Merge pull request 'Bug: the memory index stores the raw utterance as a fact's recall text, and nothing ever deletes a fact vector' (#100) from task/493-bug-the-memory-index-stores-the-raw-utte into task/470-bug-a-question-writes-invented-knowledge

Reviewed-on: #100

.gitignore

@@ -9,6 +9,7 @@
 /mavwaked
 /mavmaild
 /mavupdate
+/mavgpud
 
 # Certs (private keys, don't commit)
 certs/
@@ -40,6 +41,9 @@ deploy/telegram.env
 deploy/zenmoney.token
 # IMAP password, read by mavmaild (never in argv, never committed)
 deploy/imap.password
+# Compose interpolation secrets — MAVEN_AMBIENT_TOKEN today. docker compose
+# reads this file itself; it is not an env_file on any service.
+/.env
 
 # Temp files
 /tmp/
@@ -63,3 +67,6 @@ coverage.out
 /HANDOFF.md
 /models/stt
 /models/tts
+
+# root .env — MAVEN_AMBIENT_TOKEN and friends, same class as deploy/telegram.env
+.env

20-07-2026-BACKLOG.md

@@ -1,396 +0,0 @@
-beyond the model and tts work, the useful additions are mostly around **reliability, context, and reach**, not more intelligence.
-
-## highest-value additions
-
-### 1. unified event intake
-
-maven should receive normalized events from:
-
-* praxis
-* calendar
-* telegram
-* local notifications
-* system/service health
-* manual checklists
-* eventually email bridges
-
-one internal envelope:
-
-```go
-type Event struct {
-    Source     string
-    Kind       string
-    EntityIDs  []string
-    Title      string
-    Body       string
-    Priority   string
-    OccurredAt time.Time
-    Payload    json.RawMessage
-}
-```
-
-this gives digestion one stable input instead of source-specific logic.
-
----
-
-### 2. explicit morning routine engine — **core engine done (2026-07-20)**
-
-`internal/morning` — pure checklist engine, mirrors `internal/loop`/
-`internal/routine`'s no-I/O contract. `Evaluate(routine, facts, now)` answers
-"what's still missing" any time (order-independent — checks facts, not
-sequence); `Due(routines, facts, last, now)` fires the once-per-day nag only
-at `NudgeAt` (defaults to window end) and only when something's unevidenced,
-with a `last`-map dedupe identical in shape to `routine.Due`'s cold-start/
-last-fire tracking. Evidence is just a fact timestamped inside today's
-window — manual (voice-tapped) and inferred (another daemon writing the same
-key) are indistinguishable, satisfying the manual/inferred requirement for
-free. Weekday/weekend variants are two `Routine`s with different `Weekdays`
-sets under different names. Wired into `config.MorningRoutineConfig` +
-`cmd/mavend/tick.go`'s `fireMorningRoutines` (reads only the fact keys the
-configured items reference, dispatches through the normal severity/presence
-routing table, body is literal joined item labels — not LLM-phrased, same
-no-hallucination rationale as cron routines). 13 unit tests in
-`internal/morning/morning_test.go`.
-
-Added since (2026-07-20, same day): a read-only `/morning` page in mavweb —
-`ipc.CoreAPI.MorningStatus` (new wire method, mirrors `TickTrace`'s
-daemon-cache-only shape: the store adapter errors, `daemonAPI` serves it from
-a `tickLoop.morningStatus` closure) returns each routine's active/window/
-per-item done state, server-rendered same as `/trace` (no live-update loop —
-checklist state moves on minutes, not seconds).
-
-Not yet done: no config wired in `deploy/mavend.json` (no morning routines
-configured on homesrv yet — add items there when the medicine/water/pets
-fact keys the phone/desktop write are settled), no voice query path for
-"what did I miss this morning" (Evaluate supports it; nothing calls it yet),
-no way to create/edit routines from the web UI — construction still means
-hand-editing config, deliberately deferred: routines are operator-declared
-config (like cron routines), and a CRUD editor would mean moving them to a
-DB table + hot-reload, a bigger change than this pass.
-
-not ordinary reminders.
-
-support:
-
-* required morning items
-* order-independent completion
-* soft time windows
-* skipped-step detection
-* one nudge, not repeated spam
-* manual and inferred completion evidence
-* weekend/weekday variants
-
-example:
-
-```text
-08:00–11:00
-- medicine
-- water
-- pets
-- check praxis attention
-```
-
-maven should know what is still missing, not merely fire four timers.
-
----
-
-### 3. cross-device presence
-
-**status (2026-07-20):** the hysteresis engine and 3 of the listed signals are
-already built and wired live: `internal/store/presence.go` (noisy-OR combiner
-+ Schmitt-trigger bucket resolve), fed by `desk_active` (workstation, via
-`scripts/desk-active.sh` posting to `/api/signal`), `page_heartbeat` (mavweb
-tab, `app.js`), and `wg_handshake` (`mavpoll` polling `wg show`) — threaded
-into the tick loop via `internal/loop/gather.go`. Not done: phone-reachable,
-homesrv-available, audio-output, and active-maven-client signals from the
-list below are still missing.
-
-a small presence daemon on each trusted device:
-
-* workstation active/idle
-* phone reachable
-* homesrv available
-* last keyboard/mouse activity
-* wireguard presence
-* current audio output
-* active maven client
-
-mavend receives only compact state, not raw activity logs.
-
-useful for:
-
-* choosing delivery channel
-* suppressing voice while away
-* surfacing reminders when you return
-* knowing whether an agent result should be spoken or sent as text
-
----
-
-### 4. interruption policy — **done (2026-07-20), turned out to already be built**
-
-audited the existing code before writing anything new: `internal/loop.Gate`
-already answers deliver_now vs. drop (quiet-hours/cooldown/snooze/presence/
-calendar-busy), and `cmd/mavend/tick.go`'s `digestQ` + `config.DigestConfig`
-already implement queue/digest (low-severity nudges batch into one
-notification, flushed on window elapsed or max-items reached). The four
-outcomes below were already covered by these two mechanisms; nothing new to
-build for the core policy.
-
-Gap that *was* real: `deploy/mavend.json` had no `digest` block, so batching
-was disabled in prod despite being fully implemented. Fixed — see the config
-change alongside this note.
-
-before delivering anything, evaluate:
-
-```text
-urgency
-current activity
-quiet hours
-recent nudges
-available channels
-whether already surfaced
-```
-
-result:
-
-```text
-deliver_now
-queue
-digest
-drop
-```
-
-this prevents maven from becoming annoying once praxis and other sources start producing more data.
-
----
-
-### 5. entity-aware memory — **done (2026-07-20)**
-
-`03fa52d`/`9876187` (Vikunja #279): facts gain `Subject`/`EntityID`/
-`ResolutionState`; an async enrichment worker resolves free-text subjects to
-canonical Nexus entity_ids (mirrors Praxis's enrichment pattern). Ambiguous
-or unreachable Nexus never guesses — the fact stays `pending` or terminal
-`ambiguous`. Voice-tapped facts (`IntentFact`) now flow into the enrichment
-queue automatically via an optional `Subject` field on `WriteFactReq` (old
-callers unaffected).
-
-Landed alongside this in the same session (not originally on this list, but
-closes the plumbing gaps the last brief flagged for Nexus/Praxis maturity):
-a typed Praxis lifecycle client (`398997f` — surface/acknowledge/resolve/
-ignore/pin; fixes the surfaced≠acknowledged gap where reading an item aloud
-left no trace), correlation-ID/version headers on the Nexus/Praxis clients
-(`b743860`), entity-scoped Praxis attention queries (`0579ef9`), a durable
-delivery outbox with begin-before-send/complete-after semantics
-(`29f23e3`+`9ff726e` — closes a duplicate-send-on-crash bug), fail-closed
-handling on ambiguous IPC mutation outcomes and Nexus/Hexis dependency
-errors (`838fde1`+`d9fa4d6`), and a reusable fake-ecosystem test harness
-with fault injection (`c932cd8`).
-
-connect maven memory to nexus ids.
-
-instead of:
-
-```text
-key = "кошачий фонтан"
-```
-
-store:
-
-```text
-entity_id = ent_pet_water_fountain
-predicate = refilled_at
-value = 2026-07-19T...
-```
-
-benefits:
-
-* stable russian/english aliases
-* fewer duplicate facts
-* better “when did i last…” queries
-* easier routine detection
-* cleaner praxis correlation
-
----
-
-### 6. bounded follow-up state
-
-for short continuations:
-
-* “yes”
-* “tomorrow”
-* “the second one”
-* “not that project”
-* “do it later”
-
-store explicit pending state instead of relying on chat history:
-
-```go
-type PendingInteraction struct {
-    Kind       string
-    Candidates []string
-    Args       json.RawMessage
-    ExpiresAt  time.Time
-}
-```
-
-this matters a lot for a 1.7b model.
-
----
-
-### 7. evaluation lab — **skipped for now (2026-07-20)**
-
-runs on a different machine (GPU box), and CPT is currently in progress
-there — deprioritized until the training pipeline has a checkpoint to gate.
-Not abandoned, just off the immediate list.
-
-before every new checkpoint or lora deploy:
-
-* routing accuracy
-* slot accuracy
-* malformed json rate
-* russian/english mixed input
-* ambiguous entity handling
-* reminder vs note vs fact
-* direct answer vs tool call
-* confirmation safety
-* phrasing quality
-* latency and ram
-
-also replay real anonymized traces against old and new checkpoints.
-
-this should be a hard deployment gate.
-
----
-
-### 8. replayable full-system simulator
-
-fake:
-
-* clock
-* presence
-* caldav
-* telegram
-* praxis
-* nexus
-* hexis
-* stt
-* tts
-* llama-server
-
-scenario:
-
-```text
-08:30 user appears
-08:35 medicine not completed
-08:40 correx agent waits
-08:45 calendar sync stale
-08:50 user says “what did i miss?”
-```
-
-assert:
-
-* what tools were called
-* what was surfaced
-* what stayed unresolved
-* what maven said
-* what was not executed
-
-this will save more time than another feature daemon.
-
----
-
-## useful second-wave additions
-
-### voice session quality
-
-* barge-in
-* interrupt tts on wake word
-* partial stt display
-* confidence-aware clarification
-* retry only failed stt segment
-* per-room microphone profiles
-* noise-floor calibration
-* short response mode when speaking
-
-### notification bridge framework
-
-small adapters for:
-
-* ntfy
-* telegram
-* matrix
-* web push
-* android notification forwarding
-* local dbus notifications
-
-normalize into maven/praxis events instead of treating each as a separate feature.
-
-### local knowledge ingestion
-
-* markdown/docs ingestion
-* git repo summaries
-* project decision records
-* conversation exports
-* provenance and source links
-* incremental reindexing
-
-keep this read-only and separate from personal fact memory.
-
-### service self-diagnostics
-
-`maven doctor`:
-
-* socket reachability
-* model health
-* stt/tts readiness
-* embedder availability
-* caldav freshness
-* telegram poll state
-* praxis/nexus/hexis reachability
-* db integrity
-* disk usage
-* recent failures
-
-### config and secret management
-
-* schema-validated config
-* config migration
-* secret references instead of inline values
-* dry-run validation
-* redacted config dump
-* per-daemon health config
-* startup dependency report
-
----
-
-## things i would not build yet
-
-* autonomous multi-step planning
-* large external reasoner
-* generic workflow engine
-* self-editing memory
-* automatic hexis actions from praxis
-* emotion simulation beyond phrasing
-* full home-assistant replacement
-* more model layers before routing is stable
-
-## recommended order
-
-**status as of 2026-07-20:**
-
-1. ~~evaluation lab~~ — **skipped, GPU-box work, deprioritized while CPT is in progress**
-2. ~~entity-aware memory~~ — **done** (`03fa52d`/`9876187`, plus adjacent
-   Nexus/Praxis plumbing hardening — see item 5 above)
-3. ~~morning routine engine~~ — **core engine done** (`internal/morning` +
-   `cmd/mavend` wiring — see item 2 above; not yet configured on homesrv,
-   no voice query, no web UI)
-4. interruption/delivery policy
-5. presence agents
-6. unified event intake
-7. full-system simulator
-8. notification bridges
-9. knowledge ingestion
-10. voice-session polish
-
-the main goal should be: **maven reliably knows what is happening, knows what you meant, and chooses the least annoying correct response**. everything else can wait.
-

AGENTS.md

@@ -18,7 +18,7 @@ live in sibling repos next to this one.
 
 Division of labour: Nexus identifies, Praxis observes, Hexis acts, Maven understands
 and coordinates. Maven is not the source of truth for any of the three. The full
-contract is `MAVEN_ECOSYSTEM_ARCHITECTURE.md`, and the constraints that bite during
+contract is `docs/ecosystem.md`, and the constraints that bite during
 implementation are summarised in `CLAUDE.md`.
 
 Where things are in this repo:

CLAUDE.md

@@ -10,7 +10,7 @@ compose passes `/dev/dri` + the render gid) — the resident model stays ≤1.7B
 **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 `MODEL-BAKEOFF-31-07-2026.md`. It is a Thinking variant, so `n_ctx` is 4096
+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.
 
 The **target** is still the locally CPT'd **Qwen3-1.7B** (Vikunja #122, training in flight).
@@ -25,9 +25,24 @@ Spanish. Their strong published IFEval/BFCL numbers are English-only. Model file
 `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`.
 
-See `REARCH.md` for the target architecture, `DESIGN.md` for the folded design spec, and
+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.
 
+**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. 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.
+
+Both halves are wired as of 2026-08-03. Routing and replies prefer the workstation silently
+through `modelSeam`; nudge and reminder phrasing prefer it silently inside the phraser. A
+world question goes through `LLMPhraser.PhraseWorld` and names the gap when the card is not
+free — `worldGap` in `cmd/mavend/worldmodel.go`, which he hears instead of an invented
+answer. A box with no `workstation` block behaves exactly as it did before the seam: naming
+a gap requires a gap. The offload table in `docs/offload.md` says which caller is which.
+
 ## Build & test
 
 CGO daemons (`mavend`, `mavsttd`, `mavttsd`, `mavenclient`) need the vendored toolchain
@@ -72,7 +87,7 @@ protocol; the config in `deploy/mavend.json` (with `${VAR}` env expansion from g
 
