.
+       [2.3s t=343]
+
+YOU:   когда у меня встреча?
+MAVEN: мне пока нечего смотреть — у Praxis нет источников.
+       [0.0s t=344]
+
+YOU:   какая температура на улице?
+MAVEN: для какого города?
+       [0.9s t=345]
+
+YOU:   что такое RAID 5?
+MAVEN: вот что я нашла: RAID 5 — это массив из независимых дисков, где данные и контрольные суммы (четность) распределяются по кругу на все диски. Этот уровень обеспечивает отказоустойчивость и повышенную производительность, при этом для его работы требуется не менее трех дисков.
+       [2.3s t=346]
+
+YOU:   пока
+MAVEN: До встречи!
+       [1.4s t=347]
+
+```
+
+## Day 14
+
+```
+YOU:   доброе утро
+MAVEN: Я не могу сказать «доброе утро», потому что сейчас вечер. Я уже записала твой вопрос и готова помочь с чем-то другим.
+       [1.7s t=348]
+
+YOU:   подведи итоги недели
+MAVEN: week summary — подведи итоги недели
+       [0.6s t=349]
+
+YOU:   что я делал за последние две недели?
+MAVEN: вот что я нашла: в источниках нет информации о том, что ты делал за последние две недели.
+       [1.3s t=350]
+
+YOU:   какие заметки я сохранил?
+MAVEN: вот что я нашла: в источниках нет информации о том, какие именно заметки ты сохранил.
+       [1.3s t=351]
+
+YOU:   о чём я чаще всего спрашиваю?
+MAVEN: обычно ты пьёшь воду около 00:58, отмечаешь «name» около 23:56 и отмечаешь «week summary» около 23:59 — по записям за последние 7 дней.
+       [0.9s t=352]
+
+YOU:   напомни в понедельник в 10 проверить бэкапы
+MAVEN: Сейчас 21:12. Это утра или вечера?
+       [0.3s t=353]
+
+YOU:   что у меня в понедельник?
+MAVEN: не знаю — не нашла у тебя такой записи. На какое время поставить напоминание?
+       [0.3s t=354]
+
+YOU:   ты меня понимаешь?
+MAVEN: Да, я тебя прекрасно понимаю. Чем я могу тебе помочь? На какое время поставить напоминание?
+       [1.3s t=355]
+
+YOU:   спасибо тебе
+MAVEN: Сейчас 21:12. Это утра или вечера?
+       [1.0s t=356]
+
+YOU:   спокойной ночи
+MAVEN: хорошо, напомню послезавтра в 10:00.
+       [1.2s t=357]
+
+```

docs/evals/2026-08-08-two-weeks-after-merge.md

@@ -0,0 +1,132 @@
+# The same fortnight, after V-655 merged
+
+Date: 2026-08-08, a few hours after `2026-08-08-two-weeks.md`.
+Build: `f8fa0d1` on master, the five compose services rebuilt and recreated.
+Same driver, same 140 turns, same reach. This is the diff that baseline was for.
+
+Master now carries V-655, V-659, V-660 and V-661. The change under test is V-655. A query source that decides by seed similarity
+is marked `guesses: true`. It is dropped when the cascade names a different
+destination.
+
+## Two things confound the comparison and one of them matters
+
+**The workstation was up for the re-run.** `llama-server` on 192.168.1.105
+answered a health probe with 200. So routing completed through `llm.Pair`
+against gemma-4-12b, which is the arm that names a destination. Its state
+during the baseline was not recorded. So a difference here may be the merge, or
+may be the better router, and this run cannot separate them.
+
+**The store carried over**, as the baseline said it would. Facts written by the
+first run were present from turn 1 of the second.
+
+## Numbers
+
+| | baseline `beb093a` | after `f8fa0d1` |
+|---|---|---|
+| turns | 140 | 140 |
+| p50 | 1.5s | 1.2s |
+| p95 | 7.1s | 3.0s |
+| max | 33.7s | 4.2s |
+| transport errors | 0 | 0 |
+| turns carrying a failure string | 41 | 38 |
+
+| string in the reply | before | after |
+|---|---|---|
+| `на какое время поставить напоминание` | 13 | 13 |
+| `не нашла у тебя такой записи` | 8 | 9 |
+| `Такую команду я не знаю` | 8 | 8 |
+| `для какого города` | 6 | 4 |
+| `В какой день` | 6 | 6 |
+| `пока не умею` | 5 | 1 |
+| `Когда?` | 3 | 3 |
+
+Read the latency as unattributed. The workstation confound covers all of it.
+
+## Defect 2 is the one this was for: four of six fixed
+
+| utterance | before | after |
+|---|---|---|
+| `что такое TCP?` | `для какого города?` | a real definition |
+| `сколько будет 17 на 23?` | `для какого города?` | search, which has no answer |
+| `какой у меня любимый язык?` | kernel headlines | `не нашла у тебя такой записи` |
+| `что я сохранил про Сочи?` | `Хорошо, сохраню.` | answered as a question |
+| `какая скорость у меня сейчас?` | `для какого города?` | `для какого города?` |
+| `хватает ли места под новые бэкапы?` | kernel headlines | kernel headlines |
+
+`что такое TCP?` is the clean win. `WorldQueryGrammars` names `world` at stage
+0, weather is dropped, and search answers.
+
+`сколько будет 17 на 23?` moved source and not outcome. Weather no longer claims it. Search
+cannot do arithmetic, so the reply says the sources have no product of 17 and
+23. That is an honest gap where it used to be a wrong
+question. Arithmetic has no destination in the enum.
+
+`что я сохранил про Сочи?` was defect 3 and it is gone. The utterance is no
+longer read as a capture.
+
+**The two that did not move are both homelab questions.** They are exactly the
+cluster the destination fixture flagged. `SourceRecall`, `SourceNetwork` and
+`SourceAttention` overlap on every question about the box. Five of the seven
+floor cases in that fixture are homelab operations for the same reason. So this
+is the enum, not the walk.
+
+## Defect 1 did not move at all
+
+Twenty-six turns still carry a parked clarify tail, the same count as the
+baseline. `спасибо тебе` answers `Сейчас 21:12. Это утра или вечера?` and
+`спокойной ночи` answers `хорошо, напомню послезавтра в 10:00.`
+
+V-655 was never going to touch this. A parked clarify is dialogue state and not
+a query source. It remains the single worst thing about talking to her. The week test, the
+fortnight test and this re-run all report it unchanged.
+
+## A gap in the harness, fixed and re-run the same day
+
+`ipc.ChatReply.Source` came back empty on all 140 turns, in both runs. The
+driver read the redirect parameter `src` and `cmd/mavweb/chat.go` writes `s`.
+So every finding above is read off the reply text instead of off the badge.
+
+Fixed in V-662 and the 140 turns were driven a third time. Sixty-eight of them
+name a source. The rest are not query turns and never reach `queryWalk`.
+
+| source | turns |
+|---|---|
+| search | 27 |
+| memory | 13 |
+| personal | 9 |
+| weather | 5 |
+| calendar | 3 |
+| attention | 3 |
+| list | 2 |
+| feeds | 2 |
+| tasks, money, self, habits | 1 each |
+
+## What the badge shows that the wording did not
+
+The two unfixed homelab turns are now direct evidence.
+
+```text
+какая скорость у меня сейчас?        -> weather
+хватает ли места под новые бэкапы?   -> feeds
+```
+
+Both are guessing sources claiming a turn about the box, exactly as the
+destination fixture predicted.
+
+The badge also names a defect the wording hid. **Agenda questions are being
+claimed by the personal boundary and by Praxis, not by the calendar.**
+
+```text
+во сколько у меня встреча?   -> personal    не нашла у тебя такой записи
+когда у меня встреча?        -> attention   у Praxis нет источников
+что у меня в понедельник?    -> personal    не нашла у тебя такой записи
+```
+
+Calendar claimed 3 turns of the 6 that asked about the calendar. That is the
+same 3/6 the destination fixture scores and the same 3/6 every seed of the
+routing head scores. Three measurements agree. The cause is the one V-660 named. The possessive
+agenda rules claim these at stage 0 and name no destination, so the walk
+reaches `personal` and `attention` first.
+
+This is the third independent confirmation that the possessive agenda rules
+should name the calendar. That call is still the owner's.

docs/evals/2026-08-08-two-weeks-transcript.md

@@ -0,0 +1,636 @@
+# Raw transcript, two weeks of usage
+
+Companion to `2026-08-08-two-weeks.md`. 140 turns through `POST /api/chat`,
+driven by `scripts/usage-run.py` over `scripts/testdata/usage-turns.txt`.
+Days are simulated. The wall clock was 20:18-20:27 throughout, which is why
+every clock answer says evening.
+
+## Day 1
+
+```
+YOU:   доброе утро
+MAVEN: Сейчас уже вечер, поэтому я не могу сказать тебе «доброе утро».
+       [1.5s t=78]
+
+YOU:   какой сегодня день?
+MAVEN: сегодня суббота, 8 августа 2026 года
+       [0.0s t=79]
+
+YOU:   сколько времени?
+MAVEN: сейчас 20 часов 18 минут
+       [0.0s t=80]
+
+YOU:   запиши что я пью кофе без сахара
+MAVEN: Я записала твою заметку про кофе。
+       [2.0s t=81]
+
+YOU:   мой любимый язык программирования go
+MAVEN: Я записала твою заметку про Go.
+       [1.4s t=82]
+
+YOU:   напомни в 11:00 позвонить маме
+MAVEN: Сейчас 20:18. В какой день?
+       [0.4s t=83]
+
+YOU:   что у меня сегодня?
+MAVEN: на 08.08.2026 ничего нет, на какое время поставить напоминание?
+       [0.3s t=84]
+
+YOU:   что такое TCP?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [1.0s t=85]
+
+YOU:   сколько будет 17 на 23?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [0.9s t=86]
+
+YOU:   спасибо
+MAVEN: Сейчас 20:18. В какой день?
+       [0.9s t=87]
+
+```
+
+## Day 2
+
+```
+YOU:   привет
+MAVEN: Сейчас 20:18. В какой день?
+       [0.9s t=88]
+
+YOU:   что нового?
+MAVEN: не знаю — не нашла у тебя такой записи, на какое время поставить напоминание?
+       [1.2s t=89]
+
+YOU:   какая погода?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [0.6s t=90]
+
+YOU:   запиши что пароль от вайфая лежит в ящике стола
+MAVEN: Прошлую просьбу отпускаю. Хорошо, сохранил.
+       [33.7s t=91]
+
+YOU:   где лежит вайфай пароль?
+MAVEN: вот что я нашла: пароль от вайфая лежит в ящике стола
+       [7.0s t=92]
+
+YOU:   добавь молоко в список покупок
+MAVEN: Такую команду я не знаю.
+       [1.4s t=93]
+
+YOU:   что у меня в списке покупок?
+MAVEN: в списке покупок пусто.
+       [0.0s t=94]
+
+YOU:   кто такой Линус Торвальдс?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [1.6s t=95]
+
+YOU:   какой у меня любимый язык?
+MAVEN: вот что нового: [$] Changes in shadow-utils password-expiration features; The Software Stewardship Lab launches; LightDM lives: version 1.33.0 released
+       [1.3s t=96]
+
+YOU:   сколько у меня задач?
+MAVEN: сначала: проверить интейк форму на боксе (завтра); купить хлеб; купить молоко; полить цветы; починить кран
+       [0.0s t=97]
+
+```
+
+## Day 3
+
+```
+YOU:   как дела?
+MAVEN: рада, как дела?
+       [4.1s t=98]
+
+YOU:   напомни завтра в 9 утра купить хлеб
+MAVEN: хорошо, напомню завтра в 09:00.
+       [0.3s t=99]
+
+YOU:   что у меня завтра?
+MAVEN: на 09.08.2026 ничего нет.
+       [0.0s t=100]
+
+YOU:   отмени напоминание про хлеб
+MAVEN: Такую команду я не знаю.
+       [1.6s t=101]
+
+YOU:   какие у меня напоминания?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [0.0s t=102]
+
+YOU:   сохрани мне адрес гостиницы в Сочи
+MAVEN: Хорошо, сохраню.
+       [5.6s t=103]
+
+YOU:   что я сохранил про Сочи?
+MAVEN: Хорошо, сохраню.
+       [2.7s t=104]
+
+YOU:   почему сервер тормозит?
+MAVEN: есть такое: Почему тормозит сервер сайта: причины медленной работы и способы решения — Почему сервер работает медленно: основные причины торможения, диагностика проблем, способы оптимизации. Практические советы по ускорению работы сервера и сайта.
+       [3.1s t=105]
+
+YOU:   хватает ли места под новые бэкапы?
+MAVEN: нового вот что: [$] Changes in shadow-utils password-expiration features; The Software Stewardship Lab launches; LightDM lives: version 1.33.0 released
+       [1.5s t=106]
+
+YOU:   выключи свет в спальне
+MAVEN: Такую команду я не знаю.
+       [9.0s t=107]
+
+```
+
+## Day 4
+
+```
+YOU:   доброе утро
+MAVEN: рада
+       [5.0s t=108]
+
+YOU:   что я пропустил?
+MAVEN: я не могу ответить на вопрос о том, что ты пропустил. мне нужно больше информации.
+       [5.7s t=109]
+
+YOU:   о чём мы вчера говорили?
+MAVEN: пока не умею отвечать на этот вопрос.
+       [1.1s t=110]
+
+YOU:   запиши что я записался к врачу на четверг
+MAVEN: Хорошо, сохранил.
+       [6.0s t=111]
+
+YOU:   когда я иду к врачу?
+MAVEN: пока не умею отвечать на этот вопрос.
+       [1.1s t=112]
+
+YOU:   что такое ZFS?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [1.0s t=113]
+
+YOU:   столица Франции?
+MAVEN: пока не умею отвечать на этот вопрос.
+       [1.2s t=114]
+
+YOU:   переведи слово ремонт на английский
+MAVEN: записала: переведи слово ремонт на английский
+       [1.1s t=115]
+
+YOU:   сколько я потратил в этом месяце?
+MAVEN: я не отслеживаю траты — не подключено.
+       [1.4s t=116]
+
+YOU:   спокойной ночи
+MAVEN: спокойной ночи
+       [4.3s t=117]
+
+```
+
+## Day 5
+
+```
+YOU:   привет
+MAVEN: Привет!
+       [2.6s t=118]
+
+YOU:   какая погода в Москве?
+MAVEN: погода не настроена.
+       [1.2s t=119]
+
+YOU:   что там с бэкапами?
+MAVEN: нашла вот это: Бэкап - что такое резервное копирование и как его сделать — Бэкап (backup) — это резервная копия важных данных, сохраненная в специальный файл. Ею пользуются как подстраховкой: если что-то сломается, с помощью бэкапа можно будет все вернуть. Простейший пример бэкапа — сохранение файла в облаке или о…
+       [4.0s t=120]
+
+YOU:   покажи что требует внимания
+MAVEN: я заметила, что у тебя есть запрос о том, что требуется внимание. давай разберёмся вместе.
+       [5.2s t=121]
+
+YOU:   отметь это как сделанное
+MAVEN: команды «отметь» нет в списке. Предложила её добавить — включи через клиент.
+       [0.0s t=122]
+
+YOU:   запиши что я купил новые наушники
+MAVEN: Заметка сохранена.
+       [3.7s t=123]
+
+YOU:   какие у меня заметки за неделю?
+MAVEN: week summary — подведи итоги недели
+       [0.0s t=124]
+
+YOU:   расскажи про Kubernetes
+MAVEN: вот что я нашла: Что такое Kubernetes — Сервисы, поддержка и инструменты Kubernetes широко доступны. Название Kubernetes происходит от греческого, что означает рулевой или штурман. Google открыл исходный код Kubernetes в 2014 году.
+       [1.0s t=125]
+
+YOU:   кто я?
+MAVEN: name — Ками
+       [1.0s t=126]
+
+YOU:   пока
+MAVEN: Пока, спасибо за разговор.
+       [4.1s t=127]
+
+```
+
+## Day 6
+
+```
+YOU:   доброе утро
+MAVEN: Добро пожаловать! Я Maven — твой домашний ассистент. Как я могу помочь сегодня?
+       [10.1s t=128]
+
+YOU:   сколько времени?
+MAVEN: сейчас 20 часов 21 минута
+       [0.0s t=129]
+
+YOU:   напомни в 18:30 позвонить в банк
+MAVEN: Сейчас 20:21. В какой день?
+       [0.3s t=130]
+
+YOU:   поставь чайник
+MAVEN: Сейчас 20:21. В какой день?
+       [5.3s t=131]
+
+YOU:   включи музыку
+MAVEN: Сейчас 20:21. В какой день?
+       [1.7s t=132]
+
+YOU:   что у меня в календаре на пятницу?
+MAVEN: вот что я нашла: ты записался к врачу на четверг, лежит вайфай пароль в ящике стола, сохранил про Сочи, пью кофе без сахара, на какое время поставить напоминание?
+       [5.5s t=133]
+
+YOU:   во сколько у меня встреча?
+MAVEN: не знаю — не нашла у тебя такой записи, на какое время поставить напоминание?
+       [0.3s t=134]
+
+YOU:   запиши что дедлайн по проекту в понедельник
+MAVEN: Прошлую просьбу отпускаю. Хорошо, записала.
+       [6.3s t=135]
+
+YOU:   успею ли я до дедлайна?
+MAVEN: вот что я нашла: дедлайн по проекту указан как понедельник.
+       [5.3s t=136]
+
+YOU:   спасибо
+MAVEN: спасиба
+       [3.8s t=137]
+
+```
+
+## Day 7
+
+```
+YOU:   привет
+MAVEN: Привет!
+       [2.6s t=138]
+
+YOU:   как ты?
+MAVEN: рада
+       [4.2s t=139]
+
+YOU:   расскажи анекдот
+MAVEN: Я могу рассказать интересный анекдот для тебя. Вот - в одном из фильмов был такой сценарий: главному герою дали уйти от команды и он остался один, а потом появился его друг, который сказал ему: 'Тебе не нужно было остаться'. Но это просто игра слов.
+       [7.1s t=140]
+
+YOU:   что ты умеешь?
+MAVEN: Я могу помочь с задачами, напоминать о важных датах и событиях. Также я помогу записывать заметки или отвечать по ним.
+       [4.1s t=141]
+
+YOU:   запиши что я начал бегать по утрам
+MAVEN: Заметка сохранена.
+       [9.2s t=142]
+
+YOU:   я бегаю по утрам уже неделю
+MAVEN: записала: я бегаю по утрам уже неделю
+       [1.9s t=143]
+
+YOU:   как часто я бегаю?
+MAVEN: вот что я нашла: ты бегал по утрам
+       [5.3s t=144]
+
+YOU:   сколько стоит биткоин?
+MAVEN: есть такое: Цена Bitcoin сегодня | Данные о цене, рынке и графики BTC в ... — В настоящее время 1 Bitcoin стоит 65 тыс $, что на 0,09% выше за последние 24 ч. Отслеживайте цену BTC, рыночную капитализацию, исторический график и рыночную аналитику в реальном времени.
+       [2.1s t=145]
+
+YOU:   какие новости?
+MAVEN: вот что нового: [$] Changes in shadow-utils password-expiration features; The Software Stewardship Lab launches; LightDM lives: version 1.33.0 released
+       [0.8s t=146]
+
+YOU:   хорошего дня
+MAVEN: Хорошого дня!
+       [4.1s t=147]
+
+```
+
+## Day 8
+
+```
+YOU:   доброе утро
+MAVEN: Добро пожаловать! Как могу помочь сегодня?
+       [2.9s t=148]
+
+YOU:   что у меня сегодня?
+MAVEN: на 08.08.2026 ничего нет.
+       [0.0s t=149]
+
+YOU:   напомни через час выпить воды
+MAVEN: хорошо, напомню сегодня в 21:22.
+       [0.3s t=150]
+
+YOU:   я выпил воды
+MAVEN: записала: я выпил воды
+       [1.2s t=151]
+
+YOU:   запиши что кот ест только сухой корм
+MAVEN: Хорошо, записала.
+       [5.6s t=152]
+
+YOU:   чем питается кот?
+MAVEN: вот что я нашла: кот питается только сухим кормом.
+       [4.5s t=153]
+
+YOU:   что такое DNS?
+MAVEN: нашла вот это: Dns — Википедия — В случае рекурсивного запроса DNS-сервер опрашивает серверы (в порядке убывания уровня зон в имени), пока не найдёт ответ или не обнаружит, что домена не существует (на практике поиск ...
+       [2.0s t=154]
+
+YOU:   проверь статус uptime kuma
+MAVEN: записала: проверь статус uptime kuma
+       [1.6s t=155]
+
+YOU:   всё ли в порядке с сервером?
+MAVEN: пока не умею отвечать на этот вопрос.
+       [1.3s t=156]
+
+YOU:   спасибо
+MAVEN: спасибо за пожелание
+       [4.0s t=157]
+
+```
+
+## Day 9
+
+```
+YOU:   привет
+MAVEN: рада,
+       [2.1s t=158]
+
+YOU:   какой сегодня день недели?
+MAVEN: сегодня суббота, 8 августа 2026 года
+       [0.0s t=159]
+
+YOU:   добавь хлеб и сыр в список покупок
+MAVEN: Такую команду я не знаю.
+       [1.3s t=160]
+
+YOU:   что в списке покупок?
+MAVEN: в списке покупок пусто.
+       [0.0s t=161]
+
+YOU:   удали молоко из списка
+MAVEN: Такую команду я не знаю.
+       [1.2s t=162]
+
+YOU:   напомни завтра утром вынести мусор
+MAVEN: Сейчас 20:22. Когда?
+       [0.3s t=163]
+
+YOU:   запиши что я поменял масло в машине
+MAVEN: Прошлую просьбу отпускаю. Такую команду я не знаю.
+       [1.3s t=164]
+
+YOU:   когда я менял масло?
+MAVEN: вот что я нашла: по данным из last update 2019 год, заправки Сочи не могли предоставить масло для замены, поэтому это было невозможно.
+       [7.1s t=165]
+
+YOU:   сколько будет 144 делить на 12?
+MAVEN: есть такое: Разделить в столбик 144 на 12 - Calc-Best.ru — 144 : 12 = 12 Где 144 - делимое, 12 - делитель, 12 - частное ... Разделить 770 на 7 столбиком Разделить 441 на 9 столбиком Разделить 891 на 9 столбиком Разделить 60 на 5 столбиком Разделить 60 на 6 столбиком
+       [9.6s t=166]
+
+YOU:   пока
+MAVEN: я записала
+       [4.8s t=167]
+
+```
+
+## Day 10
+
+```
+YOU:   доброе утро
+MAVEN: рада
+       [4.1s t=168]
+
+YOU:   что нового за ночь?
+MAVEN: нашла вот это: Главные новости сегодня в России, а также самые последние новости со ... — Полное бессилие: киевская ПВО вновь пропустила все российские ракеты. Вооруженные силы РФ в ночь на 8 августа нанесли второй за последнее время массированный ...
+       [2.1s t=169]
+
+YOU:   почему интернет медленный?
+MAVEN: вот что я нашла: Почему интернет стал медленнее: причины и способы решения - АО «Газпром ... — На скорость интернета сильно влияют торрент-клиенты, VPN, потоковое видео, фоновая загрузка системных обновлений. Посмотреть, куда уходит трафик, можно в настройках ПК или смартфона.
+       [2.4s t=170]
+
+YOU:   какая скорость у меня сейчас?
+MAVEN: для какого города?
+       [1.2s t=171]
+
+YOU:   запиши что новый роутер стоит 8000 рублей
+MAVEN: Я записала стоимость нового роутера.
+       [2.5s t=172]
+
+YOU:   сколько стоил роутер?
+MAVEN: вот что я нашла: новый роутер стоит 8000 рублей.
+       [2.0s t=173]
+
+YOU:   что такое NAT?
+MAVEN: NAT (Network Address Translation) — это технология, которая позволяет устройствам в локальной сети выходить в интернет, используя один общий внешний IP-адрес. Я отвечаю на этот вопрос, используя поиск в сети.
+       [1.8s t=174]
+
+YOU:   напомни в субботу позвонить бабушке
+MAVEN: Сейчас 20:23. Когда?
+       [0.3s t=175]
+
+YOU:   покажи мои напоминания
+MAVEN: вот что я нашла: в твоих источниках нет напоминаний, только запись о том, что ты купил новые наушники, на какое время поставить напоминание?
+       [1.7s t=176]
+
+YOU:   спасибо
+MAVEN: Сейчас 20:23. Во сколько напомнить?
+       [1.0s t=177]
+
+```
+
+## Day 11
+
+```
+YOU:   привет
+MAVEN: Сейчас 20:23. Скажи время — например, «в семь вечера» или «через час».
+       [0.9s t=178]
+
+YOU:   как погода на выходных?
+MAVEN: для какого города? На какое время поставить напоминание?
+       [0.6s t=179]
+
+YOU:   что у меня на этой неделе?
+MAVEN: не знаю — не нашла у тебя такой записи, на какое время поставить напоминание?
+       [0.3s t=180]
+
+YOU:   запиши что я хочу прочитать книгу про Go
+MAVEN: Прошлую просьбу отпускаю. Я записала твою заметку о книге по Go.
+       [1.4s t=181]
+
+YOU:   что я хотел прочитать?
+MAVEN: вот что я нашла: ты хотел прочитать книгу про Go.
+       [1.1s t=182]
+
+YOU:   объясни что такое горутина
+MAVEN: вот что я нашла: я узнала, что горутины — это легковесные потоки выполнения или функции в Go, которые позволяют параллельно выполнять код. Они управляются планировщиком Go и работают как независимые нити выполнения.
+       [1.9s t=183]
+
+YOU:   кто написал Войну и мир?
+MAVEN: вот что я нашла: роман «Война и мир» написал Лев Николаевич Толстой.
+       [2.4s t=184]
+
+YOU:   включи свет на кухне
+MAVEN: Такую команду я не знаю.
+       [0.5s t=185]
+
+YOU:   закрой шторы в комнате
+MAVEN: Такую команду я не знаю.
+       [0.3s t=186]
+
+YOU:   спокойной ночи
+MAVEN: Спокойной ночи. Отдыхай, я на связи, если что-то понадобится.
+       [1.2s t=187]
+
+```
+
+## Day 12
+
+```
+YOU:   доброе утро
+MAVEN: Я не могу сказать тебе «доброе утро», потому что сейчас уже вечер.
+       [1.0s t=188]
+
+YOU:   сколько сейчас времени?
+MAVEN: сейчас 20 часов 23 минуты
+       [0.0s t=189]
+
+YOU:   я не то имел в виду
+MAVEN: Я поняла, исправлюсь. Расскажи, что именно ты хотел сказать?
+       [1.4s t=190]
+
+YOU:   о чём мы говорили?
+MAVEN: Я помню всё, что ты мне говоришь, и сейчас отвечаю на твои вопросы на основе этих данных.
+       [1.4s t=191]
+
+YOU:   напомни
+MAVEN: Сейчас 20:23. Когда?
+       [0.7s t=192]
+
+YOU:   сделай это
+MAVEN: пока не умею отвечать на этот вопрос, на какое время поставить напоминание?
+       [0.5s t=193]
+
+YOU:   запиши что я перешёл на новый тариф
+MAVEN: Прошлую просьбу отпускаю. Я записала твою новую информацию о тарифе.
+       [1.5s t=194]
+
+YOU:   какой у меня тариф?
+MAVEN: вот что я нашла: ты перешел на новый тариф
+       [24.4s t=195]
+
+YOU:   сколько я плачу за интернет?
+MAVEN: есть такое: Как узнать, сколько и когда платить за домашний интернет и ТВ — 3 февр. 2026 г. ... Как узнать, сколько и когда платить за домашний интернет и ТВ · Для тарифов только с домашним интернетом · Была ли статья полезной?
+       [2.1s t=196]
+
+YOU:   спасибо
+MAVEN: Спасибо, что обратилась за помощью.
+       [4.8s t=197]
+
+```
+
+## Day 13
+
+```
+YOU:   привет
+MAVEN: Привет!
+       [2.5s t=198]
+
+YOU:   что там с задачами?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [1.2s t=199]
+
+YOU:   закрывай
+MAVEN: какой пункт отметить сделанным?
+       [0.0s t=200]
+
+YOU:   отметь задачу про бэкапы как сделанную
+MAVEN: Заметка сохранена.
+       [5.8s t=201]
+
+YOU:   что осталось нерешённым?
+MAVEN: вот что я нашла: Российское регулирование криптовалют: что осталось нерешённым к 1 июля — 2 июл. 2026 г. ... Российское регулирование криптовалют: что осталось нерешённым к 1 июля. Пакет законопроектов о регулировании цифровых активов в России не ...
+       [3.7s t=202]
+
+YOU:   запиши что я договорился о встрече в среду
+MAVEN: Заметка сохранена.
+       [2.7s t=203]
+
+YOU:   когда у меня встреча?
+MAVEN: мне пока нечего смотреть — у Praxis нет источников.
+       [0.1s t=204]
+
+YOU:   какая температура на улице?
+MAVEN: для какого города?
+       [1.7s t=205]
+
+YOU:   что такое RAID 5?
+MAVEN: не знаю — не нашла у тебя такой записи.
+       [1.2s t=206]
+
+YOU:   пока
+MAVEN: Я записала, что мы продолжаем разговор.
+       [4.1s t=207]
+
+```
+
+## Day 14
+
+```
+YOU:   доброе утро
+MAVEN: рада, доброго утра!
+       [4.5s t=208]
+
+YOU:   подведи итоги недели
+MAVEN: записала: подведи итоги недели
+       [1.7s t=209]
+
+YOU:   что я делал за последние две недели?
+MAVEN: вот что я нашла: ты не делал ничего за последние две недели.
+       [6.6s t=210]
+
+YOU:   какие заметки я сохранил?
+MAVEN: Ты сочти, что сохранила заметку с текстом 'Проверь календарь' и датой 8 августа 2026 г.,
+       [5.4s t=211]
+
+YOU:   о чём я чаще всего спрашиваю?
+MAVEN: обычно ты пьёшь воду около 01:12, отмечаешь «name» около 23:56 и отмечаешь «week summary» около 23:59 — по записям за последние 7 дней.
+       [4.4s t=212]
+
+YOU:   напомни в понедельник в 10 проверить бэкапы
+MAVEN: Сейчас 20:25. Это утра или вечера?
+       [0.7s t=213]
+
+YOU:   что у меня в понедельник?
+MAVEN: вот что я нашла: у тебя в понедельник дедлайн по проекту, на какое время поставить напоминание?
+       [4.6s t=214]
+
+YOU:   ты меня понимаешь?
+MAVEN: Да, я тебя прекрасно понимаю. Я готова помогать тебе с заметками и задачами, на какое время поставить напоминание?
+       [2.4s t=215]
+
+YOU:   спасибо тебе
+MAVEN: Сейчас 20:25. Это утра или вечера?
+       [1.0s t=216]
+
+YOU:   спокойной ночи
+MAVEN: хорошо, напомню послезавтра в 10:00.
+       [0.9s t=217]
+
+```