 Maven is one of four services. It owns conversation and personal memory. It does not
 own identity, operational state, or execution. Full contract in
-`MAVEN_ECOSYSTEM_ARCHITECTURE.md`.
+`docs/ecosystem.md`.
 
 ```text
 Nexus identifies.  Praxis observes.  Hexis acts.  Maven understands and coordinates.
@@ -113,7 +128,7 @@ Every cross-service call carries a correlation id minted once per action
 on in deploy** — this section used to say it was wired `nil`, which stopped being true on
 2026-07-31.
 
-- **LLM router (the intended design, REARCH.md):** the resident Qwen3-1.7B (`llmrouter.go`)
+- **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`
@@ -127,13 +142,23 @@ on in deploy** — this section used to say it was wired `nil`, which stopped be
 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.
 
-Measured on the 77-case RU fixture (`MODEL-BAKEOFF-31-07-2026.md`): the classifier scores
-36.8% full accuracy at p50 31ms; Qwen3-1.7B scores 67.5% intent-only / 72.7% through the
-cascade at p50 ≈825ms. Accuracy roughly doubled, latency is ~27× worse, and that trade was
-accepted deliberately. **The ≈2.7s figure that stood here until 2026-08-02 was contention,
-not the model.** See `ROUTING-EVAL-31-07-2026.md` line 61, which measures the LLM router at
+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. `Confidence: 1.0` used to be hardcoded in `llmrouter.go`, so the LLM
+work off the bakeoff table.
+
+**The numbers above are the homesrv floor, not the ceiling.** With the workstation up, routing
+completes through `llm.Pair` against gemma-4-12b and scores **84.4% full / 93.5% intent-only at
+p50 329ms** — better than the resident model and about 2.5× faster (`docs/evals/2026-08-02-workstation-gemma4-12b.md`,
+Vikunja #485). The workstation is never assumed up, so both sets of numbers are live. Judge a
+routing change against the classifier and the resident model, since those are what always answer.
+
+`Confidence: 1.0` used to be hardcoded in `llmrouter.go`, so the LLM
 path could never ask for clarification (6/6 refusal cases missed on the fixture) — Vikunja
 #359. Fixed 31-07-2026 with structural signal (single-token utterance, keyless fact, act with
 no allowlisted fn) feeding the same stage-3 gate the classifier path already had — see

Makefile

@@ -16,11 +16,11 @@ PIPER_BIN     := $(shell pwd)/deps/piper/piper
 PIPER_MODEL   := $(shell pwd)/models/tts/ru_RU-irina-medium.onnx
 PIPER_ESPEAK  := $(shell pwd)/deps/piper/espeak-ng-data
 
-.PHONY: simulate stt-fixtures test-stt-golden all build build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav clean test fmt-check vet run-stt run-tts run-web download-embedder deps-go eval-router eval-recall eval-phrasing eval-models
+.PHONY: simulate stt-fixtures test-stt-golden all build build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav clean test fmt-check vet run-stt run-tts run-web download-embedder deps-go eval-router eval-recall eval-phrasing eval-models build-gpud
 
 all: build
 
-build: build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav build-mail build-update
+build: build-stt build-tts build-daemon build-client build-waked build-web build-poll build-caldav build-mail build-update build-gpud
 
 build-stt:
 	CGO_CFLAGS="$(CGO_CFLAGS)" CGO_LDFLAGS="$(CGO_LDFLAGS)" LD_LIBRARY_PATH="$(shell pwd)/deps/lib" \
@@ -59,6 +59,12 @@ build-mail:
 build-update:
 	$(GO) build $(GOFLAGS) -o mavupdate ./cmd/mavupdate/
 
+# mavgpud runs on the workstation, not here. It is built with the rest so a
+# broken supervisor is caught by `make build` on homesrv rather than by the
+# workstation refusing to serve. Copy the binary over, do not `make deploy` it.
+build-gpud:
+	$(GO) build $(GOFLAGS) -o mavgpud ./cmd/mavgpud/
+
 run-web: build-web
 	./mavweb -addr :9200 -voice 127.0.0.1:9100
 
@@ -78,7 +84,7 @@ deps-go:
 	done
 	$(GO) version
 
-# fmt-check fails if any file needs gofmt. DESIGN.md has always said `make
+# fmt-check fails if any file needs gofmt. docs/design.md has always said `make
 # test` gates on gofmt and vet; it did not, so nine files quietly drifted.
 # Run `gofmt -w` on whatever this prints.
 fmt-check:
@@ -191,7 +197,7 @@ deps-piper:
 # multilingual-e5-small: an asymmetric retrieval model. It is trained to match
 # a short question against a longer passage, which is what note recall is.
 # The quantized file is the one we download, deploy and measure — see
-# RECALL-EVAL-31-07-2026.md.
+# docs/evals/2026-07-31-recall.md.
 EMBEDDER_DIR := $(shell pwd)/models/embedder/multilingual-e5-small
 EMBEDDER_MODEL_URL := https://huggingface.co/Xenova/multilingual-e5-small/resolve/main/onnx/model_quantized.onnx
 EMBEDDER_TOKENIZER_URL := https://huggingface.co/Xenova/multilingual-e5-small/resolve/main/tokenizer.json

PROGRESS.md

@@ -1,468 +0,0 @@
-## Maven — current state (updated 2026-07-20)
-
-### Session 2026-07-20 — ecosystem hardening + entity-aware facts
-
-Ten commits, focused on closing the Nexus/Praxis integration gaps flagged
-as "wired but immature" in the prior review, plus the entity-aware-memory
-backlog item (`20-07-2026-BACKLOG.md` item 5).
-
-- **Entity-aware fact resolution (Vikunja #279)** — facts gain
-  `Subject`/`EntityID`/`ResolutionState`; an async worker resolves
-  free-text subjects to canonical Nexus entity_ids (mirrors Praxis's own
-  enrichment pattern). Ambiguous/unreachable Nexus never guesses — stays
-  `pending` or terminal `ambiguous`. Voice-tapped facts (`IntentFact`) flow
-  into the queue automatically via an optional `Subject` field on
-  `WriteFactReq` (old callers unaffected, no signature break).
-- **Typed Praxis lifecycle client (Vikunja #271)** — `GetItem`/`Search`/
-  `Surface`/`Acknowledge`/`Resolve`/`Ignore`/`Pin`, routed through new RU/EN
-  dialogue verbs. Fixes a real lifecycle-invariant bug: reading an
-  attention item aloud now calls `Surface` — previously the digest path
-  read items without recording that they'd been surfaced, so "Maven
-  mentioned it" was indistinguishable from "never came up."
-- **Durable delivery outbox (Vikunja #270)** — `BeginDeliveryAttempt`
-  before `Send`, `CompleteDeliveryAttempt` after; a stale `pending` row
-  found at startup reconciles to `unknown` (never silently resent or
-  dropped — same rule as Hexis's execution-timeout handling). Closes a
-  crash-window duplicate-send bug. Wired into `DispatchNudge`,
-  `DispatchReminder`, `RepeatUnacked`; reconciliation runs once at boot
-  before the tick loop resumes.
-- **Fail-closed IPC/dependency handling (Vikunja #269, #272/#273)** —
-  ambiguous mutation outcomes (frame sent, reply lost) no longer blindly
-  retry; Nexus/Hexis dependency errors fail closed instead of guessing.
-- **Correlation IDs + version headers (Vikunja #273)** — the hand-rolled
-  Nexus/Praxis HTTP clients now send `X-Nexus-Version`/`X-Praxis-Version`
-  and thread the same correlation ID already generated in
-  `executeCapability` through the whole call chain, matching the Hexis
-  client's existing behavior.
-- **Entity-scoped Praxis attention queries** — callers holding a resolved
-  entity_id can ask "what needs attention for this entity" directly
-  instead of filtering the unscoped list client-side.
-- **Fake-ecosystem test harness with fault injection** — a reusable
-  `fakeServer` (Nexus/Praxis/Hexis fixtures, runtime-toggleable
-  `SetFault`, fake clock) replacing ad-hoc per-test `httptest` servers;
-  covers a gap that had zero test coverage (`handlePraxisAct`) and adds a
-  fault-then-recovery regression test for the fail-closed fixes above.
-- **Ops fix** — `deploy/mavend.json`'s phraser was pointed at a 4B model
-  with `n_gpu_layers=99`, which OOM'd under memory pressure and left a
-  zombie `llama-server` child; swapped to the 2B Qwen model matching the
-  intended resident-model size.
-
-Net effect: the Nexus/Praxis wiring described as "plumbing exists, thin
-compared to Maven's test depth" in the prior review is now materially
-hardened — typed clients, fail-closed error handling, durable delivery,
-and a proper fault-injection test harness are all in place. Evaluation lab
-(`20-07-2026-BACKLOG.md` item 7) is explicitly skipped for now — it runs
-on the GPU box, which is occupied by CPT. Morning routine engine (backlog
-item 3) is next up, not started.
-
----
-
-> **Resolved 2026-07-30 (task #318).** The resident checkpoint is
-> **Qwen3.5-0.8B** (`Q4_K_M`), set in `deploy/mavend.json`; the **target** is
-> the locally CPT'd **Qwen3-1.7B**, still training (#122). Older model claims
-> below — the LFM references, the pipeline line, and the "swapped to the 2B
-> Qwen model" ops entry above — are historical. Read them as a log of what was
-> true at the time, not as current fact. Note also that `/mnt/hdd1/llms` is
-> bind-mounted over `models/llm/`, so the LFM2.5 gguf in the repo tree is
-> never loaded.
-
-Architecture decision (as written on 2026-07-20): the target resident
-router/phraser is the locally trained Qwen3-1.7B model — still the target as
-of 2026-07-30. Older LFM references below describe the then-deployed
-historical stack, not the target checkpoint. RU CPT has a successful
-full-weight checkpoint at step 1000/8077; evaluation and Qwen3 SFT tooling are
-tracked in `docs/plans/2026-07-18-qwen3-resident-training-eval.md`.
-
-Consolidated status. The reactive↔proactive core is closed and testable through
-the web PWA. The former SPEC's open items 1–7 (now `DESIGN.md` § execution ledger) are landed (protocol doc, away-channel
-fallthrough, CalDAV poller, quiet-hours schedule, tools enable/disable, note RAG,
-passkey step-up); item 8 (multi-user) is deliberately deferred — see the tail.
-The two big infra gaps from the jul5 revision are closed on `overnight-jul5`:
-**at-rest encryption** (AES-256-GCM, tmpfs working copy — not sqlcipher, see
-`internal/store/crypt.go`) and **Docker deployment** (one image, six daemon
-containers). The `overnight-jul6` session (now on `master`) closed the biggest
-*query-surface* gaps — **calendar querying, general-knowledge answers, and
-weather** — plus a populated homelab act allowlist and two pure scaffolds
-(dialogue state, long-term-memory vector store). ~15.2k LOC + ~8.5k test, 303
-tests, `-race` in `make test`.
-
-### Access model
-
-- **Phone** → needs the wg tunnel to reach homesrv (no homesrv DNS otherwise;
-  raw IP or a DNS tweak can bypass, not the default).
-- **PC** → uses homesrv DNS, resolves the domains over local-net, **no wg needed**.
-- nginx + ufw both scope to `10.42.0.0/24` (wg) + `192.168.1.0/24` (LAN), deny all else.
-- **Surface in use now: the web PWA (`mavweb`).** Voice PTT + in-app nudges both ride it.
-
-### Works end-to-end (tested)
-
-- **Reactive voice:** PWA record → Whisper STT (`mavsttd`) → ONNX classifier →
-  resident phraser (llama-server subprocess; Qwen3.5-0.8B as of 2026-07-30 —
-  this line historically named "LFM 2.5-1.2B") → Piper TTS
-  (`mavttsd`) → reply.
-  HTTP POST path (mobile-Chrome drops WS for the audio).
-- **Capture:** `fact` (EN **and RU** — root-substring recognizers) + `reminder`
-  persist through CoreAPI (`source=tap:voice`). This is the substrate the care
-  rules read.
-- **Notes / query (semantic recall, sqlite — no chroma):** `note` → embed (the
-  classifier's ONNX embedder) → `notes` table. `query` → embed → brute-force
-  cosine top-k → confidence-gated (below `queryMinScore` 0.55 ⇒ "no note", not a
-  guess). **Note RAG (SPEC item 6):** the gated top-k feed the phraser
-  (`PhraseQuery`) to compose a natural answer ("вот что я нашла: …") instead of
-  a verbatim dump; raw-notes fallback on any LLM error. Stub is deterministic.
-- **Monitoring (`/dash`):** mavweb server-renders presence + recent nudges (by
-  outcome) + recent facts from the append-only store via CoreAPI. Read-only,
-  meta-refresh, no JS.
-- **Proactive loop:** 60s dumb ticker, pure predicates over a State snapshot,
-  universal gate (quiet-hours/presence/cooldown/snooze/calendar), one-nudge-per-
-  tick max-severity, reminders (gate-bypassing), sev4 repeat-til-ack, feedback
-  auto-tuner (outcome ratio → bounded cooldown, persisted as `source=feedback`).
-- **Rules:** water/meal/break (sev1–2 care), service_down (sev4, `poll:uptimekuma`),
-  netdata_critical (sev3, `poll:netdata`).
-- **Routines (`internal/routine`):** operator-declared clockwork — the third
-  proactive class beside reminders (user-stated) and care rules (world-state).
-  Config `routines[]` (cron + literal RU body + severity) fire through the normal
-  dispatcher on schedule (an 08:00 briefing, a 22:00 wind-down). Bodies are
-  literal (not LLM-phrased ⇒ can't hallucinate); rule name `routine:<name>` so
-  they don't pollute the care autotuner; cold-start guard seeds on first sight so
-  a restart never replays a missed schedule. Pure `routine.Due`, unit-tested; the
-  tick driver holds the last-fired map.
-- **Env facts (`mavpoll`):** netdata alarms → `netdata_alarm` (fires immediately
-  on a real CRITICAL); kuma monitor_status → `service_down`. Writes only on
-  value-change (no append-only churn).
-- **Presence:** noisy-OR decay + Schmitt hysteresis. Live via `page_heartbeat`
-  (PWA auto-pings `/api/signal` every 30s → present when a tab's open).
-- **Delivery:** ntfy / telegram / voice by `f(severity, presence)`; minimal body
-  on away channels. PWA subscribes to ntfy over **WebSocket** for in-app nudges.
-- **Away-channel fallthrough (SPEC item 2):** when the router picks voice but no
-  live session exists at push time (presence guess was wrong), the dispatcher
-  reroutes through the AWAY table — sev3→ntfy, sev4→telegram-repeat-til-ack,
-  sev≤2→drop — instead of silently dropping. Covers nudges + reminders.
-- **Calendar busy (SPEC item 3, `mavcaldav`):** new poller queries a self-hosted
-  **Radicale** CalDAV server on an interval, writes `calendar_busy` + event facts
-  through CoreAPI (value-change only). The loop gate already consumes `calendar_busy`.
-- **Quiet-hours schedule (SPEC item 4):** the gate reads `quiet_hours`; a config
-  time window (`voice.quiet_hours`, HH:MM, midnight-crossing handled) now sets it
-  on each tick — in addition to the "тихий режим" voice toggle. Both activate quiet.
-- **Client protocol (SPEC item 1):** the voice wire format (length-prefixed JSON
-  frames) is published in `PROTOCOL.md`, generated from `internal/voice/wire.go`
-  so third-party clients don't need the Go source.
-- **Passkey step-up (SPEC item 7):** `internal/webauthn` does real WebAuthn —
-  ES256/P-256 register + assert, ecdsa signature verification, rpIdHash + UP/UV
-  flag binding (UV = the gesture), sign-count regression check. `PasskeySession`
-  bumps the auth session L2→L3 for a TTL on assert. mavweb serves `/auth/passkey`
-  (enroll + step-up) + the begin/finish endpoints. Crypto is round-trip tested
-  (incl. tampered-sig / missing-UV / wrong-origin negatives).
-- **Stability:** llama-server orphan leak fixed (`Pdeathsig` kills the child on
-  any mavend death); `kill-maven.sh` reaps strays (matches the model, not a
-  bogus `llama-server.*maven` pattern); `start-maven.sh` wires `-core` + poller.
-
-### Wired but needs a deploy action (not code)
-
-- **`desk_active`** (strongest presence signal) — `scripts/desk-active.sh` runs
-  on the **desk PC** (hypridle-gated systemd timer), posts over wg to mavweb.
-- **`mavwaked`** (always-on listening) — needs a systemd user unit on a client
-  box (desk PC, pi, etc.) where the mic is attached. Connects to mavend over wg
-  or local net via `-addr`. Deferred until a client box is wired with a mic.
-
-Caveats / gotchas:
-- **desk_active is a workstation deploy, not code** — 0 facts ever written; presence
-  runs on page_heartbeat alone (dash reads "away"/"never at desk"). `scripts/desk-active.sh`
-  + a hypridle-gated `maven-desk` timer must be installed on the desk PC (not homesrv).
-- **Notes recall needs the ONNX embedder** — under the HashEmbedder floor, cosine is
-  lexical (token overlap), not semantic; scores are low, so most RU commands sit under
-  the 0.35 route threshold and clarify. Configure `voice.embedder` for confident recall+routing.
-  (The floor now at least tokenizes Cyrillic — see below — so it ranks correctly, just weakly.)
-- **Switching the embedder model silently breaks old notes** — different dim ⇒
-  cosine 0 ⇒ they stop matching; brute-force can't re-embed. Re-embed on a model change.
-- **`wg_handshake` is OFF and should stay off** — in this topology the phone only
-  runs wg when *outside*, so a fresh handshake means AWAY, not here. The `mavpoll
-  -wg` flag exists (defaults `""`) and could later back the spec's "away override"
-  by flipping the sign; as a presence-*here* signal it's inverted. desk_active +
-  page_heartbeat cover home presence.
-- **Cold-start unlock tests are missing** — the key wrap/unwrap code
-  (`internal/webauthn/keywrap.go`) and locked-mode IPC gating (`cmd/mavend/main.go`)
-  are correct but have **zero test coverage**. The roadmap (item 2.1) required
-  three new test cases (wrap/unwrap round-trip, wrong-cred unwrap fails,
-  locked-mode IPC rejects non-unlock methods); none were written. `make test`
-  is green by omission. Write these before relying on the cold-start path with
-  real keys.
-
-### Done since last revision (overnight-jul6, 2026-07-06)
-
-Seven tasks (session board `SESSION-06-07-2026.md`, deleted 2026-07-30 — see git history), one commit each, merged to `master`.
-This session was run through **opencode**, not Claude Code (co-author trailer).
-
-Since then (**2026-07-06, second session**):
-
-- **Always-on listening (gap 1, MVP)** — `cmd/mavwaked/`: 825 lines, 10 `-race`
-  tests. Energy-based VAD over 30ms windows (same RMS threshold as mavsttd's
-  `gateReason`), adaptive noise floor, speech→silence state machine. Captures
-  PCM from arecord(1) subprocess, sends `PushToTalk` with `Surface=SurfaceVoice`
-  (L0 — no destructive acts). Reply plays through aplay(1). No wake word yet
-  (pure VAD trigger); the 30ms frame shape matches silero-vad ONNX input 1:1,
-  so swapping energy-threshold for ONNX inference is a local change in vad.go.
-  `Makefile` `build-waked` target. Runs on client boxes (not docker/homesrv)
-  via systemd user unit; connects to mavend over wg or local net.
-
-Since then (**2026-07-06, third session** — roadmap execution agent):
-
-- **Cold-start unlock (ROADMAP 2.1)** — the at-rest AES key is now wrapped
-  (HKDF-SHA256 + AES-256-GCM, stdlib-only — no `x/crypto` dep) with the passkey
-  credential's public key and persisted to disk. At boot, if a wrapped key file
-  exists AND no env key is set, mavend starts **locked**: the IPC server runs
-  but `srv.Check` rejects everything except `MethodAssertStepUp` +
-  `MethodUnlock`. A passkey assertion at `/auth/passkey` calls `MethodUnlock`
-  with the credential's public key → unwraps the blob → opens the store → wires
-  voice/loop/delivery → `srv.SetAPI` swaps the locked stub for the real
-  CoreAPI. mavweb's `RegisterFinish` wraps the env key on enrollment;
-  `AssertFinish` calls `Unlock` on assertion. Env-key fallback preserved
-  (dev/CI path unchanged). **Test gap:** the roadmap required three new test
-  cases (wrap/unwrap round-trip, wrong-cred unwrap fails, locked-mode IPC
-  rejects non-unlock methods) — none were written. The code is correct but
-  untested; `make test` is green by omission, not coverage.
-- **Conversation depth (ROADMAP 3.2)** — cross-intent anaphora + fact-by-key
-  lookup. `AnaphoraResolver` in `router/slots.go` detects RU pronouns
-  (это/он/она/оно/тот/мой + inflected forms). `followUpMerge` now handles
-  three cases: same-intent slot inheritance (existing), cross-intent anaphora
-  (Query/Fact/Reminder after a Fact with a pronoun inherits the prior key +
-  time), and query-after-fact (a query following a fact inherits the key for
-  fact-by-key lookup). `Session.History []Turn` added as the multi-turn
-  scaffold (capped at 4). 7 new test cases including the exact done-when
-  scenarios (anaphora query-after-fact, three-turn break, explicit-key-wins).
-- **Routing quality + persona (ROADMAP 4.1/4.4)** — `QueryMinScore` is now a
-  config knob (`voice.query_min_score`, default 0.55) instead of a hardcoded
-  const. `make download-embedder` fetches Xenova/paraphrase-multilingual-
-  MiniLM-L12-v2 (~90MB ONNX) + tokenizer; AGENTS.md documents the embedder +
-  libonnxruntime setup. `Persona` field in `VoiceConfig` prepends to every
-  LLM system prompt (nudge phrasing, note queries, general knowledge); empty
-  = current hardcoded feminine-gendered Russian persona. Also fixed two
-  pre-existing data races found by `-race`: `voice/server.go` wg.Add vs
-  wg.Wait (accept mutex), `mavweb/server.go` s.api field (atomic.Value).
-
-- **Calendar querying (task 3)** — "что у меня завтра?" now answers from the
-  CalDAV facts the poller already writes. Added `store.CalendarEvents(from,to)`,
-  a RU date-scope parser («сегодня»/«завтра») in `router/slots.go`, and an
-  IPC `CalendarEvents` RPC (api/client/server/wire) feeding the `IntentQuery`
-  handler. Empty day → «на сегодня ничего нет». Previously calendar only *gated*
-  nudges; it's now queryable.
-- **General-knowledge routing (task 4)** — when notes-RAG misses `queryMinScore`,
-  the query now falls through to the phraser with an anti-hallucination system
-  prompt (`router.KnowledgePrompt`, single tested source) instead of giving up.
-  Empty/errored/Stub phraser → «не знаю.», never a fabrication.
-- **Weather (task 5)** — new `internal/weather/`: `Provider` interface, a stub
-  («погода не настроена»), and a real **keyless Open-Meteo** provider (geocode +
-  current_weather, injectable `*http.Client`, mocked in tests — no live network).
-  Wired into `IntentQuery` (keywords погода/градус/температура) with a ~5s
-  context timeout; selected by `voice.weather.provider` ("open-meteo" | "" → stub).
-- **Homelab act allowlist (task 2)** — `voice.tools` seeded with read-only acts
-  (`systemctl status`, `docker ps`, `uptime`, `df`, `free`, `journalctl` reads)
-  as `destructive:false` and mutating ones (restart/stop/start/reboot,
-  docker-restart/stop) as `destructive:true`. Guardrail verified: no dangerous
-  verb is `destructive:false`. RU phrasings seeded in `act.txt`.
-- **Embedder config validation (task 1)** — a partially-filled `voice.embedder`
-  block (some of model/tokenizer/lib paths missing) is now a load error instead
-  of a silent fall-through to the Hash floor; the floor fallback logs explicitly.
-- **Dialogue state scaffold (task 6)** — `internal/dialogue/`: `Session` +
-  TTL `SessionStore` + pure `InheritSlots`. **Now wired** (post-merge follow-up):
-  the voice handler carries slots across same-intent turns within a 2-min window
-  (`followUpMerge`, unit-tested) — bounded gap-filling, not full multi-turn yet.
-- **Long-term memory interface (task 7)** — `internal/memory/`: `Store` interface
-  + `InMemoryStore` (cosine). Wired into `IntentNote` (best-effort insert) and,
-  post-merge, into `IntentFact` (facts indexed) + `IntentQuery` (read-back after
-  notes-RAG misses). In-memory only — no persistent backend yet (gap #8).
-
-Follow-ups (Claude Code, post-merge): gofmt'd `handlers_test.go` (the jul6
-verification commit left it misaligned, so `gofmt -l` still flagged it despite the
-"all gates green" claim); deduped the task-4 knowledge prompt to the single tested
-`router.KnowledgePrompt()`. Tree is now genuinely green (gofmt/vet/303 tests).
-
-### Done since the jul5 revision (overnight-jul5, 2026-07-05)
-
-The overnight session (`SESSION-05-07-2026.md`, deleted 2026-07-30 — see git history; 25 tasks) closed the previous
-"not built yet" items 1–3 and added feature depth:
-
-- **At-rest encryption** — the on-disk db is AES-256-GCM ciphertext; the daemon
-  works on a tmpfs (RAM) plaintext copy, sealed back atomically on close. Wrong
-  key / tamper ⇒ fail closed, never a plaintext fallback. Legacy plaintext dbs
-  upgrade on first clean shutdown. Key via config/env (`db_key_env`); no KDF —
-  raw 32-byte key, base64. The passkey cold-start unlock plugs into the same
-  `store.OpenEncrypted` seam later.
-- **Docker deployment** — single image, one container per daemon
-  (`docker-compose.yml`); only mavend mounts the key + db volume; IPC over a
-  shared socket volume. `ipc.DialWait` (boot-order tolerance) + redial-on-drop
-  (core restarts don't kill modules). `deploy/README.md` has the runbook.
-- **Tests** — mavcaldav, mavttsd, voicesink, mavweb main/handlers covered;
-  `make test` runs `-race -coverprofile`.
-- **Recurring reminders** — `cron` + `next_fire_ts` on reminders; recurring ones
-  reschedule (instead of mark-fired) after successful delivery.
-- **Notification digest/batching** — low-severity nudges queue and flush as one
-  digest per window/max-items (`digest` config block); stale-reminder bursts on
-  boot collapse into a single digest reminder, completed only after delivery.
-- **Rule trace engine** — `ExplainTick`/`ExplainGate` record per-rule
-  predicate/gate/selection results each tick; served over IPC (`tick_trace`)
-  and rendered at mavweb `/trace` ("why didn't she nudge me").
-- **Web UI** — new `/history` (facts + revert buttons), `/notifications` (nudge
-  history), `/trace` pages; nav links on `/dash`; RU/EN cheatsheet toggle in the
-  PWA; manifest icons (`icon.svg`). POST `/tools` now requires an in-process
-  passkey step-up when WebAuthn is configured.
-- **Revert/undo** — `RevertFact` voids the latest fact for a key (append-only
-  void-marker, audit trail intact); exposed at `/api/revert` from `/history`.
-- **Tool scopes** — `scope` column on tools, threaded through propose/enable/UI.
-  `DisableTool` raised to AuthStepUp alongside Enable.
-- **Passkey persistence** — mavweb credentials in a JSON file (`-passkey-file`),
-  surviving restarts; rollback-on-persist-failure keeps memory and disk in sync.
-- **STT silence gate** — min-duration + RMS floor drop non-speech before whisper
-  hallucinates on it (`-min-ms`, `-silence-rms` flags on mavsttd).
-- **Housekeeping** — `db_key.env` gitignored (+`.env.example`), `build-caldav`
-  target, zero-timestamp "never" fix on /dash.
-
-### Not built yet (ranked by ROI)
-
-1. **Multi-user (SPEC item 8)** — deliberately deferred, see the tail.
-
-Closed (jul6 follow-ups): `/api/revert` now sits behind the same passkey
-step-up as POST `/tools`; `go.mod` direct deps (`onnxruntime_go`,
-`coder/websocket`, `robfig/cron`) are labeled correctly — `go mod tidy` can't
-run here because it walks the vendored `deps/go` toolchain tree.
-Purge+rotate leaked db key (#12) — investigated and closed: the key was
-**never committed** to git history (gitignored at introduction, no commit
-ever tracked `deploy/db_key.env`), so nothing to scrub. File stays on disk
-and in deploy env by design — at-rest encryption needs it at boot.
-
-Done earlier (2026-07-03): **act tool executor, store-backed, full flow**
-(`internal/tool` + `internal/store/tools.go` + `tools` CoreAPI methods).
-- **Execution:** IntentAct runs the matched fn against the store's ENABLED
-  allowlist. argv, no shell → STT text can't inject. Live store read, so a
-  newly-enabled tool runs without a daemon restart.
-- **proposed→enabled→disabled (SPEC item 5):** an act whose verb isn't enabled is
-  scaffolded as a `proposed` tool (maven suggests). A human enables it (fills argv
-  + destructive) on the authed **`mavweb /tools`** page — never voice — and can
-  disable it back to `proposed` (kept in the store, won't run). `EnableTool`/
-  `DisableTool` sit at `AuthStepUp`; the gate is now **live** via `PasskeySession`,
-  so /tools enable requires a passkey assertion at `/auth/passkey` first.
-- **Confirm turn:** a destructive enabled tool replies "выполнить X? да/нет" and
-  parks; the next utterance (ru/en yes-no) confirms or cancels (90s TTL).
-- **Config:** `voice.tools` seeds enabled tools at boot (editing mavend.json =
-  the human enable act); mavweb enables ad-hoc ones on top.
-- **Russian:** fixed grammar in reply strings + seed files; maven's self-
-  reference is feminine ("she") — [[maven-persona-gender]].
-
-Also fixed:
-- **HashEmbedder was blind to Cyrillic** (`tokenize` iterated bytes, kept only
-  `a-z0-9`) → every RU utterance embedded to the zero vector → cosine 0 across
-  all intents → misrouted to `act` (alphabetical tie-break). Now rune-based
-  (`unicode.IsLetter`). This was the real cause of "Найди заметку" (a query)
-  landing in `notes`; added note-retrieval query seeds too.
-- **Notes are now browsable on `/dash`** — `RecentNotes` plumbed through the
-  store + CoreAPI; voice-captured notes were previously only reachable via
-  semantic `query`.
-Earlier: notes/query recall, `/dash` monitoring, `wg_handshake` poller (NO-OP).
-
-### Gaps — why "voice assistant" is still aspirational (2026-07-06)
-
-What separates Maven today from the thing the spec describes. Dealbreakers
-first — these define the category:
-
-1. **Always-on listening is code-complete (MVP).** `cmd/mavwaked` captures
-   PCM from arecord → energy-based VAD → PushToTalk with `Surface=SurfaceVoice`
-   (L0). Gap narrowed: no wake word yet (pure voice-activity trigger; every
-   utterance fires). The 30ms frame shape and 16kHz PCM match silero-vad's
-   ONNX input exactly, so a wake-word model swap is a local change in vad.go.
-   Hardware: the mic lives on a client box (desk PC, pi, etc.) — never the
-   homesrv. Deploy action: systemd user unit on whichever box has the mic,
-   connects to mavend over wg or local net.
-2. **Conversation is deeper now, still not full dialogue.** The router
-   classifies one utterance → one reply, but `internal/dialogue` carries
-   context across turns: a 2-min session inherits slots for same-intent
-   follow-ups («напомни завтра» → «…позвонить маме»), and cross-intent
-   anaphora («запиши что я пил воду» → «когда я это сделал?») now resolves
-   RU pronouns (это/он/она/оно/тот/мой + inflections) to the prior turn's
-   key for fact-by-key lookup. `Session.History []Turn` is the scaffold for
-   real multi-turn. Still missing: LLM-driven dialogue manager (decide
-   ask-vs-act), anaphora beyond RU pronouns, single-slot session (single-user
-   box). The sub-1B phraser only words replies.
-3. **Latency/shape of a turn.** Clip-based STT (record → upload → whisper →
-   route → phrase → piper → play). No streaming either direction, no barge-in;
-   every exchange is a full round trip.
-
-Capability-class gaps — built but thin:
-
-4. **Act surface is a small argv allowlist.** propose→enable works and the
-   allowlist now ships a homelab starter set (jul6 task 2 — status/ps/uptime/
-   df/free/logs read-only, restart/stop/reboot gated). Still bounded to what's
-   seeded; broadening it is config, not code.
-5. **Query answers now cover notes + calendar + weather + general knowledge**
-   (jul6 tasks 3/4/5). Calendar querying, keyless Open-Meteo weather, and a
-   phraser knowledge-fallback all landed; caveat — general-knowledge quality is
-   only as good as the sub-1B phraser, and weather needs `voice.weather.provider`
-   set. The cheatsheet and router are now roughly aligned.
-6. **Routing quality depends on the ONNX embedder being configured** — the
-   HashEmbedder floor makes RU recall lexical/weak; many commands fall to
-   "clarify". `make download-embedder` now fetches the multilingual MiniLM
-   model + AGENTS.md documents libonnxruntime setup; `voice.query_min_score`
-   is a config knob (default 0.55) so the floor can be tuned without recompile.
-7. **Presence is effectively one signal** (page_heartbeat); desk_active is
-   still an undeployed script — "voice when near" routing runs on a guess.
-8. **Long-term memory is now persistent (store-backed), not the spec's chroma.**
-   `internal/memory` has a `Store` interface; the daemon now wires
-   `store.MemoryStore` (`internal/store/memory.go`) — a **persistent** backend
-   in the **same encrypted sqlite db** (survives restarts; recall text inherits
-   at-rest encryption, so no plaintext sidecar). Vectors are float32 blobs,
-   search is brute-force cosine (fine at single-user scale; ANN is the later
-   swap behind the same interface). Notes **and facts** are indexed on capture;
-   `IntentQuery` reads it back (after notes-RAG misses, before general-knowledge)
-   — fact recall («когда я пил воду?») is its distinct payoff. The in-memory
-   impl remains the test/no-store floor. Remaining: an ANN/external index is
-   optional-scale, not a gap. Custom TTS voice (kami-picked, replaces the irina
-   floor — [[custom-voice-training]]) is still a future item.
-
-Ops footnote: voice-over-web verified 2026-07-06 — mavend binds 0.0.0.0:9100
-and mavweb reaches it cross-container at mavend:9100 (nc -z confirmed).
-mavpoll uses network_mode=host to reach localhost services (netdata, kuma).
-
-### Future / logged, not now
-
-Custom TTS voice training (kami-picked voice, replaces irina floor); listening
-modes 2–3 (meeting-record, ambient-derive).
-
-### Services & layout
-
-- `mavend` (core, IPC unix socket) — store + loop + phraser; the only key-holder.
-- `mavsttd` / `mavttsd` — STT/TTS worker modules (unix sockets).
-- `mavweb` — PWA bridge (HTTP), `/api/ptt` voice, `/api/signal` presence ingest,
-  `/api/ntfy` WS-subscribe config, `/dash` read-only monitoring.
-- `mavpoll` — env poller (netdata/kuma → facts via CoreAPI).
-- `mavcaldav` — CalDAV poller (Radicale → `calendar_busy` + events via CoreAPI).
-- All behind wg + nginx deny-all; no phone-home. CGo only in `mavsttd`.
-- Start/stop: `./start-maven.sh [build]`, `./kill-maven.sh`.
-- Config: `~/.config/maven/mavend.json` (or `mavend.json` in repo root).
-
-### Key files
-
-- `cmd/mavend/{main,tick,voice}.go` — daemon wiring, loop driver, voice handler
-- `internal/loop/{loop,rules,gather,feedback}.go` — proactive engine
-- `internal/store/` — append-only facts/reminders/nudges/presence/notes
-- `cmd/mavweb/{main.go,dash.html}` — PWA bridge + `/dash` monitoring
-- `internal/router/{classifier,slots,stage0}.go` — reactive routing + slot parse
-- `internal/delivery/` — dispatcher + ntfy/telegram/voice sinks
-- `internal/auth/` — scope/gate/policy; `FloorEnrollment` (same-uid = device
-  trust) + `webauthn.PasskeySession` (real step-up for L3)
-- `internal/webauthn/`, `cmd/mavweb/webauthn.go` — passkey register/assert
-- `cmd/mavcaldav/`, `cmd/mavpoll/`, `scripts/desk-active.sh` — env producers
-
-### Why multi-user (SPEC item 8) is deferred
-
-Not neglect — the one item where doing nothing now beats doing something:
-
-- **No second user exists yet** (the "gf phase"). Building per-user partitioning
-  now means code exercised by zero users and validated by nobody — YAGNI.
-- **The append-only schema makes it a migration, not a rewrite.** No row is ever
-  mutated, so adding `facts/notes/reminders.user_id` later is add-columns +
-  backfill-to-"kami" — no reshaping, no dual-write window. Deferral is cheap.
-- **The hard part is speaker attribution, and it needs the second voice.** A
-  voice-print discriminator (kami vs gf vs unknown) can't be trained or tuned
-  with one voice in the house. Plumbing before the model is pipe with no water.
-- **It's fenced deliberately** (`DO NOT TOUCH THIS PHASE` in `DESIGN.md` § Users) so an
-  autonomous agent doesn't add `user_id` columns while touching the store and
-  commit us to a schema before the constraints that shape it exist.

QA-PLAN.md

@@ -1,184 +0,0 @@
-# QA plan: checking Maven properly
-
-Written 2026-08-01, after the 35-PR stack landed and the box came back up.
-
-44 of the 50 open Vikunja tasks are `QA:` tasks. They are verification work, not
-build work. Most sat unverifiable while Maven was down for 11 days. That
-blocker is gone.
-
-This plan orders them by what unblocks what. Do sessions 1 and 2 first. Almost everything
-downstream assumes the voice loop works, and nobody has confirmed that since
-the redeploy.
-
----
-
-## Before you start
-
-Two things bite anyone running these checks on homesrv.
-
-**curl needs `--noproxy '*'`.** The shell exports `http_proxy=http://127.0.0.1:18080`.
-Without the flag, every local check returns 503 from the proxy and looks like a
-dead service. This cost me a false regression report today.
-
-**The database is not readable with sqlite3.** Four older QA steps say
-`docker compose exec mavend sqlite3 /data/maven.db "select ..."`. That cannot
-work: the container has no `sqlite3` binary, and the store is AES-256-GCM at
-rest with a tmpfs working copy. Read state through mavweb instead, at
-`/history`, `/trace`, `/routines` and `/dash`.
-
----
-
-## Session 1: the voice loop (half a day)
-
-Nothing here has been confirmed since the redeploy, and everything else assumes
-it works. Do this first.
-
-Closes or advances: **44** (conversation), **45** (text chat), **287** (voice
-session quality), **321** steps 3-5 (quiet mode), **288** (STT fixtures).
-
-1. Open `http://127.0.0.1:9201/chat` and hold a short conversation in Russian.
-   Watch for three things: she answers in feminine forms (`рада`, `поняла`), she
-   says `ты` and never `вы`, and no pet names appear.
-2. Press push-to-talk on `/dash`. Say `привет`. Confirm a spoken reply comes
-   back. This is the only check that covers mic to STT to core to TTS to
-   speaker as one path. It is also the path the eleven-day outage most likely
-   broke.
-3. Say `тихий режим`. Expect `тихий режим включён. буду реже напоминать.`
-4. Say `выключи тихий режим`. Expect `тихий режим выключен.` Negation must win.
-5. Say `в комнате тихо`. Quiet mode must NOT flip. Confirm on `/history` that no
-   `quiet_hours` fact was written.
-6. Say `включи режим тишины`, then `сделай потише`. Both must flip quiet mode
-   on. These are the noun form and the comparative, added 01-08-2026.
-7. Wait for a nudge, then say `потом` within twenty minutes. Expect `хорошо,
-   вернусь к этому позже.` and the nudge row on `/notifications` reading
-   `snoozed`. Say `потом` again with nothing pending: it must route as an