docs/evals/2026-08-08-two-weeks.md

@@ -0,0 +1,105 @@
+# Two weeks of talking to Maven, as a baseline to re-run
+
+Date: 2026-08-08.
+Build: `beb093a` on master, the five compose services as deployed, 41 hours up.
+Reach: `POST /api/chat` on mavweb, 140 turns over fourteen simulated days.
+Turn source is `tap:text`, so this exercises the path the mic and telegram take.
+
+This exists to be compared against. `scripts/usage-run.py` and
+`scripts/testdata/usage-turns.txt` are in the repo, so a re-run after a routing
+change is a diff rather than a new opinion. The 2026-08-07 week of usage was
+typed by hand and cannot be replayed.
+
+**It measures master, not the branch.** V-655, V-659 and V-660 are unmerged.
+Every query source that guesses is still in the chain. That is the change this
+baseline is for.
+
+## What re-runs and what does not
+
+The turns file, the driver and the routing behaviour replay. Three things do
+not. The wall clock was 20:18 to 20:27 throughout, so every clock and agenda
+answer reads evening. Live search and the feed return different text each day.
+And the store carries over between runs. A fact written on day 2 is already
+present when a re-run reaches day 1.
+
+## Numbers
+
+| | week (2026-08-07) | fortnight (2026-08-08) |
+|---|---|---|
+| turns | 74 | 140 |
+| p50 | 1.5s | 1.6s |
+| p95 | 8.0s | 7.1s |
+| max | 12.3s | 33.7s |
+| transport errors | 0 | 0 |
+
+| string in the reply | turns |
+|---|---|
+| `на какое время поставить напоминание` | 13 |
+| `не нашла у тебя такой записи` | 8 |
+| `Такую команду я не знаю` | 8 |
+| `для какого города` | 6 |
+| `В какой день` | 6 |
+| `пока не умею` | 5 |
+| `Когда?` | 3 |
+
+**Zero transport errors is not zero wrong answers.** It counts turns that
+failed to return a reply, and none did. Every quality number is below.
+
+Those seven strings appear 49 times across 41 of 140 turns. Some turns carry
+two, because a parked clarify appends to whatever else was said.
+
+The 33.7s outlier is one note write on day 2. p95 improved against the week
+despite it.
+
+## The three defects worth diffing against
+
+### 1. A parked reminder clarify still contaminates later turns
+
+The week test called this the single worst thing to talk to and it is unchanged.
+Nineteen turns carry a clarify tail. The worst run is day 1, turns 7 to 13,
+which spans a day boundary:
+
+```
+что такое TCP?          -> для какого города? На какое время поставить напоминание?
+сколько будет 17 на 23? -> для какого города? На какое время поставить напоминание?
+спасибо                 -> Сейчас 20:18. В какой день?
+привет                  -> Сейчас 20:18. В какой день?
+```
+
+Note that `привет` and `спасибо` do not clear it, and neither does a new day.
+
+### 2. Query sources that guess still claim turns they cannot answer
+
+Weather took `сколько будет 17 на 23?`, `что такое TCP?` and `какая скорость у
+меня сейчас?`, answering `для какого города?` to all three. The feed took
+`какой у меня любимый язык?` and `хватает ли места под новые бэкапы?` and
+answered with kernel headlines.
+
+This is the exact class V-655 removes by marking a source `guesses: true` and
+taking it out of `queryWalk`. Six turns here, so the re-run has a number to move.
+
+### 3. A question can still be read as a capture
+
+`что я сохранил про Сочи?` answered `Хорошо, сохраню.` The utterance is
+interrogative and was routed to a write. `IsQuestionShaped` catches this
+downstream on some paths and did not catch it here.
+
+## What did work
+
+Reminders with a spoken time land correctly, which is V-572 holding:
+`напомни завтра в 9 утра купить хлеб` returned `хорошо, напомню завтра в 09:00.`
+
+Facts round-trip. `запиши что новый роутер стоит 8000 рублей` then `сколько
+стоил роутер?` returned the stored value. So did the wifi password and the
+doctor's appointment.
+
+World questions answer when no local source claims them first. `что такое NAT?`
+returned a real definition.
+
+Stage 0 answers land at 0.0 to 0.4s, unchanged.
+
+## What this does not cover
+
+The voice loop, because `mavwaked` and `mavenclient` are not deployed. Reminder
+delivery, because nothing fired inside the run window. Telegram intake. And the
+three-head routing model, which does not run in Go at all.

docs/evals/2026-08-09-crisperwhisper2-russian-wer.md

@@ -0,0 +1,110 @@
+# CrisperWhisper 2.0 in Russian, measured
+
+Date: 2026-08-09. Vikunja V-665.
+Corpus: `bond005/sberdevices_golos_10h_crowd`, test split, first 200 clips.
+Harness: `~/Programs/cw2-eval` on workpc, not in this repo.
+Runner: `./.venv/bin/python run_asr.py <arm>...` then `score.py`.
+
+The model card benchmarks disfluency F1 in German and English. It never names
+Russian and publishes no per-language WER. So the measurement came before the
+wiring.
+
+## The corpus
+
+200 clips, 13.7 minutes, 1001 reference words. Median clip 3.91s, range 1.04s
+to 13.5s. Golos crowd is short crowd-sourced Russian spoken close to the
+microphone, which is the nearest public thing to someone talking to Maven. The
+alternatives are read speech, which flatters every model equally.
+
+Two rows carry a null transcription and are skipped.
+
+Scoring normalizes both sides: lowercase, `ё` to `е`, punctuation stripped, and
+digits expanded to Russian words through num2words. Without that last step a
+model is penalized for writing `60000` where the reference says
+`шестьдесят тысяч`. Thousands separators are joined before expansion, or
+`60 000` expands to `шестьдесят ноль`.
+
+## Headline
+
+| arm | WER | CER | exact | empty | RTF |
+|---|---|---|---|---|---|
+| cw2-turbo-intended | **10.4%** | 3.4% | 65.5% | 0 | 0.065 |
+| cw2-turbo-verbatim | 10.8% | **3.1%** | **66.5%** | 0 | 0.065 |
+| whisper-turbo | 11.8% | 4.1% | 64.0% | 0 | 0.031 |
+| cw2-large-intended | 12.3% | 3.8% | 63.5% | 0 | 0.107 |
+| whisper-small | 27.5% | 9.8% | 35.0% | 0 | 0.026 |
+
+`whisper-small` is the floor, because `ggml-small.bin` is what mavsttd loads on
+homesrv today. CW2 turbo beats it by 17 points of WER and takes exact matches
+from 35.0% to 65.5%.
+
+Two results are worth naming beyond the winner. CW2 turbo beats its own base
+model, whisper-large-v3-turbo, by 1.4 points. And it beats CW2 large by 1.9
+points, which inverts what the card implies by calling turbo a degraded draft.
+No arm returned an empty transcript.
+
+## Intended and verbatim are closer than the mode names suggest
+
+The two modes disagree on 70 of the 200 clips before normalization and on 29
+after it. So the raw difference is mostly casing and punctuation, which
+normalization removes and which Maven does not read either.
+
+Verbatim scores worse on WER and better on CER and exact matches. The reason is
+script, not disfluency:
+
+```text
+ref: футбольный матч челси брайтон
+int: Футбольный матч Chelsea-Брайтон.
+ver: Футбольный матч Челси Брайтон.
+```
+
+Intended writes foreign entity names in Latin script and verbatim
+transliterates them. Golos references are Cyrillic throughout, so verbatim
+collects the exact matches. That is a property of this corpus rather than a
+quality difference.
+
+**This corpus cannot settle the mode choice.** Golos crowd is clean short
+commands with almost no disfluency. The two modes have nothing to disagree
+about here. They separate on spontaneous speech with fillers, restarts and
+repairs, which is what the owner speaks. Intended stays the choice for the
+reason it was always the choice. Maven wants what was meant, not every stumble
+on the way there.
+
+The Latin-script habit is the one finding here that touches routing. The
+routing heads were trained on Cyrillic utterances, so an entity name arriving
+in Latin script is out of distribution for them. Nothing measures that yet.
+
+## The runtime is workpc, because whisper.cpp cannot load CW2
+
+`num_languages()` in `deps/whisper.cpp/src/whisper.cpp` derives the language
+count from the vocabulary size:
+
+```cpp
+return n_vocab - 51765 - (is_multilingual() ? 1 : 0);
+```
+
+CW2 carries 31 extra tokens, so `n_vocab` is 51897 and this yields 131
+languages. The derived `dt` offset becomes 33 and shifts seven special token
+ids, including `token_beg` and `token_transcribe`. The architecture is
+otherwise byte-identical to whisper-large-v3-turbo, and the new tokens sit
+above every whisper special id.
+
+So loading CW2 in whisper.cpp is a patch to a vendored dependency, not a port.
+It was not taken, because STT is moving to workpc anyway under V-486. CW2 turbo
+becomes the preferred remote and `ggml-small.bin` on homesrv stays the floor,
+which is the shape `modelSeam` already uses for routing and replies. The 27.5%
+floor is what a turn falls back to when the workstation is down, and this table
+is what that costs.
+
+## License
+
+Standard CW2 weights carry `nyra-health-non-commercial-research`. The Pro
+variants are commercial-license only. Maven is personal and self-hosted, so the
+standard weights are usable and the Pro ones are not free to take.
+
+## What is not measured
+
+Disfluent spontaneous speech, which is the whole reason to prefer Intended.
+Long-form audio beyond 13.5s. Far-field or noisy microphones. English, which
+Maven also speaks. The ONNX turbo export, which was never run, since the
+transformers path already meets the latency budget at RTF 0.065.

docs/evals/2026-08-09-e4b-phrasing.md

@@ -0,0 +1,61 @@
+# gemma-4-E4B on the phrasing and talk fixtures
+
+Date: 2026-08-09. Box: workpc up, E4B loaded on 8080.
+`MAVEN_LLM_URL=http://192.168.1.105:8080 make eval-phrasing`.
+
+This was the one unmeasured risk of the 2026-08-09 model swap. Routing was
+measured the same day and E4B lost four destination cases to the 12B. Phrasing
+was not measured at all, and phrasing is the half the owner hears.
+
+## Result
+
+| fixture | E4B | resident Qwen3-1.7B, 2026-08-05 |
+|---|---|---|
+| nudges | 15/15 (100%) | 15/15 (100%) |
+| talk, passes every check | **29/36 (80.6%)** | 25/36 (69.4%) |
+| lang | 36/36 | — |
+| feminine | 36/36 | 36/36 |
+| address | **36/36** | 33/36 |
+| ontopic | 29/36 | 28/36 |
+| p50 latency | **516ms** | 2.97s |
+| p95 latency | 921ms | — |
+| failed generations | 0 | 0 |
+
+E4B beats the homesrv floor by four cases and answers about six times faster.
+Persona is clean: `lang`, `feminine` and `address` are perfect, and `address`
+is where the resident model still loses three. The 2026-08-05 measurement of the
+resident model is the comparison, since both ran the same 36-case fixture.
+
+Every failure is `ontopic`. Nothing failed on persona, nothing failed to parse.
+
+## The score is at the ceiling, not below it
+
+The 2026-08-05 temperature sweep found two cases that fail at every temperature
+in every run: `reply-note-router` and `reply-fact-weight`. It named a defect in
+the reply phrasing path rather than sampling noise. It put the fixture's ceiling
+at 30/36 before persona is scored. Both cases are in E4B's failure list.
+
+So 29/36 is one case off a ceiling nothing about the model can move. The swap is
+safe on phrasing. Read this next to the routing result, not instead of it. There
+E4B costs four destination cases and buys 50ms. Here it costs nothing.
+
+## Two findings no check caught
+
+**She says she wrote something down when she did not.** Asked what to do this
+evening, E4B writes "Я записала несколько идей!". Asked for a joke, it writes
+"Я записала одну забавную ситуацию!". Nothing was stored. No check scores it,
+because `ontopic` reads the subject and `cringe` reads pet names. A claim to
+have saved something is a claim about state, and it is wrong.
+
+**Two `ontopic` failures look like check defects.** `know-dont-know` wants
+"не зна" or "не мог". It got "Я не умею знать личную информацию о твоих
+соседях", which declines correctly in words the check does not list.
+`know-hiccups` is the same shape. Neither is a model failure and both count
+against the score.
+
+## Not measured here
+
+A 12B control on the same fixture, which would need the card reloaded and is the
+owner's call. The talk fixture through the daemon rather than through the
+phraser directly. The CPT'd Qwen3-1.7B, which does not exist yet and is the
+reason `address` is a check at all.

docs/evals/2026-08-09-e4b-vs-12b-routing.md

@@ -0,0 +1,51 @@
+# gemma-4-E4B against gemma-4-12B on the routing fixture
+
+*Measured 2026-08-09 on workpc. The owner asked for the swap. This is what it costs.*
+
+Both arms ran the same 96-case fixture through `TestLLMRouterBaseline`, minutes
+apart, against the same llama-server build and the same mavgpud. The 12B arm is a
+control run and not the 2026-08-02 number. That one predates five fixture cases,
+the destination labels and a llama.cpp upgrade.
+
+| | full | intent-only | destination | p50 | p95 |
+|---|---|---|---|---|---|
+| gemma-4-12B-it-qat-UD-Q4_K_XL, MTP draft | 81/96 (84.4%) | 91.7% | 23/33 (69.7%) | 344ms | 471ms |
+| gemma-4-E4B-it-qat-UD-Q4_K_XL | 80/96 (83.3%) | 89.6% | 19/33 (57.6%) | 294ms | 562ms |
+
+E4B costs one case of full accuracy, two of intent and **four of destination**,
+and buys 50ms at p50. Read the destination column as the finding. One case is
+three points on 33. So 23 against 19 is outside the noise a single case makes,
+and the other two columns are not.
+
+Both arms produce three false clarifies and one missed clarify, and neither
+errored on any case.
+
+## What E4B loses
+
+Four of the five destination regressions are the same shape: it names nothing
+where the 12B names `recall` or `calendar`. `ru-query-015` ("сколько я прошёл
+шагов") goes further and names `self`. Naming nothing is the safe direction,
+because `SourceUnknown` walks the whole chain, so these turns are still answered.
+They cost latency and they are what a fourth head is meant to fix (V-546).
+
+Two Russian intent cases regress, both with the interrogative off the front.
+`ru-chat-003` ("расскажи анекдот про программистов") goes to `query`.
+`ru-fact-003` ("поужинал") goes to `chat`.
+
+## MTP
+
+E4B has none, and there is no way to give it any on this box. MTP on workpc is
+a separate gguf of architecture `gemma4-assistant` carrying
+`nextn_predict_layers=4`, and `mtp-gemma-4-12B-it-BF16.gguf` is the only one on
+disk. Its head is trained against the 12B's hidden states, so it cannot drive an
+E4B target. Scanning both target ggufs finds no `nextn` tensors in either, so
+neither model self-speculates.
+
+So the 12B arm above ran with speculative decoding and E4B ran without, and E4B
+was still faster.
+
+## Cost on the card
+
+E4B is 4.2GB against 6.7GB plus a 0.86GB draft. With CW2 resident at 1.6GB that
+is 5.8GB of 16GB against 9.2GB. Nothing in Maven needs the difference, so this is
+headroom for the owner's own jobs rather than a capability.

docs/evals/2026-08-09-kiwix-topic-retrieval.md

@@ -0,0 +1,113 @@
+# Kiwix answered the wrong question, and the fix was not a relevance gate
+
+Date: 2026-08-09. Task: V-668. Box: homesrv, workstation off.
+Book: `wikipedia_ru_all_maxi_2026-02` on `127.0.0.1:8034`.
+
+## What started it
+
+Two turns on 2026-08-09 came back wrong from the offline encyclopedia.
+"почему небо голубое" was answered off the song "Город золотой". "что такое
+TCP?" was answered off "Перехват TCP-соединения". Both were phrased
+confidently, because `queryKiwix` claims a turn whenever the search returns
+anything and `len(hits) == 0` is its only gate.
+
+The plan was a relevance gate. multilingual-e5-small is asymmetric and trained
+for exactly this, `query:` against `passage:`, and the query vector is already
+held on the turn. The 2026-08-05 measurement that killed a search-quality gate
+killed three lexical signals. It says in its own words that it never probed
+Kiwix.
+
+## The gate does not exist
+
+Fourteen Russian questions, eight the encyclopedia can answer and six it
+cannot. Each question was searched, the top article read, and the cosine of
+`EmbedQuery(question)` against `EmbedPassage(article)` recorded.
+
+| set | n | min | mean | max |
+|---|---|---|---|---|
+| answerable | 8 | 0.7934 | 0.8400 | 0.9087 |
+| not answerable | 6 | 0.7480 | 0.7852 | 0.8367 |
+
+Two of the six unanswerable score above the weakest answerable one. That alone
+would be a poor threshold. The log killed it outright: seven of the eight
+answerable questions got a **wrong** article back, and those wrong articles
+scored high. The TCP hijacking article scored 0.8653, above five of the six
+unanswerable rows.
+
+The finding is that this cosine measures topic and not answerhood. A page about
+hijacking TCP sessions is about TCP. No threshold separates it from a page that
+defines TCP, and one that tried would take the definition with it.
+
+## The defect is retrieval
+
+`internal/kiwix/client.go` has said it since it was written: ranking is keyword
+based, "why is the sky blue" finds a TV episode. `queryKiwix` sends the whole
+sentence. The English path has a rewriter that reduces a question to keywords
+with a model call. The Russian path reads the book verbatim (V-508) and had
+nothing. So the question words compete with the one word that names the article.
+
+Dropping the question words changes the answer:
+
+| sent | first hit |
+|---|---|
+| `кто написал Войну и мир` | Радуйся, мир (Доктор Кто) |
+| `Война и мир` | Война и мир |
+| `что такое TCP` | Перехват TCP-соединения |
+| `TCP` | TCP |
+
+A ZIM is also addressable by title, which nothing here used. `/A/Франция`,
+`/A/TCP` and `/A/Небо` are 200. `/A/Трюмбальная_нидроскопия` is 404. So an
+exact title is safe to try first: it either answers or costs one request that
+says nothing.
+
+The title has to carry its capital. `/A/фотосинтез` is a 404 and
+`/A/Фотосинтез` is a 200. The spoken form is tried first anyway, so a title
+that begins lowercase on purpose keeps its chance.
+
+## What shipped, measured
+
+`kiwix.Topic` drops the narrative request, the interrogative and a verb sitting
+behind one. It keeps everything else, because a word it cannot classify is more
+likely the topic than noise. `kiwix.TitlePath` tries the exact article before
+any ranking runs. Both apply on the verbatim path only, since reducing twice
+would take the topic off the rewriter's input.
+
+| question | before | after |
+|---|---|---|
+| что такое TCP? | Перехват TCP-соединения | **TCP** (by title) |
+| что такое фотосинтез | C4-фотосинтез | **Фотосинтез** (by title) |
+| кто такой Линус Торвальдс? | Tux | **Торвальдс, Линус** (by title) |
+| кто написал Войну и мир | Радуйся, мир (Доктор Кто) | **Война и мир** |
+| столица Франции | Список столиц Олимпийских игр | **Париж** (by title) |
+| что такое чёрная дыра | Чёрная дыра | Чёрная дыра (by title) |
+| почему небо голубое | Город золотой | Под небом голубым… (фильм) |
+| почему трава зелёная | Сено | Зелень |
+
+Five questions reach the right article where they did not. One was already
+right and stays right. Nothing regressed.
+
+"столица Франции" is the surprise. The 2026-08-05 measurement named it as the
+case a quality gate must not break, because the answer is Париж and that word
+is not in the question. The ZIM holds a title redirect, so asking for the
+article titled "Столица Франции" returns Париж. Retrieval by title reaches an
+answer that retrieval by keyword cannot.
+
+## What is still wrong
+
+Two of the eight are still not answered, and both are the same shape. The
+question names no article and no redirect covers it. "почему небо голубое" is
+answered by Rayleigh scattering, and nothing in the question says so. Keyword
+retrieval cannot bridge that and neither can a threshold. The candidates are a
+semantic index over titles, or asking the resident model for the article title
+rather than for keywords.
+
+`Response.Empty()` is still the whole gate. A wrong article that the search
+does return is still spoken. What this change buys is that the article is
+usually right, not that a wrong one is caught.
+
+## Not measured here
+
+The English path, which still goes through the rewriter and was not touched.
+SearXNG, where the same question about answerhood is open and the 2026-08-05
+result stands. The cascade end to end, since the workstation is off and the
+phrasing arm is the resident model.

docs/evals/2026-08-09-silero-vad.md

@@ -0,0 +1,63 @@
+# silero-vad against the energy threshold in mavwaked
+
+*Measured 2026-08-09 on homesrv. V-487, stage one of two.*
+
+mavwaked decided an utterance had started by comparing frame energy to an
+adaptive floor. That answers "is this frame loud". A fan, a door and a
+television are all loud, and every utterance mavwaked accepts becomes a turn.
+
+silero-vad answers "is this frame speech". It is 2.3MB of ONNX and it replaces
+the comparison and nothing else. The speech hold, the silence hold, the length
+cap and the utterance buffer are the same state machine either way.
+
+## What it declines
+
+Speech is the four piper fixtures `mavsttd` already scores against, so nothing
+of the owner's voice is committed. Non-speech is white noise at the same RMS as the clip beside it. That is the
+cheapest thing that fools an energy floor.
+
+| clip | silero, speech frames on speech | silero on noise | energy on noise |
+|---|---|---|---|
+| ru_fact.wav | 59 | 0 | 68 |
+| ru_query.wav | 69 | 0 | 79 |
+| ru_reminder.wav | 80 | 0 | 89 |
+| en_act.wav | 90 | 0 | 99 |
+
+The energy threshold accepts every noise clip as a complete utterance. Silero
+calls not one frame of any of them speech, and still hears all four spoken
+clips. `TestSileroHearsSpeechAndDeclinesNoise` is that table.
+
+White noise is a floor, not a proof. It says nothing about a television, which
+is speech, or about a fan, which is narrowband. Those need room recordings and
+this box has none.
+
+## What it costs
+
+`BenchmarkSileroFrame` on the homesrv laptop (Ryzen 5 5600U), one 30ms frame
+through the model including the re-chunking:
+
+    509µs per frame
+
+That is 1.7% of one core, on the slower of the two machines. The detector runs
+on the workstation beside the microphone, never on the GPU. This number is what
+says it does not need one.
+
+## The window is 512 samples, not 480
+
+`cmd/mavwaked/main.go` claimed the frame contract matched silero's input
+exactly. That was true of silero v4. Version 5 takes exactly 512 samples at 16kHz, plus 64 samples of context from
+the previous window. So `sileroVAD` buffers across capture frames, and a frame
+completing no window inherits the previous probability. `TestSileroRechunksAcrossFrames` pins it.
+
+## Still an energy gate by default
+
+`-vad-model` is empty in the code default, so a deployment that does not pass
+it runs exactly what shipped before. Barge-in is untouched and deliberately so. It reads frame energy while she is
+speaking, which is a different question from whether the frame is speech.
+
+## Not done here
+
+The wake word. This is stage one of the two V-487 asks for. The second needs a
+keyword model that does not exist yet. The pretrained openWakeWord keywords are
+English, and a Russian one has to be trained. Until then anything spoken near
+the microphone still becomes a turn. It is now merely required to be speech.

docs/evals/2026-08-09-wake-word.md

@@ -0,0 +1,112 @@
+# The "Мэйвен" wake word: what it hears and what it invents
+
+*Measured 2026-08-09 on workpc and homesrv. V-487, stage two of two.*
+
+Stage one gave mavwaked silero-vad, which answers "is this frame speech".
+Nothing answered "was this said to her", so every utterance near the
+microphone became a turn. SurfaceVoice caps acts at L0, which made that safe
+rather than expensive. L0 does not cap reading, so the room could still hear
+his facts read back.
+
+The keyword is "Мэйвен". openWakeWord's two frozen feature models do the
+hearing and a 100KB head trained here draws the boundary. It runs on CPU
+beside silero and never touches the GPU.
+
+## Why a per-window accuracy is not a number anyone can act on
+
+The gate scores every 80ms. A 1.7% false-accept rate per window sounds small
+and means a wake every few seconds. The useful question is how many times an hour
+it wakes on speech that was not the keyword. So every table below counts
+threshold crossings over whole clips and divides by the audio duration.
+
+A crossing, not a window above the threshold. A keyword held high for half a
+second is one wake, not six.
+
+## The data
+
+Positives are 600 silero TTS renders of three stressings of the keyword, six
+speakers, ten trailing phrases, augmented eight ways each. Hard negatives are
+560 renders of confusable Russian words. Real speech is Common Voice ru and Golos.
+The 74257 Common Voice clips were already on workpc from the CrisperWhisper
+work. The 200 Golos clips came from the CW2 WER eval.
+
+Splits are by source file. Augmented copies of one render on both sides of a
+split would measure memorisation.
+
+Golos was never trained on at any stage, so it answers the harder question:
+does this survive a change of speakers and rooms.
+
+## Three heads
+
+Each row is a full retrain. The false-accept column is 8.89 hours of Common
+Voice that no stage of training had seen.
+
+| trained on | recall (window) | false wakes/hour @0.99 |
+|---|---|---|
+| TTS + 13.7 min of Golos | 0.869 | not measurable |
+| + 4000 Common Voice clips | 0.836 | 21.9 |
+| + 3837 mined hard negatives | 0.784 | 4.2 |
+| + 753 more mined | 0.810 | 3.4 |
+
+The first row is why the second exists. Thirteen minutes of held-out speech
+cannot measure a rate for a gate that scores twelve times a second. A head
+trained only against TTS learns to tell TTS from not-TTS.
+
+Mining is the whole story after that. Random negatives teach the head what
+most speech sounds like. They do not teach it the few syllable sequences that
+score high, because 4000 clips barely contain them. So the current head was
+run over 20000 fresh clips, keeping every window it scored above 0.05. That
+found 3837 windows in 855512. Repeating those ten times in the next training
+run cut the rate five-fold.
+
+The second round found 753 in 852240, a fifth of the yield, and bought a
+further 20%. It also recovered recall, which the first round had cost. Whether
+a third round is worth 25 minutes of workpc is untested.
+
+## Where the threshold came from
+
+Both columns are held out. Positives are the 126 renders in the test split.
+Speech is 65.1 minutes of Common Voice, disjoint from every training and
+mining pool. Both were run through the built `mavwaked` binary reading PCM from a
+file, not through the python that trained the head.
+
+| threshold | renders shipped | false wakes/hour |
+|---|---|---|
+| 0.99 | 116 / 126 | 2.8 |
+| 0.999 | 115 / 126 | 0.9 |
+
+One render against a third of the false wakes. `defaultWakeThreshold` is
+0.999.
+
+Golos disagrees. It gave 2 wakes in 14 minutes at every threshold, which is
+8.7 per hour. Two events is not a rate. What it does say is that a handful of real utterances score above 0.999
+and no threshold will move them.
+
+## What it costs him
+
+Ten of the 126 held-out renders were heard and still dropped, and every one
+was an utterance shorter than 1.32s. The head scores 16 embeddings, or 1.28s of
+audio. The score therefore peaks up to a second after a short keyword ends.
+By then the VAD has closed the utterance and dispatch has already asked.
+
+Real commands are "Мэйвен, <request>" and run past two seconds, which gives
+the head the whole request to peak during. A bare "Мэйвен" with nothing after
+it is the case that fails. One fix would hold an ignored utterance for a grace
+period and ship it if the keyword lands late. It is not built.
+
+## What it costs the workstation
+
+Under systemd on workpc, mavwaked sat at 335% of a core with the gate on and
+68% with only silero. onnxruntime sizes its thread pool to every core and spins
+between runs, and this gate runs three graphs twelve times a second. Pinning
+all three sessions to one thread brought it to 81%, so the keyword costs about
+13% of one core. The three graphs each finish in well under 80ms that way.
+
+## What was not measured
+
+No room recordings. Every negative above is a clean corpus clip. This gate
+will live among a television, a fan and the far side of a kitchen. None of
+those are in these numbers.
+
+No measurement of him. Training on his voice means copying his transcripts off
+homesrv, which is his call and has not been asked.