-   ordinary utterance, not be swallowed.
-8. Wait for the water nudge, then say `выпил воды`. Expect the ordinary fact
-   reply and nothing extra — she must not congratulate you. Check
-   `/notifications`: the row reads `acted`. Then trigger another nudge and say
-   `готово`; expect `отлично, отметила.` and the same outcome.
-9. Note anything where she is slow, cuts off, or talks over herself. That is
-   287's whole content and it has no written acceptance criteria yet.
-
-**319 is fixed** (01-08-2026). Single-word Russian utterances no longer come
-back as `не совсем поняла — можешь переформулировать?`. `привет` and `поужинал`
-both pass now: `thinSingleToken` spares social singles and any token carrying a
-verb ending, and only thins a bare nominal like `вода`. If a one-word utterance
-still gets clarified during the smoke test, that is a new case for the lexicon,
-not the old bug.
-
----
-
-## Session 2: measurement (half a day, mostly waiting)
-
-Closes or advances: **320** items 2-4, **278** (make the eval lab routine),
-**319** (gate recalibration).
-
-The resident llama-server cannot be reached by the eval harness. It binds
-`--host 127.0.0.1 --port 0` inside the container, so the port is kernel-assigned
-and never published. Start a second one on a fixed port instead:
-
-```sh
-llama-server -m /mnt/hdd1/llms/qwen3/Qwen3-1.7B-UD-Q4_K_XL.gguf \
-  --host 127.0.0.1 --port 18100 -c 4096 -ngl 99 --no-webui
-```
-
-`-c 4096` matters. The recorded numbers were measured at that context size, and
-a mismatch invalidates the comparison.
-
-Then:
-
-```sh
-make eval-models MAVEN_LLM_URL=http://127.0.0.1:18100   # want ~72.7% cascade
-make eval-router                                         # classifier baseline
-MAVEN_LLM_URL=http://127.0.0.1:18100 make eval-phrasing  # persona checks, slow
-make eval-recall
-```
-
-A large miss against 72.7% means the deploy differs from the bench harness.
-
-Two things to decide while the numbers are in front of you:
-
-- **319's gate recalibration.** The single-token rule needs narrowing or
-  dropping. This needs your judgement, not a threshold sweep. The fixture and the
-  daemon disagree about what is correct on two of the three false clarifies.
-- **278's real ask** is making the eval lab routine rather than building it. It
-  is built. Decide whether it runs on a timer, on every merge, or on demand, and
-  the task can close.
-
-Item 4 of **320** needs a permission I do not have. Kill the `llama-server`
-pid under `maven-mavend-1`, post a turn, and confirm it still completes
-through the classifier. Either grant it or run it yourself. It is the only
-check that the failure floor catches a mid-session model death.
-
----
-
-## Session 3: the interaction batch (a day, or three sittings)
-
-These need real use rather than a command, grouped by what one sitting covers.
-
-**Morning and delivery** (**280**, **281**, **128**, **282**): open `/morning`,
-walk the seven required behaviours, then check the four interruption outcomes
-and the digest gap. **282** needs the `desk_active` script enabled on the desk
-PC first, which is **15** and needs you at that machine.
-
-**Tasks and calendar** (**129**, **130**, **127**, **126**, **246**): capture a
-task by voice, confirm it lands, check prioritisation ordering is not nonsense.
-**246** (mail reader) also exercises the `IngestMail` rung that moved to
-`AuthWrite` this morning.
-
-**Routines and patterns** (**43**, **46**, **247**, **254**): these need history
-to detect against. If the database is thin after the outage, they may have
-nothing to propose, which is not a failure. Check `/routines` before
-concluding anything.
-
-**Ecosystem** (**272**, **273**, **276**): nexus, hexis and praxis are wired and
-logged clean at boot. **276** is the degraded-mode suite, which means taking
-siblings down on purpose. Worth doing while you are already in there.
-
----
-
-## Housekeeping (one sitting, no box needed)
-
-Four QA tasks will not close no matter how long they sit, because they are
-gated on something that does not exist:
-
-- **125** zenmoney: needs a token you have not minted.
-- **256** Home Assistant: needs HA configured.
-- **257** Bluetooth: BLOCKED, no bluez on the box. Says so in the title.
-- **288** STT golden audio: needs fixtures generated.
-
-Relabel these so they stop reading as backlog. They are not verification work
-that is pending, they are work that has not started.
-
-Same treatment for the five plan-only tasks (**251** MCP, **252** vision,
-**253** hearing, **255** speaker recognition, **259** crawler). A `QA:` prefix on
-a plan is misleading.
-
----
-
-## Needs you specifically
-
-Not QA. These are blocked on a decision or a credential only you have.
-
-| # | what |
-|---|---|
-| 16 | Create the Kuma API key. `-kuma-key uk5_mavpoll-key` in `docker-compose.yml` is still the placeholder. |
-| 15 | Deploy `desk_active` on the desk PC. Blocks **282**. |
-| 122 | Finish the CPT run for Qwen3-1.7B. The persona fix depends on it. |
-| 355 | Deploy the Hexis auth change. Was blocked on Maven being under construction, which it no longer is. The client half is vendored and wired. |
-| 357 | Decide whether entity-existence validation is the permanent target guard or whether blessing lands in Nexus. |
-| 275 | Hexis native API and MCP parity. |
-| — | Decide on `-require-stepup`. Making it the default needs WebAuthn configured first, or it locks you out of your own admin surfaces. See **317**. |
-| — | Three nginx sites bind wildcard `:80` (`acme.conf`, `matrix`, `panel`), so the ecosystem's bind-level protection is not in effect and `allow`/`deny` is carrying it alone. See **354**. |
-
----
-
-## Suggested order
-
-1. Session 1. If the voice loop is broken, nothing else matters.
-2. The `-require-stepup` and Kuma decisions. Five minutes, unblocks **317** fully
-   and **16**.
-3. Session 2. The numbers tell you whether the router is worth its 90x latency.
-4. Housekeeping. Cheap, and it makes the remaining backlog honest.
-5. Session 3, split whichever way suits you.

cmd/mavenclient/main.go

@@ -1,6 +1,6 @@
 // Package main is mavenclient — maven's reference client.
 //
-// Per DESIGN.md § Voice pipeline (STT / TTS): capture lives on the client;
+// Per docs/design.md § Voice pipeline (STT / TTS): capture lives on the client;
 // the server transcribes + synthesises on demand. The PC client runs the
 // wake-word / VAD gate (cmd/mavwaked) and ships ONE clean audio blob per
 // utterance on activation. The server never owns a mic.

cmd/mavend/actions_fact.go

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

cmd/mavend/actions_query.go

@@ -13,6 +13,7 @@ import (
 	"github.com/kami/maven/internal/ipc"
 	"github.com/kami/maven/internal/memory"
 	"github.com/kami/maven/internal/morning"
+	"github.com/kami/maven/internal/phraser"
 	"github.com/kami/maven/internal/router"
 	"github.com/kami/maven/internal/rss"
 	"github.com/kami/maven/internal/store"
@@ -433,6 +434,14 @@ func (h *reactiveHandler) queryMemory(ctx context.Context, t *queryTurn) (string
 		return "", false
 	}
 	text := hit.Meta["text"]
+	// The score cleared the gate and the topic still has to match (#470). A
+	// note about his slow network scored high enough to answer "почему небо
+	// синее?", because the right-note and must-be-silent score ranges overlap
+	// and no threshold sits between them.
+	if !memory.RecallAllowed(t.dec.Utterance, text) {
+		log.Printf("voice: recall %q rejected for %q: a world question and no shared topic word", text, t.dec.Utterance)
+		return "", false
+	}
 	// A note is phrased in Maven's voice; a fact is read back as it was
 	// stored.
 	if hit.Meta["type"] == "note" {
@@ -468,6 +477,12 @@ func (h *reactiveHandler) queryNotes(ctx context.Context, t *queryTurn) (string,
 	if !memory.ConfidentScores(noteScores, h.queryMinScore, h.queryMinMargin) {
 		return "", false
 	}
+	// Same topic veto as queryMemory above: the best note must be about what
+	// he asked, not merely the nearest vector in the index.
+	if !memory.RecallAllowed(t.dec.Utterance, notes[0].Text) {
+		log.Printf("voice: note %q rejected for %q: a world question and no shared topic word", notes[0].Text, t.dec.Utterance)
+		return "", false
+	}
 	texts := make([]string, len(notes))
 	for i, n := range notes {
 		texts[i] = n.Text
@@ -523,10 +538,7 @@ func (h *reactiveHandler) queryWeb(ctx context.Context, t *queryTurn) (string, b
 	// the question he actually asked. She answers the question, she does not
 	// recite the page.
 	snippet := page.Title + "\n" + crawl.TrimRunes(page.Text, webPageContextRunes)
-	reply, perr := h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{snippet})
-	if perr != nil {
-		log.Printf("voice: web: phrase: %v", perr)
-	}
+	reply := h.phraseSource(ctx, "web", t.dec.Utterance, []string{snippet})
 	if reply == "" {
 		// No phraser (or it failed): read back the top of the page rather than
 		// pretend the fetch did not happen.
@@ -592,14 +604,7 @@ func (h *reactiveHandler) querySearch(ctx context.Context, t *queryTurn) (string
 	// question he asked, not something to recite. The trim is one budget over the
 	// joined block, so a long first snippet cannot crowd out the rest.
 	evidence := crawl.TrimRunes(strings.Join(resp.Snippets(), "\n"), h.search.runes)
-	var reply string
-	if h.phraser != nil {
-		var perr error
-		reply, perr = h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{evidence})
-		if perr != nil {
-			log.Printf("voice: search: phrase: %v", perr)
-		}
-	}
+	reply := h.phraseSource(ctx, "search", t.dec.Utterance, []string{evidence})
 	if reply == "" {
 		// No phraser, or it failed. Read back the best evidence rather than
 		// pretend the search did not happen.
@@ -680,14 +685,7 @@ func (h *reactiveHandler) queryKiwix(ctx context.Context, t *queryTurn) (string,
 	// Handed over the same way a note or a page is: context for the question he
 	// asked, not something to recite.
 	snippet := top.Title + "\n" + crawl.TrimRunes(page.Text, h.kiwix.runes)
-	var reply string
-	if h.phraser != nil {
-		var perr error
-		reply, perr = h.phraser.PhraseQuery(ctx, t.dec.Utterance, []string{snippet})
-		if perr != nil {
-			log.Printf("voice: kiwix: phrase: %v", perr)
-		}
-	}
+	reply := h.phraseSource(ctx, "kiwix", t.dec.Utterance, []string{snippet})
 	if reply == "" {
 		// No phraser, or it failed. Read back the best hit rather than pretend
 		// the search did not happen.
@@ -755,11 +753,28 @@ func isPersonalQuery(utterance string) bool {
 	return false
 }
 
-// queryGeneral — general knowledge from the phraser, the last source before
-// giving up. It always claims: either the model answers or Maven says she
-// doesn't know.
+// queryGeneral — general knowledge, the last source before giving up. It always
+// claims: either a model answers, or Maven names the gap, or she says she does
+// not know.
+//
+// This is the sharpest case for the naming half. Nothing has been fetched, so
+// there is no passage to fall back on and no floor under the answer except the
+// model's weights — and a 1.7B's weights are where the invented answers come
+// from. With a workstation configured and asleep he is told that, rather than
+// told something false in a confident voice. With no workstation configured at
+// all the resident model answers exactly as it does today: naming a gap requires
+// a gap, and on that box the 1.7B is the whole product.
 func (h *reactiveHandler) queryGeneral(ctx context.Context, t *queryTurn) (string, bool) {
-	reply, err := h.phraser.PhraseQuery(ctx, t.dec.Utterance, nil)
+	if h.phraser == nil {
+		// No model of any size. That is not the workstation being asleep, so it
+		// is not that gap: it is simply not knowing.
+		return "не знаю.", true
+	}
+	reply, err := h.phraseWorld(ctx, t.dec.Utterance, nil)
+	if errors.Is(err, phraser.ErrNoWorldModel) {
+		log.Printf("voice: %q needs the world model and it is not available", t.dec.Utterance)
+		return worldGap, true
+	}
 	if err != nil || reply == "" {
 		return "не знаю.", true
 	}

cmd/mavend/confirm.go

@@ -107,14 +107,18 @@ func (h *reactiveHandler) confirmResolvers(ctx context.Context) []confirmResolve
 				return pr != nil && !h.now().After(pr.expiry)
 			},
 			yes: func() string {
-				// Only record the acceptance. The tick loop reads accepted
-				// routines and nudges on their own interval. Building a
-				// reminder here made a routine fire exactly once (Vikunja #366).
-				if err := h.dataStore.AcceptProposedRoutine(ctx, pr.routineID, h.now()); err != nil {
-					log.Printf("voice: accept proposed routine: %v", err)
-					return "не получилось запомнить рутину."
-				}
-				return "буду напоминать."
+				// Voice does NOT accept (Vikunja #367). Accepting hands the
+				// tick loop a standing new reason to speak, which is the same
+				// tier as enabling a tool — and DESIGN.md § "surface caps
+				// authority" says a room mic, reachable by anyone present, is
+				// structurally incapable of layer 3. So a spoken "да" leaves
+				// the row 'proposed' and points at the authed page, where the
+				// accept button is gated at step-up. The convenience of
+				// answering out loud stays; the authority does not move.
+				//
+				// Acceptance itself is recorded by /routines, and the tick
+				// loop nudges on the interval from there (Vikunja #366).
+				return "поняла — подтверди на странице рутин, и начну напоминать."
 			},
 			no: func() string {
 				if err := h.dataStore.DismissProposedRoutine(ctx, pr.routineID); err != nil {

cmd/mavend/factgate.go

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

cmd/mavend/factgate_test.go

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

cmd/mavend/followup.go

@@ -12,13 +12,21 @@ import (
 // which waits on voice-print attribution (see PROGRESS multi-user deferral).
 const voiceDialogueID = "voice"
 
-// toDialogueSlots projects the router's slots onto the dialogue layer's subset
-// (everything except the fact Value, which the dialogue layer doesn't carry).
+// toDialogueSlots and applyDialogueSlots are the only bridge between
+// router.Slots and dialogue.Slots. dialogue must not import router (import
+// cycle), so the two structs are hand-kept copies and every field has to be
+// carried by hand here. Adding a field to either struct without adding it to
+// BOTH functions loses a slot silently — nothing fails to build. The tests in
+// slotsparity_test.go fail when the field sets or the converters stop matching;
+// when they do, fix these two functions, not the tests.
+
+// toDialogueSlots projects the router's slots onto the dialogue layer's copy.
 func toDialogueSlots(s router.Slots) dialogue.Slots {
 	return dialogue.Slots{
 		Time:    s.Time,
 		HasTime: s.HasTime,
 		Key:     s.Key,
+		Value:   s.Value,
 		HasKey:  s.HasKey,
 		Text:    s.Text,
 		Fn:      s.Fn,
@@ -27,11 +35,10 @@ func toDialogueSlots(s router.Slots) dialogue.Slots {
 	}
 }
 
-// applyDialogueSlots writes inherited dialogue slots back onto router slots,
-// preserving router-only fields (Value) the dialogue layer never touched.
+// applyDialogueSlots writes dialogue slots back onto router slots.
 func applyDialogueSlots(base router.Slots, d dialogue.Slots) router.Slots {
 	base.Time, base.HasTime = d.Time, d.HasTime
-	base.Key, base.HasKey = d.Key, d.HasKey
+	base.Key, base.Value, base.HasKey = d.Key, d.Value, d.HasKey
 	base.Text = d.Text
 	base.Fn, base.Args, base.HasFn = d.Fn, d.Args, d.HasFn
 	return base

cmd/mavend/modelseam_test.go

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

cmd/mavend/patterns_test.go

@@ -9,6 +9,7 @@ import (
 
 	"github.com/kami/maven/internal/config"
 	"github.com/kami/maven/internal/delivery"
+	"github.com/kami/maven/internal/ipc"
 	"github.com/kami/maven/internal/loop"
 	"github.com/kami/maven/internal/pattern"
 	"github.com/kami/maven/internal/store"
@@ -283,3 +284,81 @@ func TestTickProposalCooldownSpacesAnnouncements(t *testing.T) {
 		}
 	}
 }
+
+// TestVoiceYesDoesNotAcceptRoutine — Vikunja #367. Accepting a routine hands
+// the tick loop a standing new reason to speak, which DESIGN.md puts at layer
+// 3, and voice is structurally incapable of layer 3. A spoken "да" must park
+// the decision for the authed page, not flip the row itself.
+func TestVoiceYesDoesNotAcceptRoutine(t *testing.T) {
+	st := newTestStore(t)
+	ctx := context.Background()
+	now := refNow()
+	seedRefillEvents(t, st, ctx, now, pattern.MinEvents-1)
+
+	h := &reactiveHandler{api: ipc.NewStoreAPI(st), dataStore: st, now: func() time.Time { return now }}
+
+	// The MinEvents'th event is the one that makes the pattern detectable, and
+	// it goes through the voice path so the proposal is parked for a y/n.
+	last := now.Add(time.Duration(pattern.MinEvents-1) * 7 * 24 * time.Hour)
+	factID, err := st.WriteFact(ctx, last, store.KindSelf, "cat_water", "refill", "voice", 1.0, sql.NullInt64{})
+	if err != nil {
+		t.Fatalf("write fact: %v", err)
+	}
+	if phrase := h.detectPattern(ctx, factID, "cat_water", "refill", last); phrase == "" {
+		t.Fatal("expected a parked routine proposal")
+	}
+
+	reply, handled := h.resolveConfirm(ctx, "да")
+	if !handled {
+		t.Fatal("the spoken yes should be consumed by the routine confirm")
+	}
+	if !strings.Contains(reply, "рутин") {
+		t.Fatalf("reply should send him to the routines page, got %q", reply)
+	}
+
+	rows, err := st.ListProposedRoutinesByStatus(ctx, store.RoutineAccepted)
+	if err != nil {
+		t.Fatalf("list accepted: %v", err)
+	}
+	if len(rows) != 0 {
+		t.Fatalf("voice accepted a routine: %+v", rows)
+	}
+	proposed, err := st.ListProposedRoutinesByStatus(ctx, store.RoutineProposed)
+	if err != nil {
+		t.Fatalf("list proposed: %v", err)
+	}
+	if len(proposed) != 1 {
+		t.Fatalf("proposed routines = %d, want 1 (still waiting for the page)", len(proposed))
+	}
+}
+
+// TestVoiceNoStillDismissesRoutine — declining does not move the boundary
+// outward, so voice keeps it. Only acceptance is gated.
+func TestVoiceNoStillDismissesRoutine(t *testing.T) {
+	st := newTestStore(t)
+	ctx := context.Background()
+	now := refNow()
+	seedRefillEvents(t, st, ctx, now, pattern.MinEvents-1)
+
+	h := &reactiveHandler{api: ipc.NewStoreAPI(st), dataStore: st, now: func() time.Time { return now }}
+
+	last := now.Add(time.Duration(pattern.MinEvents-1) * 7 * 24 * time.Hour)
+	factID, err := st.WriteFact(ctx, last, store.KindSelf, "cat_water", "refill", "voice", 1.0, sql.NullInt64{})
+	if err != nil {
+		t.Fatalf("write fact: %v", err)
+	}
+	if phrase := h.detectPattern(ctx, factID, "cat_water", "refill", last); phrase == "" {
+		t.Fatal("expected a parked routine proposal")
+	}
+
+	if _, handled := h.resolveConfirm(ctx, "нет"); !handled {
+		t.Fatal("the spoken no should be consumed by the routine confirm")
+	}
+	rows, err := st.ListProposedRoutinesByStatus(ctx, store.RoutineDismissed)
+	if err != nil {
+		t.Fatalf("list dismissed: %v", err)
+	}
+	if len(rows) != 1 {
+		t.Fatalf("dismissed routines = %d, want 1", len(rows))
+	}