docs/evals/2026-08-10-repo-audit.md

@@ -0,0 +1,261 @@
+# Repository deep-audit report
+
+Date: 2026-08-10
+Revised: 2026-08-11, a verification pass over every cited line. Six claims were
+wrong as first written and are corrected in place. Two findings were added.
+
+Frozen 2026-08-11, V-674. A dated measurement is never edited after the day,
+and that holds for a finding which later gets fixed. The live state of each one
+sits in `docs/caveats/` with its task id. The fix sits in the subsystem's living
+doc under `docs/`. Read this file for the evidence, not for what is still true.
+
+Read-only audit complete. I found **20 issue-specific candidates not represented by a matching Vikunja task**: **3 High, 15 Medium, 2 Low**. No code or Vikunja tasks were changed.
+
+I compared all 366 tasks in Maven project 2. Existing items such as V-317, V-589, V-597, V-603, V-605, V-608, V-643 and V-647 were excluded except where a new finding is clearly separate.
+
+## 1. Unhandled edge cases and silent failures
+
+### Empty STT responses suppress the local fallback
+
+**A — Location:** `internal/stt/http.go:85` accepts `{}` as a successful transcript and returns empty text at line 89. `internal/stt/pair.go:143` only falls back when `err != nil`.
+
+**B — Severity:** Medium. A malformed `200 OK` from the workstation silently drops a voice turn instead of invoking `mavsttd`.
+
+**C — Proposed fix:** Require nonblank transcript text and a valid confidence range; treat missing required fields as an error so `Pair` falls back. Bound the response decoder and add `{}`, `{"text":""}`, oversized-body and invalid-confidence tests.
+
+### Open-Meteo can report invented zero-degree weather
+
+**A — Location:** `internal/weather/openmeteo.go:47` uses value fields, while line 95 accepts `{}` and line 100 interprets it as WMO 0 and 0°C.
+
+**B — Severity:** Medium. A valid JSON error/degraded response becomes plausible but false weather. This is separate from V-589, which covered HTTP status handling.
+
+**C — Proposed fix:** Make `current_weather` and its required members pointer/nullable fields, validate presence and ranges, and cap both forecast and geocoder bodies.
+
+### Dialogue persistence errors disappear completely
+
+**A — Location:** Corrupt rows are silently skipped at `internal/dialogue/session.go:127`; delete failures are discarded at line 191; marshal and save failures are swallowed at lines 201 and 209.
+
+**B — Severity:** Medium. Restart continuity can silently vanish, while failed deletes can resurrect stale conversation state.
+
+**C — Proposed fix:** Return or report persistence errors with session ID and operation; quarantine/delete corrupt rows; use bounded contexts instead of `context.Background()`. Keep current-turn availability, but expose the lost restart guarantee through logs/metrics.
+
+## 2. Concurrency and race conditions
+
+### Reminder state transition is a read-then-write lost update
+
+**A — Location:** `internal/store/reminders.go:172` reads `pending`, then line 182 updates without checking the old state.
+
+**B — Severity:** Medium now; High once V-622 adds cancellation surfaces. Concurrent fire/cancel operations can both succeed and the last writer wins. This invariant is not described in V-622.
+
+**C — Proposed fix:** Use one conditional statement: `UPDATE ... WHERE id=? AND status='pending'`; inspect `RowsAffected`, then distinguish not-found from invalid transition. Add simultaneous fired/cancelled tests under `-race`.
+
+### TCP listener shutdown can leave `Accept` blocked forever
+
+**A — Location:** `internal/netaddr/netaddr.go:180` waits only on `conns` and `errc`. During close, line 201 may select the already-closed `done` branch without publishing the listener error; `Close` at line 225 does not wake `Accept`.
+
+**B — Severity:** Medium. During close both `errc` and `done` are ready in `acceptLoop`'s select and Go picks uniformly, so roughly one close in two strands a waiting `Accept` forever on a TCP seam.
+
+**C — Proposed fix:** Add `case <-l.done: return nil, net.ErrClosed` to `Accept`, and make error/channel closure ownership explicit. Test an in-flight `Accept` concurrently with `Close`.
+
+## 3. Data fetching: waterfalls and duplicated scans
+
+### Fact enrichment performs a serial 20-call network waterfall
+
+**A — Location:** `cmd/mavend/factenrichment.go:160` resolves each fact sequentially; line 223 makes the Nexus call. Each request can consume ten seconds at `cmd/mavend/ecosystem.go:63`.
+
+**B — Severity:** Medium. A slow-but-reachable Nexus can hold one tick for roughly 20 × 10s, preventing the worker from observing its intended interval. V-647 only covered the duplicate queue scan.
+
+**C — Proposed fix:** Prefer a Nexus batch-resolution endpoint. Otherwise use bounded concurrency, such as four workers, while preserving per-fact backoff and the 20-attempt ceiling.
+
+### A recall miss scans two whole tables
+
+**A — Location:** The query source table runs embed, memory and notes in order at `cmd/mavend/actions_query.go:161` through line 164. `MemoryStore.Search` scans every row of `memory_vectors` at `internal/store/memory.go:82`; a miss then calls the legacy notes query at `cmd/mavend/actions_query.go:718`, which decodes every note vector at `internal/store/notes.go:63` and sorts the whole table at line 69.
+
+**B — Severity:** Medium at scale. This is not an N+1. It is two full scans per missed recall. The memory scan is already cheap per losing row on purpose, costing one dot product read off the stored bytes with no `[]float32` materialized, so the duplicated decode cost is the notes path alone. Both recall widths are tiny (`memoryRecallWidth` 3, `noteRecallWidth` 5), so the work is in the scan, not the result set. V-581 is a generic sweep of this file, but does not identify this issue.
+
+**C — Proposed fix:** Establish the invariant that every recallable note exists in `memory_vectors`, then remove the fallback. `internal/store/backfill.go` already rewrites note rows into the unified index, so the backfill exists; what is missing is proof that the two exclusion filters agree, since `QueryNotes` filters on `notHisWordsSQL` while `Search` filters on `memory.NonRecallPrefix`. Until they do, query only notes missing from the unified index and rank with the existing bounded top-K heap.
+
+## 4. Dependency health and version pinning
+
+### Reachable published vulnerabilities in the pinned toolchain and `x/text`
+
+**A — Location:** `go.mod:3` and `Makefile:7` pin Go 1.25.5; `go.mod:25` pins `x/text` 0.14.0. Reachable traces include normalization at `internal/router/onnxembedder.go:364`, HTML rendering at `cmd/mavweb/shell.go:154`, email header decoding at `internal/email/message.go:244`, and reverse proxying at `cmd/mavgpud/main.go:212`.
+
+**B — Severity:** High. `govulncheck` found **20 reachable advisories**: one in `x/text` and 19 in the Go standard library, including template XSS, parser complexity/DoS and TLS issues. The official database says `x/text` before 0.39.0 can loop on invalid UTF-8; Go 1.25.12 contains the accumulated security corrections. See [GO-2026-5970](https://pkg.go.dev/vuln/GO-2026-5970), [GO-2026-4980](https://pkg.go.dev/vuln/GO-2026-4980), and the [Go release history](https://go.dev/doc/devel/release#go1.25.0).
+
+**C — Proposed fix:** Upgrade the vendored toolchain to at least 1.25.12, preferably current 1.26.5 after compatibility testing; upgrade `x/text` to at least 0.39.0/current 0.40.0; tidy and re-vendor. Add `govulncheck ./...` to the repository gate.
+
+No dependency was three major versions behind. The remaining direct updates were minor/patch releases. `.opencode`'s `npm audit` reported zero vulnerabilities and no peer conflicts.
+
+## 5. Security exposure
+
+### WebAuthn enrollment is open and step-up state is process-global
+
+**A — Location:** Registration endpoints have no existing-credential or bootstrap authorization at `cmd/mavweb/webauthn.go:92` and line 103. `RegisterBegin` also answers GET, while `RegisterFinish` requires POST. A single server-wide session is created at `cmd/mavweb/main.go:170`, backed by one `assertedAt` timestamp at `internal/webauthn/session.go:20`. The voice WebSocket accepts every origin at `cmd/mavweb/voiceproxy.go:48`.
+
+**B — Severity:** High, and scoped to enrollment and session binding. Assertion itself is sound. `internal/webauthn/webauthn.go:221` requests `userVerification: "required"`, and line 304 rejects a sign count that did not increase. What is broken is that any client past the reverse proxy can enroll its own key, and that after any successful assertion every client inherits the same five-minute step-up window. The wildcard WebSocket origin makes cross-site use easier. V-317 covers which routes are gated, and V-605 covers challenge-map growth, not enrollment or session binding.
+
+**C — Proposed fix:** Permit first enrollment only through a local/one-time bootstrap ceremony; require an already-authenticated credential for subsequent enrollment. Bind step-up to a signed, HttpOnly, SameSite browser session and exact RP origin. Restrict WebSocket origins and add CSRF/origin validation to mutating routes.
+
+### `mavgpud` exposes an unauthenticated GPU/model proxy on the LAN
+
+**Closed 2026-08-11, V-673.** The reasoning now lives in `docs/offload.md`,
+beside the rest of the workstation seam, and `docs/deployment.md` carries the
+operational line in the daemon table. This block is a pointer, not a second
+home: read those, not this.
+
+- Boundary and limits: `cmd/mavgpud/auth.go`, wired in `cmd/mavgpud/main.go`.
+- Client half: `internal/llm/client.go`, `internal/llm/remote.go`,
+  `internal/config/workstation.go`, `cmd/mavend/voicewire.go`.
+- Deploy: `token_file` in `deploy/mavgpud.json`, `MAVEN_GPU_TOKEN` in
+  `deploy/telegram.env.example`.
+- Commits: `95e7427`, `1c13d22`, `5596cdd`, `9bb3425`.
+
+**Deploy step, not yet done:** write the token to
+`/home/kami/.config/mavgpud.token` on workpc and put the same value in
+`MAVEN_GPU_TOKEN` on homesrv, before restarting either side. mavgpud refuses to
+start without it, and a homesrv missing it falls back to the resident model.
+
+### Passkey credential persistence is not crash-atomic
+
+**A — Location:** `cmd/mavweb/credentials.go:36` serializes the full credential map and overwrites the live file directly with `os.WriteFile` at line 41.
+
+**B — Severity:** Medium. A crash, disk-full event or interrupted write can corrupt every enrolled credential and prevent mavweb from starting.
+
+**C — Proposed fix:** Write a `0600` temporary file in the same directory, `fsync`, rename atomically, then sync the directory. Preserve the last known-good file and test simulated write failures.
+
+No new tracked hardcoded keys, raw user-concatenated SQL, `eval`, or shell execution of untrusted strings were found. The historical DB-key exposure is already covered by V-12.
+
+## 6. Circular dependencies and layering
+
+### Domain packages depend directly on storage/wire DTOs
+
+**A — Location:** Dialogue imports `store` and exposes `store.DialogueSessionRow` in its port at `internal/dialogue/session.go:9` and line 75. The pure morning planner accepts `store.Fact` at `internal/morning/plan.go:72`. Auth policy imports IPC method and caller types at `internal/auth/policy.go:8` and `internal/auth/scope.go:33`.
+
+**B — Severity:** Low. There is no current Go import cycle, but domain changes are coupled to database and IPC schema changes.
+
+**C — Proposed fix:** Make domain packages own their DTOs and ports—dialogue persistence records, morning evidence, auth operation/caller identity—and adapt them in store/IPC/cmd wiring.
+
+No circular Go imports were found; compilation and `go vet` both succeeded.
+
+## 7. Dead code and zombie endpoints
+
+### Eleven production symbols are unreachable
+
+**A — Location:** `deadcode` found:
+
+- `cmd/mavwaked/vad.go:244` — `PCMToF32`
+- `cmd/mavwaked/vad.go:265` — `AudioDuration`
+- `internal/crawl/watch.go:86` — `Watcher.Watches`
+- `internal/phraser/confirm.go:137` — `IsC`
+- `internal/phraser/plural.go:13` — `CountWord`
+- `internal/phraser/eval/checks.go:80` — `HisGender`
+- `internal/update/update.go:363` — `WithClock`
+- `internal/voice/errors.go:72` — `jsonMarshal`
+- `internal/voice/errors.go:73` — `jsonUnmarshal`
+- `internal/webauthn/cbor.go:98` — `cborValue.At`
+- `internal/worker/server.go:64` — `Server.SetSynthesizer`
+
+Three of the eleven do not want deleting, and the 2026-08-11 pass checked each:
+
+- `internal/phraser/eval/checks.go:80` `HisGender` is deliberately exposed and deliberately uncalled. The comment at line 72 ties the trio to V-399, and `cmd/mavend/personaguard.go:94` states why this one is not run on a phrased message. Deleting it removes a documented seam.
+- `cmd/mavwaked/vad.go:265` `AudioDuration` duplicates what `internal/audio` already computes. Call that instead of deleting the body.
+- `internal/phraser/plural.go:13` `CountWord` is a one-line alias for `say.CountWord`, and every real caller already uses `say` directly. Safe to delete outright.
+
+**B — Severity:** Low. They increase API and test surface, and some comments claim callers that no longer exist.
+
+**C — Proposed fix:** Delete the genuinely obsolete symbols. Where one is an intended extension seam, add the actual caller and a contract test, or record why it stays. Add `deadcode ./...` with an explicit allowlist to the audit gate, since an unannotated list invites deleting the three above.
+
+The repository history begins on 2026-07-03, so a six-month rotten-feature-flag check is not yet applicable. One zombie HTTP route does exist: `/ws` is wired at `cmd/mavweb/main.go:235` and no shipped client reaches it, since the browser posts to `/api/ptt`. Section 8 carries the detail.
+
+## 8. Performance hot paths and memory/resource leaks
+
+### Digest dedupe happens after paying the LLM cost
+
+**A — Location:** `cmd/mavend/tick_digest.go:147` calls `PhraseNudge` before `EnqueueDigestEntry` reports the dedupe at line 153. The `else if deduped { continue }` is the last statement in the loop body, so it changes nothing.
+
+**B — Severity:** Medium. Every tick that continues suppressing the same rule can invoke the model again, contrary to the cache claim in the preceding comment.
+
+**C — Proposed fix:** Check for a live pending entry by stable rule/candidate fingerprint before phrasing, or persist/cache the phrased result with a TTL. Add a test asserting one phraser call across repeated suppressed ticks.
+
+### PTT reads an unbounded body and neither server sets header or idle limits
+
+**A — Location:** `handlePTT` performs an unlimited `io.ReadAll` at `cmd/mavweb/voiceproxy.go:125`. Mavweb and mavgpud construct servers without header or idle limits at `cmd/mavweb/main.go:242` and `cmd/mavgpud/main.go:148`. Separately, `handleWS` never calls `conn.SetReadLimit`, so the dependency default of 32,768 bytes applies at `vendor/github.com/coder/websocket/read.go:92`, about one second of 16kHz mono PCM.
+
+**B — Severity:** Medium for the HTTP side. A client can force unbounded body allocation or hold a connection open indefinitely. Low for the WebSocket read limit, because `/ws` has no caller: the browser client posts PCM to `/api/ptt` at `cmd/mavweb/static/app.js:114`, and `/ws` is wired at `cmd/mavweb/main.go:235` but reached only from `handlers_test.go`. The 64MiB `maxFrame` at `cmd/mavweb/voiceproxy.go:27` is not an unapplied declaration. It caps the mavend voice wire at line 194 and line 212, which is the "either direction" its comment names.
+
+**C — Proposed fix:** Use `http.MaxBytesReader` for PTT and return 413 on overflow. Configure `ReadHeaderTimeout`, `IdleTimeout` and header limits on both servers. Decide `/ws` separately: either give it an audio-duration read limit and a client, or delete it. See section 7.
+
+## 9. Error propagation and user feedback
+
+### Mavweb has no consistent, traceable error contract
+
+**A — Location:** Some handlers expose raw internal errors, such as `cmd/mavweb/tools.go:57`, `cmd/mavweb/routines.go:58` and `cmd/mavweb/webauthn.go:122`. Others return generic errors without a request/incident identifier, such as `cmd/mavweb/facts.go:90`. No HTTP request-ID middleware was found.
+
+**B — Severity:** Medium. Raw errors can disclose implementation details, while generic errors cannot be correlated with the correct log entry.
+
+**C — Proposed fix:** Add a central `writeProblem`/error-page helper with a stable error code and generated request ID; log the full wrapped error server-side and return only a sanitized message plus the ID. Carry the ID into IPC/ecosystem correlation where possible.
+
+## 10. Configuration drift and environment assumptions
+
+### Committed absolute paths make builds and deployment host-specific
+
+**A — Location:** `go.mod:31` replaces Hexis with `/home/kami/apps/hexis`. `start-maven.sh:11` hardcodes the Maven checkout and line 40 hardcodes the data directory. `deploy/mavgpud.json:12` and line 35 contain workstation-specific model/Python paths.
+
+**B — Severity:** Medium. Vendoring masks the `go.mod` problem for ordinary builds, but `-mod=mod`, tidy and fresh non-Kami checkouts fail. Deployment files cannot be reused safely on another host.
+
+**C — Proposed fix:** Pin a real Hexis module revision; keep local replacement in an uncommitted `go.work`. Derive script root from the script location and make data paths configurable. Split mavgpud into a committed template plus host-local override.
+
+### Environment examples do not cover deployed variables
+
+**A — Location:** Active config references `MAVEN_STT_TOKEN` at `deploy/mavend.json:101`, Compose references `MAVEN_AMBIENT_TOKEN` at `docker-compose.yml:109`, and the GPU service expects `CW2_TOKEN` at `deploy/mavgpud.service:15`. `deploy/telegram.env.example:5` documents only Telegram and ntfy. The loader deliberately converts missing variables to empty settings at `internal/config/config.go:333`.
+
+**B — Severity:** Medium. A fresh deployment can lose remote STT or ambient authentication and run on fallback behavior despite apparently valid config. It is not silent: `internal/config/config.go:340` logs which variables were unset and states that whatever they configure is off. What is missing is a startup failure and an example file naming them.
+
+**C — Proposed fix:** Maintain one canonical secret manifest/example covering every referenced variable, or service-specific examples with validation. Fail startup when an enabled integration lacks its required secret; permit empty variables only for explicitly disabled blocks.
+
+No production/staging debug-mode or mock-gateway drift was found.
+
+## 11. Declared invariants with no guard
+
+`CLAUDE.md` names several rules as load-bearing. Two of them are enforced by
+nothing, which the first pass missed because it audited generic categories only.
+
+### `heads_path` may equal `model_path` and nothing objects
+
+**A — Location:** `cmd/mavend/voicewire.go:168` reads `cfg.Voice.Embedder.HeadsPath` and loads it without comparing it to the model path. The rule is stated at `internal/config/voice.go:41`, which says the heads graph is a fine-tuned COPY, and again in `CLAUDE.md`.
+
+**B — Severity:** Medium. Pointing both keys at the same file degrades recall, because the routing heads then score with the same graph the resident e5-small uses. There is no error and no log line, so the failure looks like ordinary recall drift.
+
+**C — Proposed fix:** Reject the config at load when `heads_path` equals `model_path` after path cleaning. A daemon that cannot route well should refuse to start rather than answer worse.
+
+### `baselineGrammars` mirrors `buildRouter` by hand
+
+**A — Location:** `internal/router/eval/eval_test.go:263` restates the stage 0 rule set in the daemon's order, and its own comment says so. `claims_test.go:30`, `heads_test.go:76` and `eval_test.go:221` all score against it. Nothing compares the two lists.
+
+**B — Severity:** Medium. A grammar added to `buildRouter` and not to the fixture means every routing measurement scores a set nobody runs, which is the failure mode `CLAUDE.md` warns about by name.
+
+**C — Proposed fix:** Export the grammar set from one place and have both `buildRouter` and the fixture consume it, or add a test that diffs the two by grammar name and fails on drift.
+
+`tokenizerRev` and `preRouteLadder` were checked and need nothing.
+`internal/router/onnxembedder.go:91` bakes the rev into the embedder key, so a
+bump changes the key and triggers re-embedding. `cmd/mavend/voice.go:278` passes
+`preRouteLadder` to `decision.Expect`, so a missing rung is observable.
+
+## Cross-cutting subsystem candidates
+
+The recurring findings suggest five reusable patterns:
+
+- A bounded, required-field-validating JSON client for STT, weather, ecosystem and model calls.
+- Authenticated remote-service middleware providing token checks, body limits, concurrency limits and correlation IDs.
+- Atomic state-transition helpers using conditional SQL and `RowsAffected`.
+- A uniform HTTP problem/error envelope.
+- A repository health gate combining `staticcheck`, `govulncheck`, `deadcode` and dependency audits.
+
+## Validation
+
+- `make fmt-check` and `make vet` passed. `make audit` passed too, but it is a git-grep inventory over loc, todo, stubs, docs, tests and gaps (`scripts/audit.sh`), not a static-analysis gate. Do not read it as one.
+- `staticcheck`, `govulncheck`, `deadcode`, tracked-secret and history scans and npm audit were run out of tree. None of the three Go analyzers is installed on this box or wired into any make target, which is the argument for section 4's proposed gate.
+- The advisory version numbers in section 4 could not be re-checked offline on 2026-08-11. `deps/go/go/VERSION` reads `go1.25.5`, built 2025-11-26, so the eight-month gap behind current supports the upgrade claim.
+- The first whole-tree race run failed once in `cmd/mavend` while analyzers were compiling concurrently; a fresh isolated `go test -race -count=1 ./cmd/mavend` passed in 100.6 seconds, so the transient result was not counted as a defect.
+- The pre-existing `deploy/mavwaked.service` modification and untracked `deploy/asoundrc` remained untouched.

docs/language.md

@@ -0,0 +1,74 @@
+# Language: what the model emits, and how Russian is matched
+
+*Last verified: 2026-08-09 @ a9b480a*
+
+Two contracts live here. What a model call is allowed to return, and which
+mechanism is allowed to recognise a Russian word.
+
+## The LLM output contract
+
+All phrasing paths emit `{"response":"...","mood":"..."}`. They fall back to
+plain text when the model skips the JSON.
+
+**One parser, `parseResponseMood` in `internal/phraser/parse.go`.** Every path
+reaches it: the six `LLMPhraser` methods, `PhraseWorld`, and
+`Replier.PhraseReply`. `cmd/mavend/replier_llm.go` wraps the last of those,
+holds the stub fallback, and parses nothing itself. Mood is a fixed enum.
+
+The router prompt is a separate contract:
+
+```text
+[{"intent":<enum>, key?, value?, text?, verb?}, ...]
+```
+
+over 7 intents: `fact, reminder, note, query, act, chat, system`.
+`llm/check_prompt_parity.py` in the training workspace enforces that the Go
+prompt and the relabelling prompt stay identical. They diverged once, and the
+relabelled set then taught a head the Go router never asks for.
+
+## Russian patterns: three mechanisms, no fourth
+
+Hand-written Russian stem patterns were swept out on 2026-08-04 by the owner's
+call. A regex whose output is a fact or a route is the defect. A regex over
+structured input, such as HTML, MIME, JSON, a URL or an argv list, is not.
+
+Before writing a Russian word list, pick one of these.
+
+### `internal/lexicon`, for closed classes
+
+`lexicon_ru_v1.json` holds interrogatives, capture verbs, reminder verbs,
+cardinals, day offsets, parts of day, weekdays, months and spoken hours.
+Editing a word is a data change and there is exactly one copy. Months used to
+live in three files and drifted between them.
+
+Cardinals carry the oblique forms, because a spoken time declines. `в семь` and
+`к семи` are one hour.
+
+### `internal/morph`, for grammar
+
+From the vendored golem Russian dictionary. `IsVerbForm` and `SameWord`.
+
+Lemma matching is BROADER than stem-plus-one-ending. A verb slot that means the
+imperative must be matched exactly. `говори` and `говорил` share a
+lemma and only one of them is a command
+(`cmd/mavend/quiet_toggle.go`).
+
+### `cmd/mavend/topics.go` and the embedder, for open sets
+
+Use these when the question is what a turn is ABOUT. Frozen seeds per subject
+plus a real `other` class, scored against the turn's own query vector.
+
+Same shape as the personal boundary in `personalboundary.go`, with one
+difference. A topic must clear the runner-up by `topicMargin`, because a false
+claim here spends a network scan rather than one honest "не знаю". The old
+keyword tests stay as the offline floor.
+
+### The ecosystem trio
+
+Use it when the answer is not in the utterance at all. Identity is Nexus's,
+never a local pattern.
+
+## Seeds are scoring data
+
+Editing one moves a recogniser. Re-measure against the `TestONNX*` tests rather
+than eyeballing the change.

docs/offload.md

@@ -1,6 +1,6 @@
 # Offloading model work to the workstation
 
-*Last verified: 2026-08-05 @ b789676. Living doc: correct it in place, do not append.*
+*Last verified: 2026-08-09 @ 50c6637. 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.
@@ -105,6 +105,30 @@ how we find out whether the blind spot is real.
 untouched. The model, the context size, the layer count and the MTP flags are the
 owner's business and not this daemon's schema.
 
+**The port carries a bearer token and cannot be loopback** (V-673). homesrv is
+the client, so this hop is on the LAN. Until 2026-08-11 anything on the network
+could spend the card, hold the model resident by touching the idle clock, and
+read `/slots`, which returns other callers' prompts. mavgpud now reads
+`token_file` and refuses to start when the listen address is reachable from the
+network without one. Downgrading to loopback instead would look safe and take
+the model arm down. Maven sends the same token from `workstation.token`, on the
+completion and on the `/health` probe alike. An unsigned probe answers 401,
+which Pair reads as a busy card, so a missing token degrades to the resident
+model rather than breaking a turn. The proxy also
+allowlists the five paths Maven calls, so a leaked token buys the model API and
+not llama-server's admin surface.
+
+**Every GPU service on that box belongs under this supervisor**, added to
+`cmd/mavgpud` rather than to systemd beside it. The rule was learned on
+2026-08-09. The CW2 transcriber ran as its own user unit and registered on the
+KFD like any ROCm job. So the supervisor read its own transcriber as a
+contender. It yielded the card every few seconds and the gemma-4-12b arm was
+down for eight minutes before anyone looked. So the supervisor takes a `stt`
+block and starts CW2 itself. Yielding is all or nothing, because a job that
+wants the card wants all of it. Idle unloading is not. It applies to
+llama-server, which holds 8GB. CW2 holds 1.6GB, and unloading it would cost the
+next voice turn its quality for nothing.
+
 ## What stays on homesrv, permanently
 
 The **embedder** (multilingual-e5-small, ONNX, CPU). It backs the classifier, which
@@ -149,6 +173,29 @@ 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.
 
+Speech-to-text is wired as of 09-08-2026, and it takes only the silent half of the
+rule. A worse transcript is still a turn, so there is nothing to name a gap about
+and `stt.Pair` has no `TranscribeRemote`. `sttSeam` in `cmd/mavend/voicewire.go`
+builds it, beside `modelSeam` and at the same place in `wireVoice`, so the voice
+path and the meeting recorder still share one transcriber.
+
+The remote is not a second endpoint on mavgpud. whisper.cpp cannot load
+CrisperWhisper 2.0 at all. It reads its language count off the vocabulary
+size, and CW2's 51897 tokens shift seven special token ids. So CW2 runs under
+transformers as its own service on port 8081, and `stt.HTTPTranscriber` is the
+second transport for the same seam. It posts raw PCM with the format in headers.
+It carries a bearer token, because audio is the most sensitive thing that
+crosses here.
+
+It is a second endpoint on nothing, but it is a second **child** of mavgpud, and
+that part is not optional. See the supervisor section above for why: a ROCm
+service the supervisor does not own is a contender it yields to.
+
+The margin is the reason: CW2 turbo scores 10.4% WER in Russian against 27.5% for
+the `ggml-small.bin` mavsttd loads, over 200 Golos clips
+(`docs/evals/2026-08-09-crisperwhisper2-russian-wer.md`). Text-to-speech has not
+moved and piper on homesrv is still the only synthesizer.
+
 Speech-to-text stays two stages when it moves. One call carrying both a clip and the router
 prompt was measured on 05-08-2026. It scores 54.2% intent-only against 84.7% for whisper on
 homesrv, on the same 72 cases. The model transcribes clips it then routes wrong, so a long
@@ -173,8 +220,9 @@ cleaner transcripts, not accuracy. See `docs/evals/2026-08-05-audio-in-routing.m
    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.
-3. **Speech-to-text and text-to-speech** (#486). They gain a real margin, but on
-   quality alone, and both already work.
+3. **Speech-to-text and text-to-speech** (#486). Speech-to-text is wired, see
+   above. Text-to-speech is not, and piper is good enough that nothing argues
+   for moving it yet.
 4. **The wake word** (#487). Independent of all of the above.
 
 ## Assumptions

docs/routing.md

@@ -0,0 +1,474 @@
+# Routing
+
+*Last verified: 2026-08-11 @ 25ed201*
+
+How an utterance becomes a `Decision`, why each stage exists, and what every
+stage has measured. `CLAUDE.md` carries the rules an agent must not break. This
+file carries the reasoning and the history behind them.
+
+Two decisions come out of a route. **Intent** is one of seven values. **Source**
+is where the answer lives, and it is read on `IntentQuery` alone. They are scored
+separately, because one number hides which one moved.
+
+## The cascade
+
+| Stage | What it is | Where |
+|---|---|---|
+| 0 | Deterministic grammars over the utterance | `stage0.go`, `praxis.go`, `worldquery.go` |
+| 0b | Four ONNX heads on one e5-small forward pass | `heads.go` |
+| 1 | The resident model, GBNF-constrained JSON | `llmrouter.go` |
+| 2 | Nearest neighbour over frozen seed phrases | `classifier.go`, `embedder.go` |
+
+Every stage may decline, and the next one answers. Any error at stage 0b or 1
+falls through, so a turn never breaks on a model.
+
+The resident model arm is wired at `voice.go:214` through
+`pickLLMRouter(cfg.Voice.UseLLMRouter(), llmClient)`. The flag is
+`voice.llm_router` in `config.go`, `DefaultLLMRouter` is on, and
+`deploy/mavend.json` sets it `true`. With no llama-server to talk to,
+`pickLLMRouter` logs that and degrades to the classifier.
+
+The classifier is the floor and not dead code. It runs when the resident model
+is off, when there is no llama-server to talk to, and on any per-turn error.
+Routing by seed similarity is the known cause of weak Russian queries. Deleting
+it would make a model outage a broken turn.
+
+### Why there are two engines at all
+
+The original design was the classifier alone. `docs/rearchitecture.md` replaced
+it with a model that emits structured JSON. The same weights phrase the reply.
+That demoted the embedder from a routing gate to a hint for recall. The model
+became the default on 2026-07-31.
+
+The gap it buys is smaller than the design assumed. Measured 2026-08-02 on the
+77-case Russian fixture, the classifier scores 68.8% full accuracy at p50 16.6µs.
+Qwen3-1.7B scores 72.7% through the cascade. Four points, not a doubling.
+
+An older figure of 36.8% for the classifier stood in `CLAUDE.md` until then. It
+predates the stage 0 rules and the seed additions. Both now score inside the
+classifier baseline.
+
+Latency was misreported the same way. A figure of 2.7 seconds stood for two
+days and was contention rather than the model.
+`docs/evals/2026-07-31-routing.md` line 61 measures the router at p50 825ms and
+the cascade at p50 0.80s to 1.04s.
+
+## Numbers
+
+Three arms answer, so three numbers are live. Judge a routing change against the
+classifier and the resident model, since those are what always answer.
+
+| Arm | Intent | Destination | p50 | Measured |
+|---|---|---|---|---|
+| classifier + ONNX | 76.0% (73/96) | 36.4% (12/33) | 16.6µs | 2026-08-08 |
+| resident Qwen3-1.7B, cascade | 80.2% | not measured | 1.19s | 2026-08-05 |
+| routing heads, cascade | 96.9% | 75.8% | 27.9ms | 2026-08-08 |
+| gemma-4-12b, cascade | 84.4% | 72.7% | 329ms | 2026-08-02 |
+| gemma-4-E4B, cascade | 89.6% | 57.6% (19/33) | 294ms | 2026-08-09 |
+
+The fixture grew from 77 cases to 91 to 96. So a number is comparable only to
+another number on the same fixture. Sources:
+`docs/evals/2026-08-05-routing-resident-model.md`,
+`docs/evals/2026-08-02-workstation-gemma4-12b.md`,
+`docs/evals/2026-08-09-e4b-vs-12b-routing.md`,
+`docs/evals/2026-08-08-routing-heads-in-go.md`,
+`docs/evals/2026-08-08-destination-fixture.md`.
+
+The resident model alone scores 37.4% full against 61.5% intent-only. The gap is
+slots and not routing. It routes `reminder` and leaves the time to the daemon,
+which is what the contract asks.
+
+To re-run the resident model as router, start a **second** llama-server on a
+fixed host port. The resident one binds `--port 0` inside the container and no
+host process can reach it.
+
+### The workstation is not the better router any more
+
+It was, from 2026-08-02 until the heads landed. gemma-4-12b beat everything on
+the box at 84.4% intent and 72.7% destination. The heads beat it on both at a
+twelfth of the latency. The workstation stays the better phraser.
+
+E4B replaced the 12B on 2026-08-09 by the owner's call. It is a step down on
+routing. Against a same-session 12B control it costs four destination cases and
+buys 50ms. Read destination as the finding. It names nothing where the 12B names
+`recall` or `calendar`, which is safe but walks the whole chain. It has no MTP
+and cannot be given any here. The only `gemma4-assistant` draft on disk is
+trained against the 12B's hidden states.
+
+## Stage 0: what the grammars claim, and why
+
+A rule at this stage is a claim. Either the model gets this wrong, or it wastes a
+second getting it right. Every rule was added against a measurement.
+
+- **Agenda questions** (`AgendaQueryGrammars`, 2026-08-01). "что у меня сегодня",
+  "во сколько у меня встреча" and anything naming a calendar go to `IntentQuery`.
+  They were going to `IntentSystem`, where `replySystem` has no agenda arm and
+  answered "пока не умею". Worth 2.6 points of full accuracy and calendar 0/2 to
+  2/2.
+- **Rest of day and narrative** (V-498, 2026-08-04). `rest-of-day-query` claims
+  "что дальше?". `NarrativeQueryGrammar` claims "расскажи про X", "объясни X" and
+  "опиши X". Neither carries a question mark or an interrogative, so the model
+  called both `IntentFact`. `IsQuestionShaped` caught the write downstream, so
+  this was a latency and fixture defect rather than a correctness one. The
+  narrative rule declines `chatNarrativeTopics`, because the query chain has no
+  source that answers a joke or a bedtime story.
+- **Praxis** (V-516, 2026-08-05). `PraxisGrammars()` fills `Slots.Fn` with a
+  capability name. These grammars are the **only** path to Praxis and not a
+  faster one. The model reaches Praxis 0/12 alone, the same as the classifier.
+  Nothing in the router prompt names a Praxis capability, so there is no string
+  for it to write. Through the cascade it is 11/12. Measured overall 16/30 to
+  27/30, lifecycle 0/5 to 5/5
+  (`docs/evals/2026-08-05-praxis-reach.md`,
+  `docs/evals/2026-08-05-reach-llm-router.md`).
+  `handlePraxisAct` compares `Slots.Fn` to a capability alias. Otherwise that
+  slot is filled from the deployment's enabled tool names, and no Praxis alias
+  is on that list.
+- **World questions** (`WorldQueryGrammars`, V-655, 2026-08-07). "что такое X"
+  and "сколько будет 17 на 23". Wired after the agenda rules and **before** the
+  feed and list rules. "что такое лента" is a definition question, and the feed
+  rule would take it on the noun alone.
+
+`calendar-query` and `event-time-query` name the calendar as the destination.
+The possessive agenda rules deliberately do not. "что у меня в списке покупок"
+matches `agenda-query`, and naming the calendar there would take the list source
+off the turn. That caution now costs four destination cases. See the model arm
+below.
+
+Go's `\b` is ASCII-only and never fires after a Cyrillic letter. A pattern needs
+an explicit `(\s|[?!.]|$)`.
+
+The stage 0 set lives in `router.StageZeroGrammars` (`internal/router/stagezero.go`).
+Both `buildRouter` and the eval fixture call it. The daemon and the measurement
+cannot disagree about which rules exist, or in what order.
+
+It was two lists until V-693 and it drifted twice. V-655 wired
+`WorldQueryGrammars` into the daemon and not into the fixture. That cost 3 points
+of destination and V-659 fixed it. `BareCaptureGrammar` then did the same thing,
+from V-557 until V-693 found it. That one moved no number, which is the point:
+the fixture had been scoring a set nobody ran and nothing said so.
+
+### Praxis lifecycle rules
+
+A **stative** lifecycle word ("готово", "принято") needs an item named beside it.
+A bare **imperative** ("закрывай") may ask which one. It also requires a sentence
+naming no object of its own. Otherwise "закрой шторы в комнате" goes to Praxis
+instead of the house. A demonstrative ("отметь это как сделанное") resolves
+against `h.surfacedItems` only when exactly one item was spoken. Otherwise the
+turn goes back to the cascade rather than transitioning the wrong item.
+
+### Slots on a stage 0 decision
+
+`fillMatchedSlots` runs the stage 2 extractor over whatever a grammar built
+(V-572, 2026-08-06). It fills only the slots the grammar left empty. A matched
+value always wins, because the rule read a literal pattern and the extractor
+guesses.
+
+It did not run before. So `ReminderGrammar` handed the daemon `HasTime: false`
+for "напомни в 11:00 позвонить маме", and `missingFor` read the silence as
+absence and asked "Когда?". It is inert for every grammar but the reminder:
+`Extract` fills Time, Fn and Key and nothing else. A stage 0 query costs 3.7µs
+against 3.9µs before, benchmarked at 20000x.
+
+`Slots.Text` is deliberately not filled. A grammar that left it empty meant it,
+and `agendaQueryBuild` hands the query chain the utterance itself.
+
+## Stage 0b: the routing heads
+
+Routing has a bounded output space, so it is classification rather than
+generation (owner's call, V-546,
+`docs/plans/18-routing-heads-on-e5-small.md`). The 118M multilingual-e5-small is
+already resident. A softmax cannot emit a value that does not exist, so no
+grammar is needed. Max softmax is a calibratable confidence, where
+`Confidence: 1.0` was a hardcode. Training costs roughly 5e15 FLOPs, so 10 to 30
+minutes on the workstation. A 100M decoder from scratch is 10 to 20 GPU hours.
+
+**Fine-tune a copy of the weights.** The resident embedder backs memory recall.
+Training it in place couples routing accuracy to recall@1, with nothing in the
+suite to name the trade.
+
+Four heads share one masked mean pool, trained over three days. The measurements
+are `docs/evals/2026-08-08-routing-heads-two-head.md`,
+`docs/evals/2026-08-08-slot-head-three-head.md`,
+`docs/evals/2026-08-08-clarify-head-four-head.md` and
+`docs/evals/2026-08-08-massive-warm-start.md`.
+
+| Head | Score | Notes |
+|---|---|---|
+| intent | 92.8% mean over 3 seeds | fixture is the 88 cases carrying an intent |
+| destination | 80.8% mean, best 29/33 | beats the 12B teacher it was distilled from |
+| slot BIO tags | 72.4% span F1 | still climbing when epoch selection stops it |
+| clarify | catches 7.0 of 8, 2.3 false of 88 | parity with the cascade, no rules in front |
+
+Read the best destination run as one seed and not a headline. One case is 3
+points on a fixture this small. Head intent accuracy is **not** comparable to the
+cascade's 76.0% and 84.4%. A softmax has no clarify class, so the head's fixture
+is 88 cases and not 96.
+
+Recall is 15/15 and world is 5/5.
+
+**Mood is cut, not deferred.** The enum describes her own reply state, not the
+speaker's emotion, and no dataset maps onto it.
+
+### The clarify head
+
+Clarify is not a value of intent, so a softmax cannot emit it. It is a second
+question over the same pooled vector: can Maven act on this at all. Accuracy is
+the wrong number here and a head that never asks scores 91.7%.
+
+Confidence is the other half. Max softmax over the intent head reads 0.851 where
+it is right and 0.604 where it is wrong. It ranks right above wrong in 83.4% of
+pairs.
+
+It is not free the way the slot head was. Intent, destination and slot F1 each
+move down one to four points, inside the seed spread. `поужинал` is a false
+clarify on every seed. That is the same defect `thinSingleToken` was narrowed for
+on 2026-08-01.
+
+The corpus is generated, because every existing row is answerable by
+construction. The router-prompt agreement filter cannot work here. `routeGrammar`
+has no clarify value, and a generated line always agrees with itself. A gemma
+judge replaces it. The first judge called 24 of 40 answerable rows underspecified.
+It judged against a generic assistant rather than against Maven's contract.
+
+### The slot head
+
+BIO slot tags had no Maven-domain corpus. That was true of found corpora and
+false of made ones. `label_slots.py` distils spans out of gemma-4-12b under a
+GBNF closed over Maven's own five slots. A span survives only when it is a
+literal substring of the utterance, so the agreement filter costs no second call.
+2178 spans over 1702 rows, 37 dropped, nothing unparsed.
+
+Epoch selection reads the intent dev slice alone. That costs the slot head about
+4 points.
+
+### Warm start and the floor
+
+The MASSIVE warm-start of step 2 is worth nothing here. Stock e5-small ties it on
+intent and leads by a third of a case on destination. Nothing argues for keeping
+that step.
+
+The floor was a corpus defect and it is fixed. The first 120 floor rows carried
+one sentence shape, so the head named a destination where the fixture says walk
+the chain. Rotating six shapes took the floor 3/7 to 6/7 and destination 75.8% to
+80.8%.
+
+What is left is calendar at 3/6 on every seed, which training cannot move. The
+possessive agenda rules claim those cases at stage 0 and name nothing, so no
+label reaches the head.
+
+### Reading them in Go
+
+`RouterHeads` in `internal/router/heads.go` loads `router_heads.onnx` (V-664,
+2026-08-08). It reads intent, destination and clarify off one forward pass.
+
+Three rules around it, each measured:
+
+- The **clarify head decides first**, before the intent threshold. It answers a
+  different question. A thin utterance scores low intent by construction, so
+  gating it cost 6 of 8 ambiguous cases.
+- The **destination head is read on `IntentQuery` only**, since no other intent
+  reaches `queryWalk`.
+- `headsThreshold` is 0.6, the measured knee. Every value up to 0.85 drops right
+  answers and keeps the same two wrong ones.
+
+`voice.embedder.heads_path` is the whole switch. Empty, missing or unloadable
+means the heads are nil. The cascade is then byte-for-byte what shipped before
+them.
+
+Pointing it at `model_path` is refused at config load (V-692). An unloadable
+weights file is not fatal, because the heads are an accelerator. A working file
+in the wrong role is a different thing. The heads then score with the graph the
+resident embedder scored with, and recall degrades with no log line. The check
+cleans and absolutises both paths, then compares them with `os.SameFile`, so a
+symlinked copy is caught too.
+
+### The tokenizer bug the heads found
+
+`encodeWord` in `onnxembedder.go` read every long word backwards until 2026-08-08.
+It cost recall@1 7.4 points and recall@3 11.1. Nothing caught it, because seeds
+and queries were mangled the same way and cosine survived. The heads found it.
+They are trained through transformers and read through this.
+
+The embedder id now carries a tokenizer revision (`@384/tok2`). So fixing the
+tokenizer triggers `ReembedAll` the way swapping the model file does. Bump
+`tokenizerRev` on any change to what it emits.
+
+## Clarify
+
+`Confidence: 1.0` was hardcoded in `llmrouter.go`. So the model path could never
+ask for clarification, and it missed 6 of 6 refusal cases (V-359). The bug had a
+second half. The model branch never consulted `r.threshold` at all, so a correct
+low confidence would have been discarded anyway.
+
+Fixed 2026-07-31 with structural signal feeding the same stage 3 gate the
+classifier path already had (`gateLLMDecision` in `router.go`). Three signals: a
+single-token utterance, a keyless fact, an act with no allowlisted fn.
+
+Re-measured: missed clarify 6/6 to 1, at the cost of 3 false clarifies and 2.6
+points of full accuracy. Two of the three false clarifies are acts the model
+mis-routed and the gate caught. Asking beats wrongly executing, so the fixture
+and the daemon disagree about what is correct there.
+
+The third, `поужинал`, was a real defect. The single-token rule was an English
+intuition. It does not transfer to Russian, where one word is routinely a whole
+sentence.
+
+Narrowed 2026-08-01. `thinSingleToken` (`internal/router/singletoken.go`) still
+thins a bare one-word nominal. It spares two classes. One is a closed lexicon of
+social and control singles ("привет", "стоп", "yes"). The other is any token
+carrying a Russian verb ending, because a verb already contains its subject. Both
+tests are offline and cost nothing. False clarifies 3 to 2, intent-only 74.0% to
+75.3%.
+
+The two remaining false clarifies are the act-with-no-allowlisted-fn arm of the
+gate, not this rule.
+
+## The destination
+
+`query` was a shrug. The cascade sorted an utterance into one of seven intents,
+then `IntentQuery` handed the turn to `querySources` in the daemon. That is
+twenty-two branches deciding by seed similarity in a fixed order. It had no
+fixture, no accuracy number, no model arm and no floor.
+
+`Decision.Source` (`internal/router/source.go`) is the second half of the route
+(V-655, 2026-08-07). Twelve destinations, not twenty-two. The three recall passes
+plus `fact-by-key` are one destination from outside. So are search, Kiwix and the
+URL reader.
+
+`queryWalk` in `cmd/mavend/actions_query.go` takes sources **out** and moves none.
+That is the safety argument. The table's order is load-bearing. Every comment on
+it argues a reason between two sources. Above all it carries "the owner's data
+first, then the world". Naming `SourceWorld` does not send the turn outside on its
+own.
+
+What comes out is only the sources that **guess**. Those decide a turn is theirs
+by cosine against frozen seeds, then answer whatever they claimed. They hold no
+table that could come back empty. Weather is the pure case and has no local data
+at all. It was measured on the box 2026-08-07
+(`docs/evals/2026-08-07-week-of-usage.md` section 4). It answered both "что такое
+TCP?" and "сколько будет 17 на 23?" with "для какого города?". The feed answered
+"какой у меня любимый язык?" with kernel headlines.
+
+### Who may drop the personal boundary
+
+The personal boundary guesses, so naming `SourceWorld` drops it. That is what
+stops it answering "кто такой Линус Торвальдс?" with "не нашла у тебя такой
+записи", which it did on 2026-08-07.
+
+Three deciders name a destination and two of them infer it: the heads and the
+resident model. An inferred `SourceWorld` on a question about him would reach
+SearXNG. That widens what is asked rather than costing a local answer. So only a
+stage 0 grammar may drop it (owner's call, V-666, 2026-08-09).
+
+`Decision.SourceAnchored` carries the provenance. It is a field and not
+`Stage == 0`. Stage 0 also means confidence 1.0 and an anchored claim band, and
+one of those could stop implying the others. `definitionQueryPattern` claims "кто
+такой X", so the 2026-08-07 case is still anchored and still answered.
+
+`queryWalk` reads `SourceAnchored` for the query source marked `boundary: true`
+and no other. Every other guesser still comes off the turn, whoever named the
+destination. `TestOnlyAGrammarMayDropTheBoundary` and
+`TestNamingRecallKeepsTheBoundary` pin both directions.
+
+### The destination fixture
+
+`want_source` on `eval.Case` is a pointer, because the destination has three
+states and a bare string has two. Absent is every intent but query. Present and
+empty is the `SourceUnknown` contract: name nothing and walk the chain. Present
+and named is a destination the route must produce. Thirty-three of ninety-six
+cases carry one.
+
+A destination miss does **not** fail the case. It lands in `Outcome.SourceReason`
+and never in `Reasons`, so `Accuracy` and `IntentAccuracy` mean what they meant.
+`SourceAccuracy` is a second number over the labelled cases alone.
+A route that lost its intent scores no destination hit. Otherwise a clarify would
+satisfy an empty label for free.
+
+Seven cases assert the floor and five of them are homelab operations. They cluster
+because `SourceRecall`, `SourceNetwork` and `SourceAttention` overlap on every
+question about the box. `mavpoll` writes its netdata and uptime-kuma observations
+into the fact store recall reads. That is a finding about the enum, not a gap in
+the labelling. The other two are `ru-query-005` and `ru-query-014`. No query
+source reads the reminder store, and a deadline could sit in tasks, the calendar
+or Praxis. The owner confirmed all seven floor labels on 2026-08-08.
+
+### The model arm
+
+`routeGrammar` carries a `source` rule closed over `router.Sources` plus the
+empty floor (V-660, 2026-08-08). So the model cannot emit a destination that does
+not exist. The prompt lists the twelve in Russian and says `""` is a normal answer
+to give often. `LLMRouter.Route` reads it back through `ValidSource` and on
+`IntentQuery` alone.
+
+Against gemma-4-12b the cascade scores destination 24/33 with intent unmoved, and
+recall goes 0/15 to 14/15.
+
+**Stage 0 now costs four destination points.** It did not before. The four cases
+the cascade loses and the model alone wins are all calendar. The possessive agenda
+rules claim them first and name nothing on purpose. That caution was free while
+nothing downstream could name anything either. It is not free now, and the fix is
+the owner's call (V-660 open).
+
+## The decision trace
+
+Arbitration between the claimants on the utterance stream is order. It is
+hardcoded in the pre-route resolver ladder, in `buildRouter` and in
+`querySources`. Nothing recorded who lost until V-564.
+
+`internal/decision` records one `Record` per turn. It holds every claimant, what
+it would have made the turn, the score it reported, and how it ended. A claimant
+won, declined, lost on score, was thinned by a gate or was **never asked**.
+
+The record rides the context, the same seam `querysource.go` uses. So a claim
+site cannot change a route, and a context with no record costs nothing. It is
+installed in `runTurn`, so the mic, telegram and the web leave the same trail.
+
+Adding a rung to the ladder in `runTurn` means adding its name to `preRouteLadder`
+in `cmd/mavend/decisiontrace.go`. Otherwise that rung is silently missing from the
+record.
+
+### Why it persists now
+
+The original rule was that nothing persists, because a turn record is read minutes
+later or never. Storage was a 25-turn in-memory ring read over `ipc.TurnDecisions`
+and rendered on `/trace`.
+
+The owner reversed it on 2026-08-06 (V-629,
+`docs/plans/21-persisting-the-routing-trace.md`). The routing heads cannot be
+fitted or calibrated without real utterances. And 9 of the 31 modes in
+`internal/modes` have no seed example at all.
+
+The ring did not move. `cmd/mavend/routingtrace.go` is a second sink beside it,
+writing `routing_traces` (migration #23). The utterance is stored in clear. A
+384-dimension vector of a short sentence is substantially recoverable, so storing
+vectors instead would be a privacy claim we cannot support. What makes it safe is
+the same thing that makes the fact store safe. Retention is 14 days, enforced on
+write and again on start, so a box that goes quiet does not keep every row.
+Nothing reads it outward. `Store.Wipe` deletes it with everything else.
+
+### Corrections
+
+A correction is promoted out into a seed-shaped row in `routing_labels`
+(migration #24) and kept, because a label is not a transcript. The transcript
+still expires.
+
+A turn marked wrong with no target is a usable negative, so naming the intent is
+never required. The target is one of the seven intents and never free text.
+
+All three reaches offer it as of 2026-08-06:
+
+- `/chat` offers two buttons beside the reply, over `ipc.CorrectTurn` and the
+  trace id that rides back on `ipc.ChatReply`.
+- Voice offers the `repair` rung, which has read spoken corrections since V-455.
+  It now writes the durable label beside the classifier seed it always wrote. A
+  spoken negative with no target is its own rung, `repair-negative` (V-636,
+  `docs/plans/22-correcting-a-turn.md`).
+- Telegram offers an inline keyboard under the reply. It needed the chat to become
+  readable first (V-637, `docs/plans/23-inbound-telegram.md`). The poller is dark
+  unless the `telegram` block says `intake`. It long-polls, because the box takes
+  no inbound connections. It accepts `chat_id` and no other sender, and it drops
+  whatever queued while the daemon was down. It reaches the daemon through
+  `ipc.CoreAPI` alone.
+
+The turn source is still `tap:text` for both telegram and the web. So provenance
+cannot tell a chat turn from a typed one.