+}

cmd/mavend/simulator_test.go

@@ -1,5 +1,5 @@
 // mavend/simulator_test.go — the replayable full-system simulator
-// (Vikunja #284, 20-07-2026-BACKLOG.md item 7).
+// (Vikunja #284).
 //
 // # What it is
 //

cmd/mavend/slotsparity_test.go

@@ -0,0 +1,71 @@
+package main
+
+import (
+	"reflect"
+	"testing"
+	"time"
+
+	"github.com/kami/maven/internal/dialogue"
+	"github.com/kami/maven/internal/router"
+)
+
+// TestSlotsParity — dialogue.Slots is a hand-kept copy of router.Slots
+// (dialogue must not import router: import cycle). Drift is silent, so this
+// test compares the two field sets by name and type. If it fails, add the new
+// field to both structs AND to toDialogueSlots/applyDialogueSlots in
+// followup.go — do not relax the test.
+func TestSlotsParity(t *testing.T) {
+	fields := func(v any) map[string]string {
+		rt := reflect.TypeOf(v)
+		out := make(map[string]string, rt.NumField())
+		for i := 0; i < rt.NumField(); i++ {
+			f := rt.Field(i)
+			out[f.Name] = f.Type.String()
+		}
+		return out
+	}
+	rf, df := fields(router.Slots{}), fields(dialogue.Slots{})
+	for name, typ := range rf {
+		dt, ok := df[name]
+		if !ok {
+			t.Errorf("router.Slots.%s (%s) missing from dialogue.Slots", name, typ)
+			continue
+		}
+		if dt != typ {
+			t.Errorf("field %s: router has %s, dialogue has %s", name, typ, dt)
+		}
+	}
+	for name, typ := range df {
+		if _, ok := rf[name]; !ok {
+			t.Errorf("dialogue.Slots.%s (%s) missing from router.Slots", name, typ)
+		}
+	}
+}
+
+// TestSlotsRoundTrip — the converters carry every field. A field the parity
+// test accepts can still be dropped in transit, so round-trip a fully
+// populated value and compare.
+func TestSlotsRoundTrip(t *testing.T) {
+	full := router.Slots{
+		Time:    time.Date(2026, 8, 2, 11, 0, 0, 0, time.UTC),
+		HasTime: true,
+		Fn:      "restart",
+		Args:    []string{"nginx"},
+		HasFn:   true,
+		Key:     "water",
+		Value:   `"drank"`,
+		HasKey:  true,
+		Text:    "выпил воды",
+	}
+	// Every field must be non-zero, or the round-trip proves nothing.
+	rv := reflect.ValueOf(full)
+	for i := 0; i < rv.NumField(); i++ {
+		if rv.Field(i).IsZero() {
+			t.Fatalf("field %s is zero: extend this fixture so the round-trip covers it",
+				rv.Type().Field(i).Name)
+		}
+	}
+	if got := applyDialogueSlots(router.Slots{}, toDialogueSlots(full)); !reflect.DeepEqual(got, full) {
+		t.Errorf("round-trip lost a slot:\n got %+v\nwant %+v", got, full)
+	}
+}

cmd/mavend/voicewire.go

@@ -48,7 +48,11 @@ type voiceWiring struct {
 	// mcp — the MCP client, nil unless the `mcp` block configures an enabled
 	// server (Vikunja #251). Its tools land in the same allowlist as every
 	// other act, so nothing else here has to know about it.
-	mcp *mcpWiring
+	// pair — the workstation model with the resident one as the floor, nil
+	// unless a `workstation` block names an address. Held here only so the
+	// prober is stopped on shutdown; callers were handed it at build time.
+	pair *llm.Pair
+	mcp  *mcpWiring
 	// home — the Home Assistant client, nil unless the `smarthome` block is
 	// enabled (Vikunja #256). Its devices land in the same allowlist as every
 	// other act, so nothing else here has to know about it.
@@ -76,6 +80,9 @@ func (w *voiceWiring) close() {
 	if w.ttsClient != nil {
 		_ = w.ttsClient.Close()
 	}
+	if w.pair != nil {
+		w.pair.Stop()
+	}
 	w.mcp.close()
 }
 
@@ -139,6 +146,7 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 		emb = router.NewHashEmbedder(1024)
 	}
 	w.embedder = emb
+	repairFactVectors(dataStore, emb)
 	checkStoredEmbedder(dataStore, emb)
 
 	// ----- tool executor (the enabled act allowlist, store-backed) -----
@@ -188,6 +196,18 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 		// the new llama-server when the resident model is swapped (Vikunja #250).
 		llmClient = llmClientFor(lp, 60*time.Second)
 	}
+	// The workstation model sits above that one when it is configured and its
+	// card is free. hot is what the router and the replier complete through:
+	// either the pair, or the resident client alone, or nothing at all.
+	hot, pair := modelSeam(cfg, llmClient)
+	w.pair = pair
+	// The phraser gets the same pair, which is what carries the workstation model
+	// into the paths that do not go through `hot`: world questions (the naming
+	// half), and the digestion worker's nudge and reminder phrasing (the silent
+	// half). Wiring, so it happens once and before the voice server listens.
+	if lp, ok := phr.(*phraser.LLMPhraser); ok && pair != nil {
+		lp.UseRemote(pair)
+	}
 	// ----- router (the cascade; floor examples seed the classifier) -----
 	// The act matcher's allowlist is exactly the enabled tool names — the
 	// router only matches acts the executor can run (one source of truth).
@@ -199,7 +219,7 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 	// against the classifier's 50.0%, at about 1s a turn instead of 30ms (see
 	// config.VoiceConfig.LLMRouter). The classifier always stays wired as the
 	// fallback, so a model error never breaks a turn.
-	rtr := buildRouter(emb, matcher, threshold, pickLLMRouter(cfg.Voice.UseLLMRouter(), llmClient))
+	rtr := buildRouter(emb, matcher, threshold, pickLLMRouter(cfg.Voice.UseLLMRouter(), hot))
 
 	// ----- sessions registry (shared with voicesink) -----
 	sessions := voice.NewSessions()
@@ -233,8 +253,8 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 
 	// ----- replier (LLM-backed when the engine is on, Stub floor otherwise) -----
 	replier := voice.Replier(voice.NewStubReplier())
-	if llmClient != nil {
-		replier = newLLMReplier(llmClient, contextBlockFn(cfg, time.Now))
+	if hot != nil {
+		replier = newLLMReplier(hot, contextBlockFn(cfg, time.Now))
 	}
 
 	// ----- the handler (the reactive path; closes over stt / tts / router / coreAPI / memory) -----
@@ -291,7 +311,42 @@ func wireVoice(cfg *config.Config, coreAPI ipc.CoreAPI, phr phraser.Phraser, mem
 // pickLLMRouter returns the LLM router when the operator asked for it and there
 // is a llama-server to talk to, and nil otherwise. nil is safe: the cascade then
 // routes with the classifier, so an unusable setting costs accuracy, not turns.
-func pickLLMRouter(enabled bool, c *llm.Client) *router.LLMRouter {
+// modelSeam builds the completion seam the hot paths use: routing and replies.
+//
+// With no `workstation` block it is the resident client and nothing probes
+// anything, which is today's deploy exactly. With one, it is an llm.Pair that
+// prefers the workstation and falls back to the resident model silently — the
+// silent half of the degradation rule (docs/offload.md), because the big model
+// is only better here and the 1.7B is today's shipping quality. He is never
+// told which of the two phrased his reply.
+//
+// A nil resident client means the phraser is not an LLM phraser. There is then
+// no floor, and a Pair with no floor is a configuration mistake rather than a
+// degraded mode, so the seam is nil and the cascade routes with the classifier.
+func modelSeam(cfg *config.Config, resident *llm.Client) (router.Completer, *llm.Pair) {
+	if resident == nil {
+		if cfg.Workstation != nil {
+			log.Printf("voice: a workstation is configured but there is no resident model to floor it with — ignoring the block")
+		}
+		return nil, nil
+	}
+	if cfg.Workstation == nil {
+		return resident, nil
+	}
+	ws := cfg.Workstation
+	pair := llm.NewPair(
+		llm.New(ws.URL, time.Duration(ws.Timeout)),
+		resident,
+		ws.Health,
+		time.Duration(ws.Probe),
+	)
+	pair.Start(context.Background())
+	log.Printf("voice: workstation model at %s, probed every %s, resident model as the floor",
+		ws.URL, time.Duration(ws.Probe))
+	return pair, pair
+}
+
+func pickLLMRouter(enabled bool, c router.Completer) *router.LLMRouter {
 	if !enabled {
 		return nil
 	}
@@ -419,6 +474,36 @@ func seedTools(api ipc.CoreAPI, tools []config.ToolConfig) {
 	log.Printf("voice: seeded %d act tools from config", n)
 }
 
+// repairFactVectors brings stored fact vectors in line with the facts they name
+// (#493), once per box, before the embedder marker is even looked at.
+//
+// Automatic and not a flag, unlike -reembed: only voice-tapped facts are in
+// this index, so the work is tens of embeddings rather than the thousands of
+// notes that made the backfill a deliberate act. And the box that needs it is
+// broken in a way nobody can see — recall answers with the wrong text and
+// nothing logs an error — so waiting for an operator to know to run it is how
+// the defect survived four restarts in the first place.
+func repairFactVectors(dataStore *store.Store, emb router.Embedder) {
+	if dataStore == nil {
+		return
+	}
+	res, err := dataStore.RepairFactVectors(context.Background(),
+		// EmbedPassage, the stored side, same as every other writer of these
+		// vectors.
+		func(ctx context.Context, text string) ([]float32, error) {
+			return router.EmbedPassage(ctx, emb, text)
+		})
+	if err != nil {
+		log.Printf("voice: fact vector repair failed, no marker written and nothing half-done — retried next start: %v", err)
+		return
+	}
+	if res.Skipped || res.Rewritten+res.Dropped == 0 {
+		return
+	}
+	log.Printf("voice: fact vector repair — %d re-embedded from the fact they name, %d dropped as voided or superseded, %d already right, took %s (#493)",
+		res.Rewritten, res.Dropped, res.Kept, res.Took.Round(time.Millisecond))
+}
+
 // reembedOnStart is the -reembed flag (set in run()). Opt-in on purpose: see
 // runReembed.
 var reembedOnStart bool

cmd/mavend/worldmodel.go

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

cmd/mavend/worldmodel_test.go

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

cmd/mavgpud/gpu.go

@@ -0,0 +1,117 @@
+package main
+
+import (
+	"os"
+	"path/filepath"
+	"strconv"
+	"strings"
+)
+
+// The card is an AMD 7900 GRE with 16GB, driven by amdgpu and ROCm. Everything
+// here reads sysfs and forks nothing: rocm-smi is not even installed on the
+// workstation, and a poll that costs a subprocess every second is a poll that
+// gets tuned down until it is useless.
+
+// gpuProc — one process holding the compute engine.
+type gpuProc struct {
+	PID  int
+	Comm string
+	VRAM int64 // bytes, as the kernel accounts them to this process
+}
+
+// probe reads the two sysfs trees the supervisor decides from.
+//
+// kfdRoot is /sys/class/kfd/kfd/proc, one directory per ROCm process. The
+// directory appears when the process initialises HIP, which is well before it
+// allocates anything large. That is the whole reason this works: the job that
+// is about to want the card announces itself while it is still starting up,
+// so we see the contender rather than only the winner of an allocation race.
+//
+// drmDev is /sys/class/drm/cardN/device, which reports total and used VRAM for
+// the card as a whole.
+type probe struct {
+	kfdRoot string
+	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.
+//
+// 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 {
+	entries, err := os.ReadDir(p.kfdRoot)
+	if err != nil {
+		return nil
+	}
+	var out []gpuProc
+	for _, e := range entries {
+		pid, err := strconv.Atoi(e.Name())
+		if err != nil || pid == selfPID {
+			continue
+		}
+		out = append(out, gpuProc{
+			PID:  pid,
+			Comm: readComm(pid),
+			VRAM: p.procVRAM(e.Name()),
+		})
+	}
+	return out
+}
+
+// procVRAM sums the per-node vram_* files under one process directory. The
+// suffix is the KFD topology node id (vram_35881 on this card), so it is
+// globbed rather than named, and a machine with two cards sums both.
+func (p probe) procVRAM(pid string) int64 {
+	matches, err := filepath.Glob(filepath.Join(p.kfdRoot, pid, "vram_*"))
+	if err != nil {
+		return 0
+	}
+	var total int64
+	for _, m := range matches {
+		total += readInt(m)
+	}
+	return total
+}
+
+// freeVRAM reports the bytes the card has left. Used only to decide whether to
+// start: a shortfall here means llama-server would refuse to load anyway. It is
+// never used to decide to stop, because by the time free VRAM has dropped the
+// other job has already failed its allocation, which is exactly the outcome
+// yielding exists to prevent.
+func (p probe) freeVRAM() int64 {
+	total := readInt(filepath.Join(p.drmDev, "mem_info_vram_total"))
+	used := readInt(filepath.Join(p.drmDev, "mem_info_vram_used"))
+	if total <= 0 {
+		return 0
+	}
+	if free := total - used; free > 0 {
+		return free
+	}
+	return 0
+}
+
+func readInt(path string) int64 {
+	b, err := os.ReadFile(path)
+	if err != nil {
+		return 0
+	}
+	n, err := strconv.ParseInt(strings.TrimSpace(string(b)), 10, 64)
+	if err != nil {
+		return 0
+	}
+	return n
+}
+
+// readComm names the contender for the log. The log is the instrument for the
+// open question in Vikunja #488: whether a process can want this card without
+// ever registering on the KFD, which a Vulkan or video-decode job would.
+func readComm(pid int) string {
+	b, err := os.ReadFile(filepath.Join("/proc", strconv.Itoa(pid), "comm"))
+	if err != nil {
+		return "?"
+	}
+	return strings.TrimSpace(string(b))
+}

cmd/mavgpud/gpu_test.go

@@ -0,0 +1,103 @@
+package main
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"net/url"
+	"os"
+	"path/filepath"
+	"strconv"
+	"testing"
+)
+
+// fakeKFD builds the sysfs shape the workstation actually has: one directory
+// per ROCm process, each holding a vram_<node> file. Sampled from the live box
+// on 02-08-2026, where the CPT run appeared as proc/478104/vram_35881.
+func fakeKFD(t *testing.T, vramByPID map[int]int64) string {
+	t.Helper()
+	root := t.TempDir()
+	for pid, vram := range vramByPID {
+		dir := filepath.Join(root, strconv.Itoa(pid))
+		if err := os.MkdirAll(dir, 0o755); err != nil {
+			t.Fatal(err)
+		}
+		f := filepath.Join(dir, "vram_35881")
+		if err := os.WriteFile(f, []byte(strconv.FormatInt(vram, 10)+"\n"), 0o644); err != nil {
+			t.Fatal(err)
+		}
+	}
+	return root
+}
+
+func TestForeignExcludesOurChild(t *testing.T) {
+	root := fakeKFD(t, map[int]int64{478104: 12791693312, 999: 4096})
+	p := probe{kfdRoot: root}
+
+	all := p.foreign(0)
+	if len(all) != 2 {
+		t.Fatalf("with no child running, both processes are foreign, got %d", len(all))
+	}
+
+	ours := p.foreign(999)
+	if len(ours) != 1 || ours[0].PID != 478104 {
+		t.Fatalf("our own llama-server must not count as a contender, got %+v", ours)
+	}
+	if ours[0].VRAM != 12791693312 {
+		t.Errorf("per-process VRAM = %d, want the value from vram_35881", ours[0].VRAM)
+	}
+}
+
+// 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) {
+	if got := (probe{kfdRoot: t.TempDir()}).foreign(0); len(got) != 0 {
+		t.Errorf("empty kfd tree: got %d processes, want 0", len(got))
+	}
+	if got := (probe{kfdRoot: "/nonexistent"}).foreign(0); got != nil {
+		t.Errorf("missing kfd tree: got %+v, want nil", got)
+	}
+}
+
+func TestFreeVRAM(t *testing.T) {
+	dev := t.TempDir()
+	write := func(name, v string) {
+		if err := os.WriteFile(filepath.Join(dev, name), []byte(v), 0o644); err != nil {
+			t.Fatal(err)
+		}
+	}
+	// The live numbers from the workstation while the CPT run held the card.
+	write("mem_info_vram_total", "17163091968\n")
+	write("mem_info_vram_used", "13396389888\n")
+	p := probe{drmDev: dev}
+	if got, want := p.freeVRAM(), int64(3766702080); got != want {
+		t.Errorf("freeVRAM = %d, want %d", got, want)
+	}
+	if got := (probe{drmDev: "/nonexistent"}).freeVRAM(); got != 0 {
+		t.Errorf("unreadable card reports %d free, want 0 so nothing starts", got)
+	}
+}
+
+// With no model loaded the supervisor must still answer, and it must answer 503
+// 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, "")}
+	h := s.handler(mustURL(t, "http://127.0.0.1:1"))
+
+	for _, path := range []string{"/health", "/v1/chat/completions"} {
+		w := httptest.NewRecorder()
+		h.ServeHTTP(w, httptest.NewRequest(http.MethodGet, path, nil))
+		if w.Code != http.StatusServiceUnavailable {
+			t.Errorf("%s with no model: got %d, want 503", path, w.Code)
+		}
+	}
+}
+
+func mustURL(t *testing.T, s string) *url.URL {
+	t.Helper()
+	u, err := url.Parse(s)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return u
+}

cmd/mavgpud/main.go

@@ -0,0 +1,247 @@
+// mavgpud — the workstation's GPU supervisor.
+//
+// It runs on the workstation (an AMD 7900 GRE, 16GB), not on homesrv, and it is
+// deployed separately from the Maven daemons. Maven does not participate in any
+// of this and never asks for a start: it reads /health through internal/llm.Pair
+// and either gets the big model or falls back to the resident 1.7B.
+//
+// The rule, from Vikunja #488: keep llama-server loaded whenever the card is
+// free, unload it when it has been idle too long or when another process needs
+// 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.
+package main
+
+import (
+	"context"
+	"encoding/json"
+	"flag"
+	"log"
+	"net/http"
+	"net/http/httputil"
+	"net/url"
+	"os"
+	"os/signal"
+	"sync/atomic"
+	"syscall"
+	"time"
+)
+
+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"`
+	// 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"`
+
+	KFDRoot   string `json:"kfd_root"`
+	DRMDevice string `json:"drm_device"`
+
+	Poll        duration `json:"poll"`
+	IdleTimeout duration `json:"idle_timeout"`
+	StopGrace   duration `json:"stop_grace"`
+	MinFreeVRAM int64    `json:"min_free_vram_bytes"`
+	// EvictAfter and StartAfter are counted in polls, not seconds. Both exist
+	// to damp flapping: a one-tick blip from a short-lived rocm process must
+	// not evict the model, and a card that has just been released must not be
+	// grabbed before the previous job has finished unmapping.
+	EvictAfter int `json:"evict_after_polls"`
+	StartAfter int `json:"start_after_polls"`
+}
+
+func defaults() config {
+	return config{
+		Listen:      ":8080",
+		LlamaAddr:   "127.0.0.1:8081",
+		KFDRoot:     "/sys/class/kfd/kfd/proc",
+		DRMDevice:   "/sys/class/drm/card1/device",
+		Poll:        duration(time.Second),
+		IdleTimeout: duration(15 * time.Minute),
+		StopGrace:   duration(20 * time.Second),
+		MinFreeVRAM: 15 << 30,
+		EvictAfter:  2,
+		StartAfter:  5,
+	}
+}
+
+// duration lets the config file say "15m" instead of counting nanoseconds.
+type duration time.Duration
+
+func (d *duration) UnmarshalJSON(b []byte) error {
+	var s string
+	if err := json.Unmarshal(b, &s); err != nil {
+		return err
+	}
+	v, err := time.ParseDuration(s)
+	if err != nil {
+		return err
+	}
+	*d = duration(v)
+	return nil
+}
+
+func main() {
+	path := flag.String("config", "/etc/mavgpud.json", "config file")
+	flag.Parse()
+
+	cfg := defaults()
+	b, err := os.ReadFile(*path)
+	if err != nil {
+		log.Fatalf("mavgpud: read config: %v", err)
+	}
+	if err := json.Unmarshal(b, &cfg); err != nil {
+		log.Fatalf("mavgpud: parse config: %v", err)
+	}
+	if cfg.LlamaBin == "" {
+		log.Fatal("mavgpud: llama_bin is required")
+	}
+
+	base := "http://" + cfg.LlamaAddr
+	run := newRunner(cfg.LlamaBin, cfg.LlamaArgs, base+"/health")
+	sup := &supervisor{
+		cfg:   cfg,
+		probe: probe{kfdRoot: cfg.KFDRoot, drmDev: cfg.DRMDevice},
+		run:   run,
+	}