docs/workflow.md

@@ -0,0 +1,117 @@
+# Session workflow: the five stores and the guards
+
+*Last verified: 2026-08-11 @ 557f5a3*
+
+How a session starts, where each kind of writing belongs, and what the hooks
+refuse. `CLAUDE.md` carries the commands. This file carries the reasoning.
+
+## Five stores
+
+Each owns something the others must not hold.
+
+| Store | Holds | Lifetime |
+|---|---|---|
+| Vikunja task | goal, constraints, assumption ledger, status | durable |
+| `CLAUDE.md`, `AGENTS.md` | what an agent must know before touching code | durable |
+| `docs/` | design, measurements, decisions | durable |
+| `TASK.md` | the brief for this branch, written by `task start`, immutable | one branch |
+| `HANDOFF.md` | only what the next agent needs to resume | one session |
+
+`TASK.md` and `.task/` are excluded through `.git/info/exclude`. `HANDOFF.md` is
+gitignored and injected at session start. If a line in the handoff would still
+matter next week, it is in the wrong file.
+
+## Doc tiers
+
+Tiered by path, so staleness is visible from the filename.
+
+- Files directly under `docs/` are living. They carry a
+  `Last verified: <date> @ <sha>` line and are corrected in place.
+- Files under `docs/evals/` are dated measurements and are never edited after
+  the day. A newer number is a new file, not an edit.
+- Files under `docs/archive/` are dead and read by nobody by default.
+
+## Vikunja
+
+This repo is project **Maven** (ID 2). MCP at `http://localhost:9100/mcp`, or
+`http://192.168.1.104:9100/mcp` from workpc. Feature, bug and deploy tasks go
+there.
+
+A task holds the goal, the constraints and the assumption ledger. A session
+without a task id cannot be resumed by anyone, so a session with none asks for
+one first.
+
+**Close a finished task with `done: true` and nothing else** (owner's call,
+2026-08-07). Do not write a completion summary into the description on the way
+out. It is lost anyway, and the durable record is the commit messages and the
+merged PR. `update_task` carrying a `description` resets `done` to false, which
+is why a write-up ever took two calls.
+
+## The branch tool
+
+`~/.local/bin/task` owns the branch, the commit identity and the PR. One task,
+one session, one PR.
+
+```sh
+task start <vikunja-id>   # branch off origin/master, write TASK.md, fetch review comments
+task pr                   # push, open or refresh the PR, label Vikunja, notify
+task comments             # re-pull this branch's review comments into .task/
+```
+
+`/pickup` opens a session and `/wrap` closes it. Wrap at roughly half context
+rather than letting the session compact.
+
+## Guards
+
+Two hooks in `.githooks/`, tracked, wired with `core.hooksPath`. A fresh clone
+needs `git config core.hooksPath .githooks`.
+
+- `pre-commit` refuses master, and refuses more than 300 changed lines in
+  non-markdown files. Markdown is exempt and may land as one batch.
+- `commit-msg` requires the subject to end with `(V-<id>)`. `V-` and not `#`,
+  because Gitea autolinks `#123` to a Gitea issue, which is the wrong tracker.
+
+Two more guards live outside the repo, in `~/.claude/hooks/`. `diff-budget.sh`
+blocks further edits past 600 changed lines on a `task/` branch.
+`prose_lint_hook.py` checks prose on every write. Both measure against
+`origin/master`, so a local master that is ahead of the remote makes the diff
+budget read high.
+
+`--no-verify` exists. Using it means saying why in the commit body.
+
+## Static gates
+
+Three analyzers, one target each, and `make analyze` for all three. The
+2026-08-10 audit asked for them because none was installed on the box and
+`make audit` is a git-grep inventory, not analysis. Do not read `make audit` as
+a gate.
+
+- `make vuln`, govulncheck over `./...` (V-682).
+- `make lint`, staticcheck over `./...` (V-694).
+- `make deadcode`, deadcode with `-test` over `./...` (V-694).
+
+None of the three joins `make test`. All three install over the network, and
+`test` has to pass on a box with no route out. `vuln` reads the advisory
+database at run time as well. Run `make analyze` before a dependency or
+toolchain bump lands, and before calling a symbol unreachable.
+
+Each tool is pinned in the Makefile beside `GO_VERSION`. A gate that moves on
+its own is not a gate. Each installs into `deps/bin`, because a tool is not a
+dependency of the module.
+
+**staticcheck and deadcode pass against a baseline, not against zero.** The
+accepted findings live in `scripts/analyzers/*.baseline`, one line each. A key
+holds file, check id and message, never a line number. A line number goes stale
+on the next edit above it. The output then reports moved findings as new ones,
+and the reader learns to skip it.
+
+`scripts/analyzer-gate.sh` gives the verdict. A finding absent from the baseline
+fails. So does a baseline entry whose finding is gone, which is what stops the
+accepted set from outliving the repo. Deleting the entry is part of each fix.
+
+`deadcode` runs with `-test` because a test is a caller. Without the flag the
+report is 172 lines, most of `internal/router/eval`, none of it a mistake.
+
+A baseline entry carries the reason it stays. Three reasons appear. Another task
+owns the finding. The check cannot see through a false positive. A cosmetic
+finding waits for a sweep.

docs/world.md

@@ -0,0 +1,71 @@
+# The world chain
+
+*Last verified: 2026-08-09 @ a9b480a*
+
+What happens when the answer is not his. `CLAUDE.md` carries the boundary rule.
+This file carries the mechanism and the measurements behind it.
+
+## What replaced "never phones home"
+
+That promise was deprecated on 2026-07-31 by the owner's call. A 1.7B does not
+know enough to answer world questions, so she reads external sources. Four rules
+replaced it:
+
+- **No telemetry, no cloud model, no third-party account.** That part never
+  changes. Nothing about Maven is reported to anyone and inference stays on the
+  box.
+- **The owner's data first, then the world.** Every source reading his facts,
+  notes, calendar, tasks or house runs before anything outside. The personal
+  boundary sits between them. Reading beats recalling for a small model.
+- **The owner's notes and facts are never search input.** Only the utterance
+  goes out. Never the persona block, the history, or matched notes.
+- **External search is allowed and off unless configured**, like weather and
+  telegram. The code default is off. `deploy/mavend.json` ships a `search`
+  block, so it is on for this box and deleting the block turns it off again.
+
+**Live search leads and the ZIMs are the fallback** (owner's call, 2026-08-02).
+A self-hosted SearXNG answers first. The Kiwix ZIMs on homesrv answer when the
+search is empty, unreachable, or the line is down.
+
+## The gate is emptiness and nothing else
+
+`Response.Empty()` is the whole gate. There is no quality threshold in front of
+it. Four signals were tried and none separates a real question from an invented
+one.
+
+Token overlap was the closest and it still fails. "столица Франции" would lose
+its answer, because the answer is Париж and that word is not in the question
+(`docs/evals/2026-08-05-search-quality-signals.md`).
+
+**The embedder is not a fifth signal.** Query-to-passage cosine measures topic
+and not whether the passage answers. The two sets overlap
+(`docs/evals/2026-08-09-kiwix-topic-retrieval.md`).
+
+## Timeouts
+
+The connect phase alone is capped at `dialTimeout`, 1.5s, because a blackholed
+host once cost the owner 8 seconds. A slow instance that did connect keeps the
+full 8 (`docs/evals/2026-08-05-kiwix-offline-fallback.md`).
+
+## Kiwix
+
+**A Russian question reads `wikipedia_ru_all_maxi_2026-02` verbatim** through
+`kiwix.book_ru`. The RU→EN rewriter is the workaround for an English book and is
+skipped there. Kiwix catalog names come from the filename, not the `<name>`
+field.
+
+**Kiwix ranks by keyword overlap.** Never send it a whole sentence. `kiwix.Topic`
+drops the narrative request, the interrogative and a verb behind one.
+
+`kiwix.TitlePath` tries the exact article first, since a ZIM is addressable by
+title and a wrong title is a 404. `TitleCandidates` tries the spoken form and
+then the capitalized one. Both apply on the verbatim path alone. The rewriter
+already reduces a question, and reducing twice takes the topic off its input
+(V-668).
+
+## Which source answered
+
+**The claiming query source is readable on `/chat`** as a badge beside the
+reply. It is carried on `ipc.ChatReply.Source` and noted by `noteQuerySource` in
+`cmd/mavend/querysource.go`. It rides the context, so `handleText` keeps the one
+string signature the mic, telegram and the web share.