+	sup.touch()
+
+	ctx, cancel := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
+	defer cancel()
+
+	target, err := url.Parse(base)
+	if err != nil {
+		log.Fatalf("mavgpud: llama_addr: %v", err)
+	}
+	srv := &http.Server{Addr: cfg.Listen, Handler: sup.handler(target)}
+	go func() {
+		log.Printf("mavgpud: listening on %s, model %s", cfg.Listen, cfg.LlamaBin)
+		if err := srv.ListenAndServe(); err != nil && err != http.ErrServerClosed {
+			log.Fatalf("mavgpud: listen: %v", err)
+		}
+	}()
+
+	sup.loop(ctx)
+
+	// The card must come back before we do. A supervisor that exits leaving
+	// llama-server holding 14GB is worse than one that never ran.
+	shut, done := context.WithTimeout(context.Background(), 5*time.Second)
+	defer done()
+	_ = srv.Shutdown(shut)
+	run.stop(time.Duration(cfg.StopGrace))
+}
+
+type supervisor struct {
+	cfg   config
+	probe probe
+	run   *runner
+
+	lastReq atomic.Int64 // unix nanos of the last request Maven sent
+
+	foreignStreak int
+	clearStreak   int
+}
+
+func (s *supervisor) touch() { s.lastReq.Store(time.Now().UnixNano()) }
+
+func (s *supervisor) idle() time.Duration {
+	return time.Since(time.Unix(0, s.lastReq.Load()))
+}
+
+// handler serves the two things the workstation exposes.
+//
+// /health is answered locally and always, with no GPU cost and no round trip,
+// because it is the only thing Maven reads and Maven reads it on a timer
+// forever. Everything else is llama-server's API, reverse-proxied. Proxying
+// rather than pointing Maven straight at llama-server is what makes the idle
+// window measurable: the supervisor cannot otherwise know when the model was
+// last used.
+func (s *supervisor) handler(target *url.URL) http.Handler {
+	proxy := httputil.NewSingleHostReverseProxy(target)
+	mux := http.NewServeMux()
+	mux.HandleFunc("/health", func(w http.ResponseWriter, r *http.Request) {
+		if !s.run.isReady() {
+			http.Error(w, "model not loaded", http.StatusServiceUnavailable)
+			return
+		}
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"status":"ok"}`))
+	})
+	mux.HandleFunc("/", 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
+}
+
+func (s *supervisor) loop(ctx context.Context) {
+	t := time.NewTicker(time.Duration(s.cfg.Poll))
+	defer t.Stop()
+	for {
+		select {
+		case <-ctx.Done():
+			return
+		case <-t.C:
+			s.tick(ctx)
+		}
+	}
+}
+
+// tick is the whole decision. Yielding is checked before starting, and presence
+// on the KFD is what triggers it — not a VRAM threshold. A ROCm process
+// registers under /sys/class/kfd/kfd/proc when it initialises HIP, before it
+// 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())
+	if len(others) > 0 {
+		s.foreignStreak++
+		s.clearStreak = 0
+	} else {
+		s.foreignStreak = 0
+		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))
+		}
+		return
+	}
+
+	if s.clearStreak < s.cfg.StartAfter {
+		return
+	}
+	if free := s.probe.freeVRAM(); free < s.cfg.MinFreeVRAM {
+		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)
+	}
+}
+
+// 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.
+func describe(procs []gpuProc) string {
+	out := ""
+	for i, p := range procs {
+		if i > 0 {
+			out += ", "
+		}
+		out += p.Comm
+	}
+	return out
+}

cmd/mavgpud/runner.go

@@ -0,0 +1,132 @@
+package main
+
+import (
+	"context"
+	"log"
+	"net/http"
+	"os/exec"
+	"sync"
+	"syscall"
+	"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
+// 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.
+type runner struct {
+	bin  string
+	args []string
+	// ready is llama-server's own /health, which answers "is a model loaded".
+	// Loading a 7-14B takes tens of seconds, so started is not ready.
+	readyURL string
+
+	mu    sync.Mutex
+	cmd   *exec.Cmd
+	ready bool
+	http  *http.Client
+}
+
+func newRunner(bin string, args []string, readyURL string) *runner {
+	return &runner{
+		bin: bin, args: args, readyURL: readyURL,
+		http: &http.Client{Timeout: 2 * time.Second},
+	}
+}
+
+// pid is the child's, or 0. The GPU probe needs it to tell our own model apart
+// from a contender.
+func (r *runner) pid() int {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	if r.cmd == nil || r.cmd.Process == nil {
+		return 0
+	}
+	return r.cmd.Process.Pid
+}
+
+func (r *runner) running() bool { return r.pid() != 0 }
+
+// isReady reports the cached readiness. The supervisor loop refreshes it; the
+// health handler only reads, so answering /health never costs a round trip.
+func (r *runner) isReady() bool {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	return r.ready
+}
+
+// start launches llama-server. It returns as soon as the process exists, not
+// when the model is loaded.
+func (r *runner) start() error {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	if r.cmd != nil {
+		return nil
+	}
+	cmd := exec.Command(r.bin, r.args...)
+	// Own process group, so stop kills anything llama-server spawned rather
+	// than leaving it holding VRAM after we have declared the card yielded.
+	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
+	if err := cmd.Start(); err != nil {
+		return err
+	}
+	r.cmd, r.ready = cmd, false
+	log.Printf("mavgpud: started llama-server pid=%d", cmd.Process.Pid)
+	go func() {
+		err := cmd.Wait()
+		r.mu.Lock()
+		r.cmd, r.ready = nil, false
+		r.mu.Unlock()
+		log.Printf("mavgpud: llama-server exited: %v", err)
+	}()
+	return nil
+}
+
+// stop ends llama-server 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.
+func (r *runner) stop(grace time.Duration) {
+	r.mu.Lock()
+	cmd := r.cmd
+	r.ready = false
+	r.mu.Unlock()
+	if cmd == nil || cmd.Process == nil {
+		return
+	}
+	pgid := -cmd.Process.Pid
+	_ = syscall.Kill(pgid, syscall.SIGTERM)
+	deadline := time.Now().Add(grace)
+	for time.Now().Before(deadline) {
+		if !r.running() {
+			return
+		}
+		time.Sleep(100 * time.Millisecond)
+	}
+	log.Printf("mavgpud: llama-server did not exit in %s, killing", grace)
+	_ = syscall.Kill(pgid, syscall.SIGKILL)
+}
+
+// refreshReady asks llama-server whether the model is loaded. Called once per
+// supervisor tick, never per request.
+func (r *runner) refreshReady(ctx context.Context) {
+	if !r.running() {
+		return
+	}
+	ok := false
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, r.readyURL, nil)
+	if err == nil {
+		resp, err := r.http.Do(req)
+		if err == nil {
+			ok = resp.StatusCode == http.StatusOK
+			resp.Body.Close()
+		}
+	}
+	r.mu.Lock()
+	was := r.ready
+	r.ready = ok
+	r.mu.Unlock()
+	if ok && !was {
+		log.Printf("mavgpud: model ready")
+	}
+}

cmd/mavweb/main.go

@@ -512,7 +512,7 @@ func main() {
 
 	// /tools — the authed enable surface. maven proposes acts she can't run;
 	// this page is where a human reviews and enables them (proposed→enabled).
-	// Enabling is the boundary-moving act (DESIGN.md § Tool registration —
+	// Enabling is the boundary-moving act (docs/design.md § Tool registration —
 	// drafting is suggest, enabling is act), so it lives ONLY here,
 	// behind wg+nginx+auth — never the voice/chat path.
 	mux.HandleFunc("/tools", func(w http.ResponseWriter, r *http.Request) {
@@ -1210,13 +1210,10 @@ func routineRows(rs []ipc.ProposedRoutine) []routineRow {
 	return out
 }
 
-// acceptRoutine creates the recurring reminder for a proposal, then marks the
-// proposal accepted and links the reminder to it. Weekly patterns get a cron
-// expression; any other interval fires once.
-//
-// TODO(vikunja#46): this mirrors the voice accept path in cmd/mavend/voice.go.
-// When the tick loop learns to read accepted proposals directly, both callers
-// should hand off to one place in core instead of each building a reminder.
+// acceptRoutine marks a proposal accepted. This page is the ONLY surface that
+// may do it (Vikunja #367): accepting gives the tick loop a standing new
+// reason to speak, which DESIGN.md puts at layer 3, and the button here is
+// behind step-up. Voice can park the question and dismiss, never accept.
 func acceptRoutine(ctx context.Context, core ipc.CoreAPI, id int64) error {
 	proposed, err := core.ListProposedRoutines(ctx)
 	if err != nil {

deploy/mavend.json

@@ -39,6 +39,23 @@
     "proxy": "socks5://192.168.240.1:10808"
   },
 
+  "//workstation": [
+    "The big model on the desk PC (workpc, 7900 GRE 16GB), fronted by",
+    "mavgpud on port 8080. It runs gemma-4-12b and it is preferred over the",
+    "resident Qwen3-1.7B for routing and replies whenever the card is free.",
+    "The machine is never assumed up: it sleeps, and the card is often held by",
+    "a CPT run, in which case mavgpud answers 503 and Maven falls back to the",
+    "resident model without saying so. Deleting this block restores exactly",
+    "the behaviour homesrv had before it existed.",
+    "Addressed by LAN address, not container name: mavgpud runs on another",
+    "machine and there is no shared docker network to name it on."
+  ],
+  "workstation": {
+    "url": "http://192.168.1.105:8080",
+    "probe": "15s",
+    "timeout": "90s"
+  },
+
   "//search": [
     "The live web, searched after his own notes and before Kiwix. Only the",
     "query string leaves the box — never a note, a fact, the persona block or",
@@ -76,6 +93,56 @@
     "snippet_runes": 1500
   },
 
+  "//morning_routines": [
+    "The daily checklist (Vikunja #280). Each item is done when its fact_key",
+    "gets a non-voided fact inside the window, so 'выпил воды' closes water and",
+    "nothing has to be ticked by hand. nudge_at fires once, at the end of the",
+    "window, and only for what is still open. Weekdays empty = every day."
+  ],
+  "morning_routines": [
+    {
+      "name": "утро",
+      "window_start": "08:00",
+      "window_end": "11:00",
+      "nudge_at": "10:30",
+      "severity": 1,
+      "items": [
+        { "key": "medicine", "fact_key": "medicine", "label": "лекарство" },
+        { "key": "water",    "fact_key": "water",    "label": "вода" },
+        { "key": "pets",     "fact_key": "pets",     "label": "покормить кота" }
+      ]
+    }
+  ],
+
+  "//feeds": [
+    "RSS reading (Vikunja #258). Every item lands as a note with source",
+    "rss:<name>, which is also what puts entries in the intake journal that",
+    "/events reads. Only the feed URL leaves the box.",
+    "This is a starting pair, not a curated set — trim or extend it."
+  ],
+  "feeds": {
+    "poll_interval": "30m",
+    "max_items": 5,
+    "max_age": "24h",
+    "sources": [
+      { "name": "lwn",       "url": "https://lwn.net/headlines/newrss", "category": "технологии" },
+      { "name": "archlinux", "url": "https://archlinux.org/feeds/news/", "category": "технологии" }
+    ]
+  },
+
+  "//crawl": [
+    "Reading a web page (Vikunja #259). on_demand answers 'посмотри <URL>'.",
+    "No allow_hosts, so any public host he names is readable; private",
+    "addresses are refused unconditionally by internal/webfetch and do not",
+    "need listing. Setting allow_hosts here would also narrow on-demand,",
+    "which is the point of leaving it empty."
+  ],
+  "crawl": {
+    "on_demand": true,
+    "timeout": "10s",
+    "max_runes": 4000
+  },
+
   "digest": {
     "enabled": true,
     "window": "30m",
@@ -119,7 +186,7 @@
     "timeout": "400ms",
     "rate": 100,
     "max_hosts": 256,
-    "enabled": false
+    "enabled": true
   },
 
   "nexus":  { "url": "http://nexus:9740" },

deploy/mavgpud.json

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

deploy/mavgpud.service

@@ -0,0 +1,24 @@
+[Unit]
+# Runs on the workstation (bugmachine), not on homesrv. Install as a systemd
+# user unit and turn on lingering, so the card is supervised after a reboot
+# with nobody logged in:
+#
+#   scp mavgpud workpc:~/.local/bin/mavgpud
+#   scp deploy/mavgpud.json workpc:~/.config/mavgpud.json
+#   scp deploy/mavgpud.service workpc:~/.config/systemd/user/mavgpud.service
+#   ssh workpc 'systemctl --user daemon-reload && systemctl --user enable --now mavgpud'
+#   sudo loginctl enable-linger kami
+Description=Maven GPU supervisor (holds llama-server while the card is free)
+After=network.target
+
+[Service]
+ExecStart=%h/.local/bin/mavgpud -config %h/.config/mavgpud.json
+Restart=always
+RestartSec=5
+# The card must come back when the supervisor goes down. mavgpud stops
+# llama-server on SIGTERM, so give it longer than stop_grace to do that.
+KillSignal=SIGTERM
+TimeoutStopSec=60
+
+[Install]
+WantedBy=default.target

docker-compose.yml

@@ -79,7 +79,16 @@ services:
     <<: *image
     # voice.bind is 0.0.0.0:9100 in deploy/mavend.json so mavweb can reach it
     # cross-container. Verified 2026-07-06.
+    # -ambient-token turns on POST /api/ambient (Vikunja #126): the phone posts
+    # notification text, mavweb keeps only a meeting time. Empty ⇒ no route at
+    # all, which is what a missing MAVEN_AMBIENT_TOKEN gives. The value comes
+    # from the gitignored .env docker compose reads for interpolation, NOT from
+    # an env_file — flags are interpolated before any service env exists.
+    # Weakness worth naming: mavweb takes this as a flag, so it is visible in
+    # `ps` inside this container, unlike the zenmoney and IMAP secrets which are
+    # read from files.
     command: ["mavweb", "-addr", ":9201", "-voice", "mavend:9100", "-core", "/run/maven/mavend.sock",
+              "-ambient-token", "${MAVEN_AMBIENT_TOKEN:-}",
               "-nexus", "http://nexus:9740", "-praxis", "http://praxis:8989", "-hexis", "http://hexis:9741"]
     depends_on: [mavend]
     # loopback-only on purpose: /tools defines+executes arbitrary argv and

docs/archive/2026-07-03-feature-ranking.md

@@ -1,6 +1,45 @@
 # maven — feature ranking
 
-> dated 2026-07-03. companion to `DESIGN.md` (folded from the former `maven.md`). ranks everything discussed post-repo-state against the infra blockers, not a replacement for the build order.
+> **Archived 2026-08-02 (V-447).** The mandatory and easy tiers are now Vikunja tasks
+> 449-458. Two of those closed immediately, because the ranking was stale. Quiet hours
+> (V-450) ship as `QuietHoursConfig` plus the care gate in `internal/loop/loop.go`.
+> Schema migrations (V-451) ship as `internal/store/migrations.go` on `PRAGMA
+> user_version`. The doable and epic tiers stay here because they are reasoning. Some of
+> them exist only to record why something is not worth doing yet. Read this for the why,
+> not as a work queue, and check the code before believing a gap.
+
+> dated 2026-07-03. companion to `docs/design.md` (folded from the former `maven.md`). ranks everything discussed post-repo-state against the infra blockers, not a replacement for the build order.
+
+---
+
+## what already shipped (checked against the code, 2026-08-02)
+
+One month old and already wrong in ten places. Everything below is marked
+built after reading the code, not the board. Read the tiers underneath with this list in
+hand.
+
+Infra 1 and 2, the two the ranking says block every feature, are both done. sqlcipher
+at-rest ships as `Store.enc` plus `OpenEncrypted`, a tmpfs working copy re-encrypted on
+`Close`, keyed from `db_key_env` in `deploy/mavend.json`. mavweb and mavcaldav are no
+longer at zero coverage: seven test files under `cmd/mavweb`, including
+`credentials_test.go` and `passkey_prf_test.go`, and two under `cmd/mavcaldav`. Infra 3
+is stale in the other direction. There are still no systemd units, but the deploy is
+`deploy/ecosystem/docker-compose.yml`, not scripts and tmux.
+
+Doable tier, built: rule trace and explanation as `/trace` plus `internal/loop/explain.go`.
+Recurring reminders as the cron column, `NextFireTs` and `RescheduleReminder`
+(`internal/store/reminders.go:206`), so "fires once right now" is wrong. Stale-reminder
+burst collapse as `collapseReminders` (`internal/loop/gather.go:209`). Revert as
+`VoidLatestFact` (`internal/store/facts.go:300`). Digest mode as `internal/store/digest.go`.
+Testing infra as the simulator and the eval lab (V-284, V-278). Passkey persistence as the
+JSON-backed `credentialStore` in `cmd/mavweb/credentials.go`. Most integrations shipped as
+their own QA tasks (V-246 mail, V-256 smarthome, V-258 rss, V-259 crawler).
+
+Doable tier, still open: correx, systemd units, memory decay and duplicate detection
+(nothing in `internal/memory` touches it), backup automation, import and export, barge-in.
+
+Epic tier, unbuilt as ranked. `event.Bus` exists (`cmd/mavend/intake.go:57`) but it is the
+intake journal from V-283, not the rewrite of facts into projections that this tier means.
 
 ---
 
@@ -20,7 +59,7 @@ nothing feature-level below should land before 1–2 are done. 3–4 can interle
 ### mandatory
 things that block correctness or safety of stuff already shipped — not new capability, just closing gaps in existing design.
 
-- **destructive-confirm policy** — open question in `DESIGN.md` § open questions, blocks correx and any new tool domain from having a coherent risk tier
+- **destructive-confirm policy** — open question in `docs/design.md` § open questions, blocks correx and any new tool domain from having a coherent risk tier
 - **quiet-hours definition** — open question, blocks proactive delivery being trustworthy
 - **schema migrations** — sqlcipher rollout alone forces a schema touch. want this mechanism before that, not after.
 
@@ -30,7 +69,7 @@ cheap, no dependencies, no new invariants.
 - **grocery / `list_items` table** — fourth append-only shape (item, status, list-tag), no predicate touches it, multi-adder just works for free
 - **go.mod tidy**
 - **capability model** (deepseek) — `homelab.docker.restart` instead of flat `tool→enabled`. cheap now, expensive to retrofit once tools surface passes ~15 entries. time-sensitive, not urgent.
-- **conversation repair** — already free: `DESIGN.md` has "misroute correction = new centroid example," this is just naming the existing mechanism as a feature
+- **conversation repair** — already free: `docs/design.md` has "misroute correction = new centroid example," this is just naming the existing mechanism as a feature
 - **command history** — read-only query over existing facts, no new mechanism
 - **clarification templates** — canned phrasing for the router's existing confidence-gate fallback, phraser-lane only
 - **pronunciation dictionary** — tts config, no architecture

docs/archive/2026-07-30-senior-review.md

@@ -30,7 +30,7 @@ critical workflows: voice turn (mic→STT→route→tool/reply→TTS); proactive
                     (/dash /chat /tools /ecosystem)
 current state:      all 34 test packages pass; vet clean; `make test` still exits 1
                     (finding 2)
-known failures:     weak RU query routing — REARCH.md names the cause; the named fix
+known failures:     weak RU query routing — docs/rearchitecture.md names the cause; the named fix
                     is wired `nil`
 maintenance burden: 4,518 lines of root markdown vs 33,319 lines of Go; 15 top-level
                     .md files, 3 of them dated session logs; several contradict
@@ -47,11 +47,11 @@ what's obsolete:    llmrouter.go (built, tested, never wired); classifier seed-p
 ```
 
 **Classification: healthy + misaligned.** Not fragile, not overbuilt, not abandoned. The
-architecture in `REARCH.md` is sound and mostly *built* — it just is not *connected*.
+architecture in `docs/rearchitecture.md` is sound and mostly *built* — it just is not *connected*.
 
 ## what it should become
 
-The thing `REARCH.md` already describes, with the switch flipped and the drift removed:
+The thing `docs/rearchitecture.md` already describes, with the switch flipped and the drift removed:
 one resident small model doing both routing and phrasing, classifier demoted from the live
 path to the failure floor, embedder demoted to RAG hint. No new architecture is needed.
 **The gap is a config/wiring decision plus doc convergence, not a redesign.**
@@ -65,19 +65,19 @@ path to the failure floor, embedder demoted to RAG hint. No new architecture is
 `architecture` / `repair`
 
 **problem:** The most load-bearing design decision in the project is stated four different,
-incompatible ways, and the code path `REARCH.md` calls "the linchpin" is disabled.
+incompatible ways, and the code path `docs/rearchitecture.md` calls "the linchpin" is disabled.
 
 **evidence** (all confirmed):
 
 - `cmd/mavend/voice.go:211` — `rtr := buildRouter(emb, matcher, threshold, nil) // LLM router disabled`,
   with comment *"the classifier handles routing reliably."*
-- `REARCH.md:11` says the same classifier is *"the structural cause of 'she messes up
+- `docs/rearchitecture.md:11` says the same classifier is *"the structural cause of 'she messes up
   queries.'"* **The code comment and the design doc make opposite claims about the same
   component.**
 - `internal/router/llmrouter.go` (139 lines) + `llmrouter_test.go` — fully built and
   tested, zero non-test callers.
-- Model identity, four ways: docs say **Qwen3-1.7B** (`CLAUDE.md:6`, `REARCH.md:15`,
-  `SPEC.md:46`, `AGENTS.md:79`, `MAVEN_ECOSYSTEM_ARCHITECTURE.md:72`);
+- Model identity, four ways: docs say **Qwen3-1.7B** (`CLAUDE.md:6`, `docs/rearchitecture.md:15`,
+  `SPEC.md:46`, `AGENTS.md:79`, `docs/ecosystem.md:72`);
   `deploy/mavend.json:9` says **Qwen3.5-2B-UD-Q4_K_XL**; `models/llm/` on disk holds
   **LFM2.5-1.2B-Thinking**; code comments in 5 files still say **LFM**.
 - `deploy/mavend.json:11` sets `"n_gpu_layers": 99` while `CLAUDE.md:4` states the target
@@ -100,7 +100,7 @@ match. Delete nothing from `internal/router` yet — the classifier is the fallb
 reconciliation, not a refactor. **Do not rewrite the router.**
 
 **alternatives:** Delete `llmrouter.go` and commit to the classifier — only defensible if
-the eval harness shows the classifier is actually adequate, which contradicts `REARCH.md`.
+the eval harness shows the classifier is actually adequate, which contradicts `docs/rearchitecture.md`.
 
 **risk:** Low-moderate. LLM route failures already fall through to the classifier
 (`router.go:88-96`), so a bad model cannot break a turn. The real risk is CPU latency.
@@ -228,21 +228,21 @@ after finding 1, not before** — and skip it if it stays purely cosmetic.
 **problem:** 15 root markdown files, 4,518 lines, several stale or superseded, at least
 three pairs contradicting each other.
 
-**evidence:** `ROADMAP.md` (759) + `MAVEN_ECOSYSTEM_ARCHITECTURE.md` (884) +
+**evidence:** `ROADMAP.md` (759) + `docs/ecosystem.md` (884) +
 `PROGRESS.md` (456) + `maven.md` (413) + `20-07-2026-BACKLOG.md` (396) +
 `SESSION-05-07-2026.md` + `SESSION-06-07-2026.md` (477 combined) + `PLANS.md` (25) +
-`START.md` + `SPEC.md` + `PROTOCOL.md`. `PROGRESS.md:61` annotates its own staleness:
-*"Older LFM references below describe the currently deployed..."*. `REARCH.md` announces it
+`docs/operations.md` + `SPEC.md` + `docs/protocol.md`. `PROGRESS.md:61` annotates its own staleness:
+*"Older LFM references below describe the currently deployed..."*. `docs/rearchitecture.md` announces it
 "supersedes" a model still described as current elsewhere.
 
 **impact:** The doc set is the reason finding 1 exists. When five documents describe the
 architecture, the code becomes the only trustworthy one — which defeats the purpose of
 having them.
 