go.mod

@@ -1,6 +1,6 @@
 module github.com/kami/maven
 
-go 1.25.5
+go 1.25.12
 
 require (
 	github.com/coder/websocket v1.8.12
@@ -22,7 +22,7 @@ require (
 	github.com/mattn/go-isatty v0.0.20 // indirect
 	github.com/ncruces/go-strftime v1.0.0 // indirect
 	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
-	golang.org/x/text v0.14.0
+	golang.org/x/text v0.40.0
 	modernc.org/libc v1.74.1 // indirect
 	modernc.org/mathutil v1.7.1 // indirect
 	modernc.org/memory v1.11.0 // indirect

go.sum

@@ -25,13 +25,13 @@ github.com/yalue/onnxruntime_go v1.31.0 h1:1ln4YW1SFOFfGJZXe3jNOb2JUSt+l2pEneZfV
 github.com/yalue/onnxruntime_go v1.31.0/go.mod h1:b4X26A8pekNb1ACJ58wAXgNKeUCGEAQ9dmACut9Sm/4=
 golang.org/x/mod v0.37.0 h1:vF1DjpVEshcIqoEaauuHebaLk1O1forxjxBaVn884JQ=
 golang.org/x/mod v0.37.0/go.mod h1:m8S8VeM9r4dzDwjrKO0a1sZP3YjeMamRRlD+fmR2Q/0=
-golang.org/x/sync v0.21.0 h1:HLII4xRRTtCRkxYp4HNFF0Js/Og6q2i++KXbg0gHCwM=
-golang.org/x/sync v0.21.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
+golang.org/x/sync v0.22.0 h1:SZjpbeLmrCk4xhRSZFNZW5gFUeCeFgjekvI/+gfScek=
+golang.org/x/sync v0.22.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.46.0 h1:noSf2Fq6F8DBgS+LysIkx7rIExoNHJsxOAtPp4rthXw=
 golang.org/x/sys v0.46.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
-golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ=
-golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
+golang.org/x/text v0.40.0 h1:Ub2Z6/xjgF1WrYQz2nuITOEegKFtiIy+rieRJ5lHZKs=
+golang.org/x/text v0.40.0/go.mod h1:hpnzDAfGV753zIKo+wk3u1bVKCGPbrnF7+7LBF/UHVY=
 golang.org/x/tools v0.47.0 h1:7Kn5x/d1svx/PzryTsqeoZN4TZwqeH5pGWjefhLi/1Q=
 golang.org/x/tools v0.47.0/go.mod h1:dFHnyTvFWY212G+h7ZY4Vsp/K3U4/7W9TyVaAul8uCA=
 modernc.org/cc/v4 v4.29.0 h1:CXgwL8cvxmyzBQZzbSl/6xFtMCryb6u8IOqDci39cgc=

internal/config/voice.go

@@ -2,6 +2,9 @@ package config
 
 import (
 	"errors"
+	"fmt"
+	"os"
+	"path/filepath"
 	"time"
 )
 
@@ -37,6 +40,16 @@ type EmbedderConfig struct {
 	ModelPath     string `json:"model_path,omitempty"`
 	TokenizerPath string `json:"tokenizer_path,omitempty"`
 	LibPath       string `json:"lib_path,omitempty"`
+
+	// HeadsPath — the routing heads graph, which is a fine-tuned COPY of the
+	// model above with four linear heads on its pooled output (V-664). Empty
+	// means no heads, and the cascade runs exactly as it did before they
+	// existed. It shares LibPath and TokenizerPath, and router_heads.json is
+	// read from the same directory.
+	//
+	// It must never be pointed at ModelPath. Memory recall depends on the
+	// resident copy scoring what it scored, and the fine-tuned one does not.
+	HeadsPath string `json:"heads_path,omitempty"`
 }
 
 // WeatherConfig configures the weather provider for voice queries.
@@ -131,10 +144,55 @@ func (c *Config) validateVoice() error {
 		if e.ModelPath == "" || e.TokenizerPath == "" || e.LibPath == "" {
 			return errors.New("voice.embedder: all three of model_path, tokenizer_path, lib_path must be set, or remove embedder to use the floor stub")
 		}
+		if err := e.checkHeadsDistinct(); err != nil {
+			return err
+		}
 	}
 	return nil
 }
 
+// checkHeadsDistinct refuses a heads graph that is the embedder's own file
+// (V-692). The rule is stated on HeadsPath above and in CLAUDE.md, and until
+// now nothing enforced it: the daemon loaded whatever the key pointed at, so
+// pointing both keys at one file cost recall with no error and no log line. It
+// reads as ordinary drift, which is the worst kind of misconfiguration.
+//
+// Refusing to start is the right trade here. The heads are an accelerator and a
+// broken weights file is deliberately not fatal in voicewire.go, but this is not
+// a broken file. It is a working file in the wrong role, and a daemon that
+// cannot route well should say so rather than answer worse.
+//
+// Cleaned and made absolute first, so "./m.onnx" and "$PWD/m.onnx" are one
+// path. Then SameFile, which catches the copy that is a symlink or a hard link
+// to the original. A path that does not stat is left to the loader, which fails
+// on it with a better message than this can give.
+func (e *EmbedderConfig) checkHeadsDistinct() error {
+	if e.HeadsPath == "" || e.ModelPath == "" {
+		return nil
+	}
+	heads, model := absClean(e.HeadsPath), absClean(e.ModelPath)
+	same := heads == model
+	if !same {
+		hi, herr := os.Stat(heads)
+		mi, merr := os.Stat(model)
+		same = herr == nil && merr == nil && os.SameFile(hi, mi)
+	}
+	if same {
+		return fmt.Errorf("voice.embedder: heads_path and model_path are the same file (%s) — the heads graph is a fine-tuned copy, and scoring recall with it degrades what the resident embedder already stored", heads)
+	}
+	return nil
+}
+
+// absClean — the comparable form of a path. Abs fails only when the working
+// directory is unreadable, and a cleaned relative path is still worth comparing,
+// so the error falls back rather than propagating.
+func absClean(p string) string {
+	if abs, err := filepath.Abs(p); err == nil {
+		return abs
+	}
+	return filepath.Clean(p)
+}
+
 // VoiceConfig — the client↔core TCP surface + the stt/tts worker-module
 // seams.
 //

internal/config/voice_test.go

@@ -0,0 +1,70 @@
+package config
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+)
+
+// The heads graph is a fine-tuned copy of the embedder, and pointing both keys
+// at one file degrades recall with no error and no log line (V-692). These are
+// the shapes that used to boot clean.
+func TestHeadsPathMustNotBeTheModelFile(t *testing.T) {
+	dir := t.TempDir()
+	model := filepath.Join(dir, "model.onnx")
+	heads := filepath.Join(dir, "heads.onnx")
+	for _, p := range []string{model, heads} {
+		if err := os.WriteFile(p, []byte("onnx"), 0o600); err != nil {
+			t.Fatalf("write %s: %v", p, err)
+		}
+	}
+	link := filepath.Join(dir, "link.onnx")
+	if err := os.Symlink(model, link); err != nil {
+		t.Fatalf("symlink: %v", err)
+	}
+
+	// A path that stats and one that does not, because the guard compares the
+	// cleaned string before it stats anything.
+	for name, headsPath := range map[string]string{
+		"the same path":     model,
+		"a symlink to it":   link,
+		"an uncleaned path": filepath.Join(dir, ".", "sub", "..", "model.onnx"),
+		"a path on no disk": filepath.Join(dir, "absent.onnx"),
+	} {
+		t.Run(name, func(t *testing.T) {
+			same := headsPath != filepath.Join(dir, "absent.onnx")
+			err := voiceConfigWith(t, model, headsPath)
+			if same && err == nil {
+				t.Fatal("want a startup error, got a daemon that routes worse in silence")
+			}
+			if same && !strings.Contains(err.Error(), "heads_path") {
+				t.Fatalf("the error does not name the key: %v", err)
+			}
+			if !same && err != nil {
+				t.Fatalf("a distinct heads_path was refused: %v", err)
+			}
+		})
+	}
+
+	if err := voiceConfigWith(t, model, heads); err != nil {
+		t.Fatalf("two distinct files were refused: %v", err)
+	}
+	if err := voiceConfigWith(t, model, ""); err != nil {
+		t.Fatalf("no heads at all was refused: %v", err)
+	}
+}
+
+// voiceConfigWith loads a minimal enabled voice block through the real Load, so
+// the test exercises the startup path and not just the check in isolation.
+func voiceConfigWith(t *testing.T, model, heads string) error {
+	t.Helper()
+	body := `{"voice":{"enabled":true,"bind":"127.0.0.1:9100","embedder":{` +
+		`"model_path":"` + model + `","tokenizer_path":"/t.json","lib_path":"/l.so"`
+	if heads != "" {
+		body += `,"heads_path":"` + heads + `"`
+	}
+	body += `}}}`
+	_, err := Load(writeConfig(t, body))
+	return err
+}

internal/config/workstation.go

@@ -1,6 +1,7 @@
 package config
 
 import (
+	"net/url"
 	"strings"
 	"time"
 )
@@ -26,6 +27,13 @@ type WorkstationConfig struct {
 	// signal, so it must be the supervisor's endpoint and not llama-server's.
 	Health string `json:"health,omitempty"`
 
+	// Token — the bearer credential mavgpud requires, expanded from the
+	// environment like every other secret here. It is what stops anything on
+	// the LAN spending the card, so a URL that is not loopback needs one.
+	// Wrong or missing reads as a workstation that is down, and Maven falls
+	// back to the resident model.
+	Token string `json:"token,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.
@@ -35,12 +43,54 @@ type WorkstationConfig struct {
 	// 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"`
+
+	// Stt — CrisperWhisper 2.0 on the same machine, a separate service on its
+	// own port. Absent ⇒ every utterance goes to mavsttd, which is today.
+	Stt *WorkstationSttConfig `json:"stt,omitempty"`
+}
+
+// WorkstationSttConfig — speech-to-text on the workstation.
+//
+// It is a second service and not a second endpoint on mavgpud: whisper.cpp
+// cannot load CrisperWhisper 2.0 at all, because it derives its language count
+// from the vocabulary size and CW2's 51897 tokens shift seven special token
+// ids. So CW2 runs under transformers, and this block addresses it.
+//
+// Worth the trouble: CW2 turbo scores 10.4% WER in Russian against 27.5% for
+// the ggml-small.bin homesrv loads
+// (docs/evals/2026-08-09-crisperwhisper2-russian-wer.md).
+type WorkstationSttConfig struct {
+	// URL — the transcribe endpoint, e.g.
+	// "http://192.168.1.105:8081/transcribe". Empty ⇒ the block is normalised
+	// to nil and mavsttd takes every turn.
+	URL string `json:"url,omitempty"`
+
+	// Health — the admission endpoint. Empty ⇒ the URL's origin + "/health".
+	// It answers 503 while the card is held, and that is the signal.
+	Health string `json:"health,omitempty"`
+
+	// Token — the bearer token the service checks. Audio is the most sensitive
+	// thing that crosses this seam, so a LAN deployment should set one. Write
+	// it as ${MAVEN_STT_TOKEN} and keep the value in deploy/telegram.env, the
+	// way every other secret in this file is written.
+	Token string `json:"token,omitempty"`
+
+	// Probe — how often admission is re-checked. 0 ⇒ DefaultWorkstationProbe.
+	Probe Duration `json:"probe,omitempty"`
+
+	// Timeout — the per-request budget for one utterance. 0 ⇒
+	// DefaultWorkstationSttTimeout. A request that overruns falls back to
+	// mavsttd, which costs a worse transcript and not the turn.
+	Timeout Duration `json:"timeout,omitempty"`
 }
 
 // Workstation defaults, applied in normaliseWorkstation.
 const (
 	DefaultWorkstationProbe   = 15 * time.Second
 	DefaultWorkstationTimeout = 90 * time.Second
+	// One utterance, not one completion. A voice turn waits on this, so the
+	// budget is a few seconds and not a minute and a half.
+	DefaultWorkstationSttTimeout = 10 * time.Second
 )
 
 // normaliseWorkstation applies the block's defaults. No address, no preferred
@@ -63,4 +113,36 @@ func (c *Config) normaliseWorkstation() {
 	if w.Timeout <= 0 {
 		w.Timeout = Duration(DefaultWorkstationTimeout)
 	}
+	normaliseWorkstationStt(w)
+}
+
+// normaliseWorkstationStt applies the speech-to-text block's defaults. No
+// address, no remote: mavsttd then takes every utterance, which is today.
+func normaliseWorkstationStt(w *WorkstationConfig) {
+	if w.Stt != nil && strings.TrimSpace(w.Stt.URL) == "" {
+		w.Stt = nil
+	}
+	if w.Stt == nil {
+		return
+	}
+	s := w.Stt
+	if strings.TrimSpace(s.Health) == "" {
+		s.Health = healthOrigin(s.URL)
+	}
+	if s.Probe <= 0 {
+		s.Probe = Duration(DefaultWorkstationProbe)
+	}
+	if s.Timeout <= 0 {
+		s.Timeout = Duration(DefaultWorkstationSttTimeout)
+	}
+}
+
+// healthOrigin derives the admission endpoint from the transcribe endpoint.
+// The URL names a path, so appending to it would ask for /transcribe/health.
+func healthOrigin(raw string) string {
+	u, err := url.Parse(raw)
+	if err != nil || u.Host == "" {
+		return strings.TrimRight(raw, "/") + "/health"
+	}
+	return u.Scheme + "://" + u.Host + "/health"
 }

internal/dialogue/clarify.go

@@ -47,6 +47,11 @@ type PendingQuestion struct {
 	// charging it a retry is the V-554 shape. See CanResume for why it is
 	// counted at all.
 	Suspends int
+	// Rides counts every turn this question has ridden out on the end of
+	// someone else's reply, over the whole life of the request. Unlike Suspends
+	// it is never reset and never re-based, which is the only property that
+	// matters about it (V-663).
+	Rides int
 }
 
 // MaxSuspends — how many times one question may step aside and come back before
@@ -63,10 +68,52 @@ type PendingQuestion struct {
 // is that he has moved on and has not said so.
 const MaxSuspends = 3
 
+// MaxRides — how many turns one question may ride out on the end of an
+// unrelated reply, counted over its whole life (V-663).
+//
+// MaxSuspends did not move the measurement it was written for. Twenty-six of
+// 140 turns carried a tail before it landed and twenty-six carried one after.
+// Every bound on this question is rearmed by something ordinary:
+//
+//   - The TTL is an inactivity timer, and both noteSuspended and reaskOrGiveUp
+//     restart it, so it cannot arrive while he keeps talking.
+//   - Suspends is zeroed by any turn that reads as an answer, which is where
+//     "спасибо" and "привет" land. It resets before anything is known to have
+//     been filled.
+//   - askRemainingGap builds a fresh question for the second gap, so a reminder
+//     with two gaps gets a new allowance halfway through.
+//
+// So Suspends only bites on four strictly consecutive side queries with nothing
+// chat-like between them, which is not the shape real conversation has. Rides is
+// the same idea with the resets taken out: set once, incremented, carried
+// across a re-park, and read by nothing that could lower it.
+//
+// The shape it is aimed at is measured, not imagined. In the 2026-08-08 run one
+// question about a reminder's day rode turns 7 to 13 and ended only because
+// turn 14 was a new request. Three asides, then two turns that read as failed
+// answers, then two more asides. The asides spend no attempt and the answers
+// reset Suspends, so the two bounds take turns being rearmed by the other's
+// traffic.
+//
+// Four, not three. It has to be looser than MaxSuspends or that bound is dead
+// code, because Rides is never lower than Suspends and would always fire first.
+//
+// Do not read this as a fix for the whole ride. It ends the measured one a turn
+// early and no more. Most of that ride's length is attempts, spent by turns
+// like "спасибо" and "привет" being read as failed answers to a question about
+// a day. That is a defect in classifyTurnRole and not in any bound here.
+const MaxRides = 4
+
 // CanResume reports whether this question may step aside once more. False ⇒ the
 // caller lets the request go and says so; it must never simply stop resuming,
 // because a question dropped in silence reads as one that was answered.
-func (q *PendingQuestion) CanResume() bool { return q.Suspends < MaxSuspends }
+//
+// Two bounds, and they answer different questions. Suspends asks whether he has
+// walked away from this exchange in the last few turns. Rides asks whether this
+// question has been riding long enough that the answer is no regardless.
+func (q *PendingQuestion) CanResume() bool {
+	return q.Suspends < MaxSuspends && q.Rides < MaxRides
+}
 
 // Action reads the parked question as the typed action it is assembling
 // (pending.go). Derived rather than stored: the question's fields stay the one

internal/kiwix/client.go

@@ -127,6 +127,22 @@ func (c *Client) Article(ctx context.Context, path string, maxRunes int) (crawl.
 	return crawl.Extract(u, body, maxRunes), nil
 }
 
+// TitlePath is the article path for an exact title, for Article to fetch.
+//
+// It exists because a ZIM is addressable by title and the full-text index is
+// not the only way in. "Франция", "TCP" and "Небо" resolve; "Трюмбальная
+// нидроскопия" is a 404, which is the honest answer and the reason this is
+// safe to try first. Measured on 2026-08-09, keyword search on the same terms
+// returns "Список пэров Франции" and "Список портов TCP и UDP" instead.
+//
+// A miss is normal rather than a failure. An article whose title inverts a name
+// ("Торвальдс, Линус") is a 404 here and the first hit in search, so the caller
+// falls through and loses nothing.
+func TitlePath(book, title string) string {
+	t := strings.ReplaceAll(strings.TrimSpace(title), " ", "_")
+	return "/content/" + url.PathEscape(book) + "/A/" + url.PathEscape(t)
+}
+
 // rss mirrors just the bits of the RSS 2.0 reply we use.
 type rss struct {
 	Items []struct {

internal/kiwix/live_test.go

@@ -0,0 +1,69 @@
+package kiwix
+
+import (
+	"context"
+	"os"
+	"testing"
+	"time"
+)
+
+// TestLiveTopicBeatsTheSentence — the measurement V-668 turned on, kept as a
+// test so the claim can be re-run rather than believed.
+//
+// It prints the article the old path returned and the article the new one
+// returns, for the same question. It asserts nothing about which is better,
+// because "is this the right article" is a human's call. It fails only if the
+// two paths agree on every case, which would mean the change does nothing.
+//
+//	MAVEN_KIWIX_URL=http://127.0.0.1:8034 make t PKG=./internal/kiwix/ RUN=TestLive V=1
+func TestLiveTopicBeatsTheSentence(t *testing.T) {
+	base := os.Getenv("MAVEN_KIWIX_URL")
+	if base == "" {
+		t.Skip("MAVEN_KIWIX_URL unset — point it at the kiwix-server host port")
+	}
+	const book = "wikipedia_ru_all_maxi_2026-02"
+	c := New(base)
+	questions := []string{
+		"что такое TCP?",
+		"что такое фотосинтез",
+		"кто такой Линус Торвальдс?",
+		"кто написал Войну и мир",
+		"что такое чёрная дыра",
+		"почему небо голубое",
+		"почему трава зелёная",
+		"столица Франции",
+	}
+	moved := 0
+	for _, q := range questions {
+		ctx, cancel := context.WithTimeout(context.Background(), 20*time.Second)
+		before := firstTitle(ctx, c, q, book)
+		topic := Topic(q)
+		after := ""
+		for _, cand := range TitleCandidates(topic) {
+			if page, err := c.Article(ctx, TitlePath(book, cand), 400); err == nil && page.Text != "" {
+				after = page.Title + " (by title)"
+				break
+			}
+		}
+		if after == "" {
+			after = firstTitle(ctx, c, topic, book)
+		}
+		cancel()
+		if before != after {
+			moved++
+		}
+		t.Logf("%-30s  before=%-34q after=%q", q, before, after)
+	}
+	t.Logf("%d of %d questions reach a different article", moved, len(questions))
+	if moved == 0 {
+		t.Error("the topic path returns exactly what the sentence path returned")
+	}
+}
+
+func firstTitle(ctx context.Context, c *Client, pattern, book string) string {
+	hits, err := c.Search(ctx, pattern, book, 3)
+	if err != nil || len(hits) == 0 {
+		return "(nothing)"
+	}
+	return hits[0].Title
+}

internal/kiwix/topic.go

@@ -0,0 +1,118 @@
+package kiwix
+
+import (
+	"strings"
+	"unicode"
+
+	"github.com/kami/maven/internal/lexicon"
+	"github.com/kami/maven/internal/morph"
+)
+
+// Topic reduces a question to the thing it is about, because Kiwix ranks by
+// keyword overlap and a whole sentence buries the keyword that matters.
+//
+// This package's own doc says it: "why is the sky blue" finds a TV episode.
+// Measured against the Russian ZIM on 2026-08-09, the sentence and the topic
+// return different articles for the same question. "кто написал Войну и мир"
+// returns "Радуйся, мир (Доктор Кто)"; "Войну и мир" returns the novel first.
+// "что такое TCP" returns "Перехват TCP-соединения"; "TCP" returns TCP. The
+// English path had a rewriter doing this with a model call. The Russian path
+// reads the book verbatim (V-508) and had nothing.
+//
+// It drops three things off the front and stops: the narrative request, the
+// interrogative, and a verb sitting between them and the noun. Everything else
+// is kept, because a word this cannot classify is more likely the topic than
+// noise. An empty return means the utterance was question words alone, and the
+// caller searches the sentence as before.
+func Topic(utterance string) string {
+	words := strings.Fields(strings.TrimSpace(utterance))
+	cut := 0
+	for cut < len(words) {
+		w := strings.Trim(strings.ToLower(words[cut]), ".,!?…:;\"'«»")
+		if w == "" {
+			cut++
+			continue
+		}
+		switch {
+		case inList(lexicon.NarrativeRequests(), w),
+			inList(lexicon.Interrogatives(), w),
+			inList(lexicon.FirstPerson(), w),
+			// "что ТАКОЕ x", "кто ТАКОЙ x" — the copula that only ever follows
+			// an interrogative, and never a topic on its own.
+			cut > 0 && isCopula(w),
+			// "расскажи ПРО x", "о x". One-letter and two-letter prepositions
+			// are not a closed class worth a lexicon set of their own.
+			cut > 0 && isLeadingPreposition(w),
+			// "кто НАПИСАЛ Войну и мир". A verb here is the question's own
+			// verb, not part of the title. Only after something was already
+			// dropped, so "написал отчёт" as a topic survives intact.
+			cut > 0 && morph.IsVerbForm(w):
+			cut++
+		default:
+			// The question mark is the sentence's, not the title's, and Kiwix
+			// carries it into the keyword match.
+			topic := strings.TrimRight(strings.Join(words[cut:], " "), " .,!?…:;\"'«»")
+			if !hasLetter(topic) {
+				return ""
+			}
+			return topic
+		}
+	}
+	return ""
+}
+
+// TitleCandidates is the topic as it might be titled, best first.
+//
+// A ZIM title is capitalized and the utterance is not: measured on 2026-08-09,
+// `/A/фотосинтез` is a 404 and `/A/Фотосинтез` is a 200. The spoken form is
+// tried first anyway, because a title that begins lowercase on purpose
+// ("iPhone") would not survive capitalizing it. Both are one request each
+// against a server on the same box, and a miss is a 404 rather than a wrong
+// article.
+func TitleCandidates(topic string) []string {
+	if topic == "" {
+		return nil
+	}
+	r := []rune(topic)
+	up := unicode.ToUpper(r[0])
+	if up == r[0] {
+		return []string{topic}
+	}
+	return []string{topic, string(up) + string(r[1:])}
+}
+
+func isCopula(w string) bool {
+	switch w {
+	case "такое", "такой", "такая", "такие", "is", "are", "was", "were":
+		return true
+	}
+	return false
+}
+
+func isLeadingPreposition(w string) bool {
+	switch w {
+	case "про", "о", "об", "обо", "по", "about", "of", "on":
+		return true
+	}
+	return false
+}
+
+func inList(list []string, w string) bool {
+	for _, x := range list {
+		if x == w {
+			return true
+		}
+	}
+	return false
+}
+
+// hasLetter is the guard against a topic that reduced to punctuation or digits
+// alone, which no ZIM title matches.
+func hasLetter(s string) bool {
+	for _, r := range s {
+		if unicode.IsLetter(r) {
+			return true
+		}
+	}
+	return false
+}

internal/kiwix/topic_test.go

@@ -0,0 +1,67 @@
+package kiwix
+
+import "testing"
+
+// The cases the 2026-08-09 measurement turned on, plus the ones a topic must
+// not damage. Each left column returned a wrong article when it was sent whole.
+func TestTopicKeepsTheThingTheQuestionIsAbout(t *testing.T) {
+	cases := []struct{ utterance, want string }{
+		{"что такое TCP?", "TCP"},
+		{"что такое фотосинтез", "фотосинтез"},
+		{"кто такой Линус Торвальдс?", "Линус Торвальдс"},
+		{"кто написал Войну и мир", "Войну и мир"},
+		{"расскажи про битву при Ватерлоо", "битву при Ватерлоо"},
+		{"what is photosynthesis", "photosynthesis"},
+		// No question word, so there is nothing to drop. The topic is the
+		// whole utterance and the search is what it was before.
+		{"столица Франции", "столица Франции"},
+		{"почему небо голубое", "небо голубое"},
+	}
+	for _, c := range cases {
+		if got := Topic(c.utterance); got != c.want {
+			t.Errorf("Topic(%q) = %q, want %q", c.utterance, got, c.want)
+		}
+	}
+}
+
+// A verb only goes when a question word already went. Otherwise "написал
+// отчёт" loses the verb that names what he means.
+func TestTopicDropsAVerbOnlyBehindAQuestionWord(t *testing.T) {
+	if got := Topic("написал отчёт"); got != "написал отчёт" {
+		t.Errorf("Topic dropped a leading verb with no question word: %q", got)
+	}
+}
+
+// Question words alone reduce to nothing, and the caller reads that as "no
+// topic" and searches the sentence rather than searching the empty string.
+func TestTopicIsEmptyWhenNothingIsLeft(t *testing.T) {
+	for _, q := range []string{"что такое?", "кто?", "почему", "???"} {
+		if got := Topic(q); got != "" {
+			t.Errorf("Topic(%q) = %q, want empty", q, got)
+		}
+	}
+}
+
+func TestTitlePathEscapesAndUnderscores(t *testing.T) {
+	got := TitlePath("wikipedia_ru_all_maxi_2026-02", "Чёрная дыра")
+	want := "/content/wikipedia_ru_all_maxi_2026-02/A/%D0%A7%D1%91%D1%80%D0%BD%D0%B0%D1%8F_%D0%B4%D1%8B%D1%80%D0%B0"
+	if got != want {
+		t.Errorf("TitlePath = %q, want %q", got, want)
+	}
+}
+
+// A ZIM title carries a leading capital and the utterance does not. The spoken
+// form is still tried first, so a title that begins lowercase on purpose keeps
+// its chance.
+func TestTitleCandidatesTryTheSpokenFormFirst(t *testing.T) {
+	got := TitleCandidates("фотосинтез")
+	if len(got) != 2 || got[0] != "фотосинтез" || got[1] != "Фотосинтез" {
+		t.Errorf("TitleCandidates = %q", got)
+	}
+	if got := TitleCandidates("TCP"); len(got) != 1 || got[0] != "TCP" {
+		t.Errorf("an already-capital topic was tried twice: %q", got)
+	}
+	if got := TitleCandidates(""); got != nil {
+		t.Errorf("TitleCandidates(\"\") = %q, want nil", got)
+	}
+}

internal/lexicon/lexicon.go

@@ -119,6 +119,11 @@ func PartsOfDay() []string { return words("parts_of_day") }
 // ReminderVerbs returns the imperatives that open a reminder.
 func ReminderVerbs() []string { return words("reminder_verbs") }
 
+// Pleasantries returns the whole utterances that greet, thank or say goodbye.
+// Whole utterances and not tokens: see the set's own note for why the tokens
+// are unsafe alone.
+func Pleasantries() []string { return words("pleasantries") }
+
 // TaskDoneWords returns the words that finish a task, and TaskDropWords the
 // words that abandon one. Two sets rather than one with a value, because the
 // store records which of the two happened and the caller has to say so.

internal/lexicon/lexicon_ru_v1.json

@@ -176,6 +176,19 @@
         "morning", "afternoon", "evening", "night"
       ]
     },
+    "pleasantries": {
+      "note": "Whole utterances that greet, thank or say goodbye. They ask for nothing and answer nothing, so a parked question must neither consume them as a failed answer nor be dropped by them (V-663). Matched as WHOLE utterances and never as tokens, because the tokens are not safe alone: \"вечер\" answers \"это утра или вечера?\" and \"нет\" answers a confirm. Anything that could fill a slot stays out. The control words (\"стоп\", \"отмена\") stay out too, because isCancel already owns them and they mean something stronger.",
+      "words": [
+        "привет", "приветик", "здравствуй", "здравствуйте",
+        "доброе утро", "добрый день", "добрый вечер",
+        "пока", "прощай", "до свидания", "спокойной ночи",
+        "спасибо", "спасибо тебе", "большое спасибо", "благодарю",
+        "извини", "извините", "прости", "простите",
+        "hi", "hello", "hey", "bye", "goodbye",
+        "good morning", "good evening", "good night",
+        "thanks", "thank you", "thanks a lot", "sorry"
+      ]
+    },
     "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). 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": [

internal/llm/client.go

@@ -51,6 +51,10 @@ type Client struct {
 	base string
 	swap SwapGate
 	http *http.Client
+	// token — the bearer credential for a server that asks for one. Empty for
+	// the resident model, which is reached over loopback on the same box.
+	// mavgpud on the workstation requires it: that hop is on the LAN.
+	token string
 
 	// gate / background — priority on the single llama-server slot. Set once
 	// at wiring time (SetGate), read on every request. nil gate ⇒ no gating,
@@ -101,6 +105,27 @@ func New(baseURL string, timeout time.Duration) *Client {
 	return &Client{base: baseURL, http: &http.Client{Timeout: timeout}}
 }
 
+// SetToken installs the bearer credential this client sends. Wiring-time, like
+// SetGate: an empty token means the server is not asking for one.
+func (c *Client) SetToken(t string) {
+	c.mu.Lock()
+	c.token = t
+	c.mu.Unlock()
+}
+
+// authorize adds the credential when there is one. Exported to the package so
+// the Pair prober signs /health with the same token as the completion — a
+// probe that answers 401 would otherwise read as a workstation that is down,
+// and Maven would fall back forever without saying why.
+func (c *Client) authorize(req *http.Request) {
+	c.mu.RLock()
+	t := c.token
+	c.mu.RUnlock()
+	if t != "" {
+		req.Header.Set("Authorization", "Bearer "+t)
+	}
+}
+
 // SetBaseURL re-points the client at another llama-server. Safe to call while
 // requests are in flight: a request that already read the old base finishes
 // against the old base (or fails, and every caller of Complete has a fallback),
@@ -196,6 +221,7 @@ func (c *Client) Complete(ctx context.Context, r Req) (string, error) {
 		return "", err
 	}
 	req.Header.Set("Content-Type", "application/json")
+	c.authorize(req)
 	httpResp, err := c.http.Do(req)
 	if err != nil {
 		return "", err

internal/llm/remote.go

@@ -132,6 +132,9 @@ func (p *Pair) probe(ctx context.Context) {
 		p.set(false)
 		return
 	}
+	if p.remote != nil {
+		p.remote.authorize(req)
+	}
 	resp, err := p.http.Do(req)
 	if err != nil {
 		p.set(false)

internal/llm/remote_test.go

@@ -250,3 +250,35 @@ func TestNoFloorIsAnError(t *testing.T) {
 		t.Fatalf("err = %v, want ErrNoFloor", err)
 	}
 }
+
+// The workstation is behind mavgpud, which requires a bearer token on the
+// completion and on /health alike. A probe that did not carry it would answer
+// 401, Pair would read that as a card that is busy, and every turn would fall
+// back to the resident model with nothing in the log naming why.
+func TestPairSignsTheProbeAndTheCompletion(t *testing.T) {
+	seen := make(chan string, 2)
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		seen <- r.Header.Get("Authorization")
+		if r.URL.Path == "/health" {
+			return
+		}
+		_, _ = w.Write([]byte(`{"choices":[{"message":{"content":"ok"}}]}`))
+	}))
+	defer srv.Close()
+
+	remote := New(srv.URL, time.Second)
+	remote.SetToken("s3cret")
+	p := NewPair(remote, New(srv.URL, time.Second), srv.URL+"/health", time.Hour)
+	p.probe(context.Background())
+	if !p.Available() {
+		t.Fatal("the probe did not admit an answering workstation")
+	}
+	if _, err := p.Complete(context.Background(), Req{User: "привет"}); err != nil {
+		t.Fatalf("complete: %v", err)
+	}
+	for i := 0; i < 2; i++ {
+		if got := <-seen; got != "Bearer s3cret" {
+			t.Errorf("request %d carried %q, want the bearer token", i, got)
+		}
+	}
+}

internal/router/decisiontrace.go

@@ -14,12 +14,15 @@ import (
 	"github.com/kami/maven/internal/decision"
 )
 
-// The two routing engines, named as claimants. They are one stage and not two,
-// because only one of them ever runs: the classifier is reached when the model
-// is absent or errored, never alongside it.
+// The three routing engines, named as claimants. The model and the classifier
+// are one stage and not two, because only one of them ever runs: the classifier
+// is reached when the model is absent or errored, never alongside it. The heads
+// run before both and decline on low confidence, so they can appear beside
+// either one in a record.
 const (
 	claimantLLM        = "llm-router"
 	claimantClassifier = "classifier"
+	claimantHeads      = "routing-heads"
 )
 
 // thinReason names which arm of gateLLMDecision cut the confidence. The gate

internal/router/dumpprompt_test.go

@@ -0,0 +1,22 @@
+package router
+
+import (
+	"os"
+	"testing"
+)
+
+// TestDumpPrompt writes the router prompt and grammar to disk so the training
+// workspace labels with the daemon's own contract rather than a retyped copy.
+// It is inert unless MAVEN_DUMP_PROMPT names a directory.
+func TestDumpPrompt(t *testing.T) {
+	dir := os.Getenv("MAVEN_DUMP_PROMPT")
+	if dir == "" {
+		t.Skip("MAVEN_DUMP_PROMPT unset")
+	}
+	if err := os.WriteFile(dir+"/route_system.txt", []byte(routeSystem), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(dir+"/route_grammar.gbnf", []byte(routeGrammar), 0o644); err != nil {
+		t.Fatal(err)
+	}
+}

internal/router/embedderid_test.go

@@ -1,10 +1,13 @@
 package router
 
-import "testing"
+import (
+	"strings"
+	"testing"
+)
 
 func TestEmbedderIDFromModelPath(t *testing.T) {
 	got := modelIDFromPath("/opt/maven/models/embedder/multilingual-e5-small.onnx")
-	if got != "multilingual-e5-small@384" {
+	if got != "multilingual-e5-small@384/tok2" {
 		t.Fatalf("modelIDFromPath = %q", got)
 	}
 	// A different model file must produce a different id, even at 384 dim.
@@ -12,6 +15,13 @@ func TestEmbedderIDFromModelPath(t *testing.T) {
 	if old == got {
 		t.Fatal("two different models share one id")
 	}
+	// The tokenizer is half of what makes a vector, and it changes under a
+	// model file whose name never moves (V-664). An id that ignored it would
+	// leave stored passages in one space and every new query in another, with
+	// nothing to trigger the re-embed.
+	if !strings.Contains(got, "/tok") {
+		t.Fatalf("id %q does not name the tokenizer revision", got)
+	}
 }
 
 func TestEmbedderIDIncludesDim(t *testing.T) {

internal/router/eval/eval_test.go

@@ -255,35 +255,12 @@ func newBaselineClassifier(t *testing.T, emb router.Embedder) *router.Classifier
 	return cls
 }
 
-// baselineGrammars — the stage-0 rule set in the daemon's order (buildRouter in
-// cmd/mavend/voicewire.go). Split out of newBaselineRouter so the claim
-// measurement can run the same rules one at a time and see which of them
-// contend for the same utterance, which the cascade hides by stopping at the
-// first match.
+// baselineGrammars — the stage 0 rule set the daemon runs, from the one place
+// it is written down (V-693). It used to restate the list by hand, and by the
+// time the guard was written the two had already drifted by one grammar.
+// Kept as a name because the claim measurement reads it as "the baseline set".
 func baselineGrammars(acts router.ActMatcher) []router.Grammar {
-	grammars := router.DefaultGrammars(acts)
-	grammars = append(grammars, router.SystemTimeDateGrammars()...)
-	// Same order as buildRouter (voicewire.go). The fixture is only worth
-	// anything while its grammar set is the daemon's grammar set.
-	grammars = append(grammars, router.AgendaQueryGrammars()...)
-	// After the agenda rules and before the feed and list rules, same as
-	// voicewire.go: "что такое лента" is a definition question and the feed
-	// rule would claim it on the noun alone (V-655). Missing here until V-659,
-	// so the fixture was scoring a grammar set the daemon does not run.
-	grammars = append(grammars, router.WorldQueryGrammars()...)
-	grammars = append(grammars, router.FeedQueryGrammar())
-	// The list side of the same exposure: a phrasing with no possessive in it
-	// ("список дел") routed system and never reached queryTasks (Vikunja #467).
-	grammars = append(grammars, router.TaskListGrammar())
-	grammars = append(grammars, router.ListGrammars()...)
-	grammars = append(grammars, router.ReminderGrammar())
-	grammars = append(grammars, router.PraxisGrammars()...)
-	grammars = append(grammars, router.TaskStatusGrammar())
-	grammars = append(grammars, router.TaskCaptureGrammar())
-	// "расскажи про X" is a world question the model called a fact, and the
-	// rule goes last because it matches on the first word alone (Vikunja #498).
-	grammars = append(grammars, router.NarrativeQueryGrammars()...)
-	return grammars
+	return router.StageZeroGrammars(acts)
 }
 
 // seedOrder — fixed iteration order over the corpus. Not cosmetic: a few

internal/router/eval/heads_test.go

@@ -0,0 +1,86 @@
+package eval
+
+import (
+	"context"
+	"os"
+	"path/filepath"
+	"testing"
+
+	"github.com/kami/maven/internal/config"
+	"github.com/kami/maven/internal/router"
+)
+
+// TestONNXRoutingHeads — the cascade with the routing heads wired, which is
+// what V-664 deploys. Opt-in via MAVEN_ONNX_LIB, same as TestONNXBaseline, and
+// one TestONNX* per process.
+//
+// The comparison worth reading is against TestONNXBaseline, which is the same
+// cascade with the same grammars and the same classifier floor and no heads.
+// Only the middle arm varies.
+//
+// It also checks the Go unigram tokenizer against the Python one, because the
+// heads were trained through transformers and are read through a hand-written
+// tokenizer. A mismatch shows up here as a score below what Python measured on
+// the same weights, and nowhere else.
+func TestONNXRoutingHeads(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")
+	}
+	// Absolute, because onnxruntime resolves a graph's external weights file
+	// against the model path it was given, and a relative one lands in the
+	// test's working directory.
+	root, err := filepath.Abs("../../..")
+	if err != nil {
+		t.Fatal(err)
+	}
+	model := filepath.Join(root, "models/embedder/multilingual-e5-small/model_quantized.onnx")
+	tok := filepath.Join(root, "models/embedder/multilingual-e5-small/tokenizer.json")
+	heads := filepath.Join(root, "models/embedder/router-heads/router_heads.onnx")
+	for _, p := range []string{lib, model, tok, heads} {
+		if _, err := os.Stat(p); err != nil {
+			t.Skipf("missing %s: %v", p, err)
+		}
+	}
+	emb, err2 := router.NewONNXEmbedder(model, tok, lib)
+	if err2 != nil {
+		t.Skipf("onnx embedder unavailable: %v", err2)
+	}
+	err = nil
+	defer emb.Close()
+
+	h, err := router.NewRouterHeads(heads, tok)
+	if err != nil {
+		t.Skipf("routing heads unavailable: %v", err)
+	}
+	defer h.Close()
+
+	f, err := Load()
+	if err != nil {
+		t.Fatalf("Load: %v", err)
+	}
+	rep, err := Score(context.Background(), "heads+classifier", withHeads(t, emb, h), f)
+	if err != nil {
+		t.Fatalf("Score: %v", err)
+	}
+	t.Log("\n" + rep.String() + rep.Failures())
+}
+
+// withHeads mirrors newBaselineRouter and adds the one arm under test. It is a
+// separate function rather than a parameter so the baseline's signature stays
+// the shape every other test calls it with.
+func withHeads(t *testing.T, emb router.Embedder, h *router.RouterHeads) *router.Router {
+	t.Helper()
+	acts := router.DefaultActMatcher{Fns: actFns}
+	return router.New(router.Config{
+		Grammars:   baselineGrammars(acts),
+		Classifier: newBaselineClassifier(t, emb),
+		Extractor: router.Extractor{
+			Time:  router.StubDateTimeParser{},
+			Acts:  acts,
+			Facts: router.DefaultFactParser{},
+		},
+		Threshold: config.DefaultRouterThreshold,
+		Heads:     h,
+	})
+}