-**recommended action:** Keep `CLAUDE.md` (agent contract), `REARCH.md` (target
-architecture), `AGENTS.md` (recipes), `PROTOCOL.md` (wire format),
+**recommended action:** Keep `CLAUDE.md` (agent contract), `docs/rearchitecture.md` (target
+architecture), `AGENTS.md` (recipes), `docs/protocol.md` (wire format),
 `20-07-2026-BACKLOG.md` (live queue). Delete the two `SESSION-*.md` and `PLANS.md` — git
-history holds them. Fold `SPEC.md` + `maven.md` + `ROADMAP.md` into one `DESIGN.md` and
+history holds them. Fold `SPEC.md` + `maven.md` + `ROADMAP.md` into one `docs/design.md` and
 mark superseded sections instead of leaving them to read as current. Target ~1,500 lines.
 
 ---
@@ -320,7 +320,7 @@ expected maintenance gain:      none over the incremental path
   suggesting Vulkan offload is intended and working — but `CLAUDE.md` says CPU-only. Likely
   the doc is stale, not the config; unverified.
 - **Whether the classifier is genuinely adequate.** `voice.go:211` asserts it is;
-  `REARCH.md` asserts it is not. Both are claims, neither is measured. The uncommitted eval
+  `docs/rearchitecture.md` asserts it is not. Both are claims, neither is measured. The uncommitted eval
   harness is the instrument to settle it — resolve before flipping the router, not after.
 - **Whether wg+nginx+auth actually fronts 9201 in production.** Not in this repo. If it
   does, finding 3 drops from "unauthenticated RCE" to "the control is not reproducible from
@@ -350,7 +350,7 @@ Key claims independently re-verified against the working tree; the verdict stand
   over WireGuard on homesrv, this is hygiene, not an emergency — but the loopback bind
   and startup warning are cheap insurance either way, so do them regardless.
 - Finding 1's "flip the router" step should be gated harder on measurement.
-  `REARCH.md`'s claim that the classifier causes weak RU queries is itself unmeasured —
+  `docs/rearchitecture.md`'s claim that the classifier causes weak RU queries is itself unmeasured —
   the review admits this under uncertainties, but the "repair now" ordering buries it.
   Run `eval_scenarios_test.go` against both paths **before** deciding to flip, not
   after. A 2B model on CPU may add enough latency that the classifier wins in practice

docs/design.md

@@ -1,10 +1,12 @@
 # Maven — Design
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 > Folded 2026-07-30 from `SPEC.md` (north star, 2026-07-03), `maven.md`
 > (consolidated decisions, 2026-06-30) and `ROADMAP.md` (execution plan,
 > 2026-07-06). Those three files are gone; git history holds them.
 > This is the single design document: principles, target state, and the
-> execution ledger. `REARCH.md` remains authoritative wherever it disagrees
+> execution ledger. `docs/rearchitecture.md` remains authoritative wherever it disagrees
 > with anything here. Everything the three sources asserted that is no longer
 > the intended design is preserved under **§ Superseded** — do not read that
 > section as current.
@@ -160,7 +162,7 @@ presence_state ( last_bucket, last_score, updated_ts )
 ```
 
 Facts additionally carry `Subject`/`EntityID`/`ResolutionState` for
-entity-aware resolution against Nexus (see `MAVEN_ECOSYSTEM_ARCHITECTURE.md`).
+entity-aware resolution against Nexus (see `docs/ecosystem.md`).
 
 ### Trigger model
 
@@ -196,7 +198,7 @@ INTO the gate as an env predicate, not the LLM's job.
 
 ## Reactive path — routing
 
-**Target design: LLM-as-router** (see `REARCH.md` and `CLAUDE.md`). One
+**Target design: LLM-as-router** (see `docs/rearchitecture.md` and `CLAUDE.md`). One
 resident model emits GBNF-constrained structured JSON, and the same model
 phrases replies; the embedder is a RAG hint, not a routing gate. The
 committed default today is the classifier/embedder cascade, which is an
@@ -655,7 +657,7 @@ daemon.
 The voice wire protocol (length-prefixed JSON frames over TCP) is designed for
 **multiple client implementations**. The reference PWA at `cmd/mavweb` is one
 client; any app (phone, desktop CLI, smartwatch) can implement the same frame
-protocol. The published spec is `PROTOCOL.md` — **generated from
+protocol. The published spec is `docs/protocol.md` — **generated from
 `internal/voice/wire.go`**, not composed freehand, so it can't drift from
 code. It covers transport (4-byte big-endian length prefix), methods
 (`PushToTalk`, `Pong`), push kinds (`AudioNudge`), surface identity
@@ -670,8 +672,8 @@ Broadening to home automation, media or comms is JSON, not code.
 
 ## Execution ledger
 
-Condensed from `ROADMAP.md` (2026-07-06). The live queue is
-`20-07-2026-BACKLOG.md`; current state is `PROGRESS.md`.
+Condensed from `ROADMAP.md` (2026-07-06). The live queue is the Vikunja board
+(project Maven, ID 2); this table is history, not a work list.
 
 | # | Item | Prio | Status |
 |---|------|------|--------|
@@ -755,7 +757,7 @@ Kept for provenance. **None of this is the current or intended design.**
   stay deterministic — "classifier owns the route, the SLM stays in its
   phrasing lane" — with an embedding + nearest-centroid stage 1 over ~10
   examples per intent, and misroutes appended as new centroid examples.
-  *Replaced by* LLM-as-router (`REARCH.md`): one resident model emits
+  *Replaced by* LLM-as-router (`docs/rearchitecture.md`): one resident model emits
   GBNF-constrained JSON and also phrases replies; the embedder is demoted to
   a RAG hint. *Landed 2026-07-31:* the LLM router is on by default and set
   `true` in `deploy/mavend.json`. The classifier cascade stays as the failure
@@ -774,7 +776,7 @@ Kept for provenance. **None of this is the current or intended design.**
   *Resolved 2026-07-30 (#318), revised 2026-07-31:* the resident checkpoint is
   stock **Qwen3-1.7B** (`UD-Q4_K_XL`, `n_ctx` 4096), which replaced
   Qwen3.5-0.8B after measuring better on both fixtures
-  (`MODEL-BAKEOFF-31-07-2026.md`). The CPT'd **Qwen3-1.7B** remains the target
+  (`docs/evals/2026-07-31-model-bakeoff.md`). The CPT'd **Qwen3-1.7B** remains the target
   (#122); what stock gets wrong is the persona, not the Russian. Note the resident
   model is no longer described as untrained — the target is trained
   end-to-end, which is the substantive change from the old claim.

docs/determinism.md

@@ -1,5 +1,7 @@
 # Deterministic logic around a small model
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 Written 2026-08-02. Branch `fix/integrated`.
 
 ## The question
@@ -268,7 +270,7 @@ rebuilt `mavend` wires it. "кто написал войну и мир?" now rou
 that turn left unsettled.
 
 - **The turn was slow, and nobody knows yet whether that is real.** Route 7s,
-  search 1s, phrasing 15s. The p50 in `ROUTING-EVAL-31-07-2026.md` is 825ms. It
+  search 1s, phrasing 15s. The p50 in `docs/evals/2026-07-31-routing.md` is 825ms. It
   was the first turn after a cold start with the model still warming, so it
   proves nothing either way. Re-run the same question warm before treating it as
   a regression. Do not plan latency work off this number.

docs/ecosystem.md

@@ -1,5 +1,7 @@
 # Maven Ecosystem Architecture
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 ## 1. Purpose
 
 This document defines Maven's role in the local ecosystem formed by:

docs/evals/2026-07-31-model-bakeoff.md

@@ -16,7 +16,7 @@ with Qwen3-1.7B.
 
 Settles Vikunja **#278 / #250**.
 
-- Same fixture and scorer as `ROUTING-EVAL-31-07-2026.md`: `internal/router/eval/`
+- Same fixture and scorer as `docs/evals/2026-07-31-routing.md`: `internal/router/eval/`
   (`ru_routing_v1.json`, 76 held-out cases).
 - Reproduce: `MAVEN_LLM_URL=http://127.0.0.1:<port> make eval-router`
   (`TestLLMRouterBaseline`). (This line used to say there is no `make eval-models` target.
@@ -148,7 +148,7 @@ Qwen3-1.7B wins every column, including against a model 20% larger than it.
 | ontopic | 16, 19, 19 | **22, 23, 23** |
 | canned fallbacks | 8, 5, 6 | **0, 2, 0** |
 
-This also fills the row `TALK-EVAL-31-07-2026.md` had to void for contamination:
+This also fills the row `docs/evals/2026-07-31-talk.md` had to void for contamination:
 **600ch/1024tok on Qwen3.5-0.8B scores 13, 11, 8.**
 
 `address` is the headline. It sat at 18-22 of 27 on the 0.8B no matter how the prompt
@@ -164,7 +164,7 @@ The 1.7B does that 0-2 times.
 
 > **Stale, corrected 2026-08-02.** The p50 figures in this table are contention on a
 > shared llama-server, not the model's cost. The router measures p50 825ms / p95 1.2s /
-> max 3.0s in `ROUTING-EVAL-31-07-2026.md`, which says so at line 61. Read this table for
+> max 3.0s in `docs/evals/2026-07-31-routing.md`, which says so at line 61. Read this table for
 > the shape of the tail only. Take absolute latency from the routing eval.
 
 | | p50 | p95 |
@@ -220,11 +220,11 @@ swapped again when the CPT lands.
   behind `voice.llm_router`, the default is on, and `deploy/mavend.json` sets it `true`.
   These numbers are the production path now. **Corrected 2026-08-02: the p50 ≈2.7s in the
   latency table above WAS a bench artifact.** It is contention on the shared llama-server,
-  not the model. `ROUTING-EVAL-31-07-2026.md` line 61 says so, and measures the router at
+  not the model. `docs/evals/2026-07-31-routing.md` line 61 says so, and measures the router at
   p50 825ms / p95 1.2s / max 3.0s. Cite that file for latency, not this one.
 - ~~`/mnt/hdd1/llms/LFM2.5/Qwen3-1.7B-UD-Q4_K_XL.gguf` is a 293 MB truncated download
   in the wrong directory.~~ **Deleted 2026-07-31.** The good 1.13 GB copy in `qwen3/` is
   what `deploy/mavend.json` loads.
 - Harness: `scratchpad/bakeoff.sh`, one server at a time, health-checked before each
   run, `/v1/models` recorded per run. Never run two LLM consumers at once — see the
-  contamination note in `TALK-EVAL-31-07-2026.md`.
+  contamination note in `docs/evals/2026-07-31-talk.md`.

docs/evals/2026-07-31-phrasing.md

@@ -1,7 +1,7 @@
 # Phrasing evaluation — 31-07-2026
 
 How Maven words a nudge, measured instead of argued. Counterpart to
-`ROUTING-EVAL-31-07-2026.md`.
+`docs/evals/2026-07-31-routing.md`.
 
 - Fixture + scorer: `internal/phraser/eval/` (`nudges_v1.json`, 15 cases; `eval.go`, `checks.go`)
 - Reproduce: `MAVEN_LLM_URL=http://127.0.0.1:18099 make eval-phrasing`

docs/evals/2026-07-31-recall.md

@@ -65,7 +65,7 @@ prefixes) is the targeted fix, and it would move findings 1 and 2 together. Sepa
 `model_quantized.onnx` — not the same file.
 
 `hard` cases score **2/11**: every one is a query where the operator did not reuse his own words.
-That is the normal case weeks later, and exactly what DESIGN.md's "recall when relevant" promises.
+That is the normal case weeks later, and exactly what docs/design.md's "recall when relevant" promises.
 
 ### 4. The memory-store recall branch is dead for notes
 

docs/evals/2026-07-31-routing.md

@@ -177,7 +177,7 @@ was silent ("не знаю" to "который час") while the one it introdu
 
 ### 1. The resident model does route better — 50.0% vs 36.8%
 
-REARCH.md's premise holds; `voice.go:211`'s comment does not. **But the classifier is only
+docs/rearchitecture.md's premise holds; `voice.go:211`'s comment does not. **But the classifier is only
 ~37% correct on held-out utterances, and the model only ~50%.** Neither is "reliable". The
 gap between them is real but both are far from a system you would describe as working.
 

docs/evals/2026-07-31-talk.md

@@ -141,7 +141,7 @@ a model check, which catches a dead server but not a loaded one.
   measured) on this fixture and the router fixture. Not the 4B — too big for
   this box, owner's call.
 - Newer sub-500M candidates (LFM2.5 200M/300M) are worth a run for routing.
-  Note `MODEL-BAKEOFF-31-07-2026.md` found LFM2.5-**1.2B** worse than
+  Note `docs/evals/2026-07-31-model-bakeoff.md` found LFM2.5-**1.2B** worse than
   Qwen3.5-0.8B at Russian routing and 2.4× slower — but those are a different,
   older generation, so that result does not predict the small ones.
 - Fix `chat-how-are-you`'s `want_any`, and re-baseline once, so `ontopic`

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

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

docs/offload.md

@@ -0,0 +1,179 @@
+# Offloading model work to the workstation
+
+*Last verified: 2026-08-03 @ 12530c8. Living doc: correct it in place, do not append.*
+
+Owner's call, 2026-08-02. Vikunja #483 is the umbrella. Tasks #484 to #487 are the
+work, and this file holds the shape and the rules all four must obey.
+
+## The goal
+
+homesrv cannot grow a GPU. The workstation has 16GB of VRAM. Move the model work
+to the workstation and leave homesrv running the logic that must be always-on,
+deterministic and cheap.
+
+## Why this is tractable
+
+The split already exists structurally. `mavsttd` and `mavttsd` are separate
+daemons that core reaches over a socket, not linked libraries. Moving them off-box
+is a transport change, not a redesign.
+
+The microphone is at the workstation, because that is where the owner sits and
+homesrv is headless. So speech-to-text and the wake word are already on the
+workstation side by construction. Audio never has to cross the LAN. Only the core
+turn does.
+
+## The constraint that shapes everything
+
+The workstation's GPU is often busy: CPT runs, experiments, Correx, the manga-recap
+pipeline. It also sleeps. homesrv does not.
+
+So an offloaded model is never *the* model. It is the preferred one, with a floor
+on homesrv. That is the shape the cascade already has, where a router error falls
+through to the classifier.
+
+## The degradation rule
+
+Two cases, and the line between them is sharp.
+
+**Fall back silently** when the workstation model would only do the job *better*:
+routing, phrasing, a nudge. Falling back costs nothing that exists today, because
+the resident Qwen3-1.7B is today's production quality. The owner should not be told
+that his reply was phrased by the smaller model.
+
+**Name the gap** when the resident model cannot do the job *at all*. A world
+question that a 1.7B answers by inventing is the case. A wrong answer is worse
+than "не могу сейчас". This is the rule CLAUDE.md already states for a sibling
+service being down.
+
+Nothing in between. A turn never breaks on the workstation being asleep.
+
+Both halves are wired, 03-08-2026. `LLMPhraser.PhraseWorld`
+(`internal/phraser/world.go`) is the naming half and has three outcomes, not two:
+
+| State | What he hears |
+|---|---|
+| no `workstation` block | the resident model answers, exactly as before the seam existed |
+| configured, card free | the workstation answers |
+| configured, asleep or busy | the gap, `worldGap` in `cmd/mavend/worldmodel.go` |
+
+The first row is the one worth stating. Naming a gap requires a gap. On a box with
+no second model the 1.7B is the whole product. Refusing every world question there
+would remove a capability the owner has today.
+
+A source holding a passage is on the naming half too: a live search, a ZIM
+article, a page he named. None of them says "не могу сейчас". They read the
+passage back, which is what `phraseSource` returning `""` selects. A real quote
+beats a gap, and neither path invents.
+
+## Admission control, not a scheduler
+
+There is no GPU arbiter. That is a service with its own failure modes, and nothing
+here needs work *distributed*. It needs admission control. The workstation
+advertises free VRAM over a health endpoint, and Maven treats it as one more query
+source that claims a turn or passes. llama-server also refuses to load when VRAM is
+short, so the failure is detectable without cooperation from the owner's other
+jobs.
+
+The caller must be able to ask "is this peer usable right now" without a turn
+hanging on a timeout. A dead remote is a normal state, not an error state.
+`internal/llm.Pair` is that check on the Maven side. A prober caches the answer,
+so `Available()` is an atomic read and no turn pays for a health check.
+
+llama-server does not stay up on the workstation. It cannot: a resident 7-14B
+would hold 16GB against the owner's CPT runs. So a supervisor there owns its
+lifecycle, keeps it loaded while the card is free, and unloads it on idle or
+when another process needs the card (owner's call, 2026-08-02, Vikunja #488).
+
+That supervisor is still not a scheduler, and the distinction is worth holding.
+It arbitrates nothing between callers. It reports whether it can take work and
+manages one process to back that answer. Maven never asks it to start anything
+and never learns that it did.
+
+Contention is decided by presence under `/sys/class/kfd/kfd/proc`, not by a VRAM
+threshold. A ROCm process registers there when it initialises HIP, before it
+allocates anything. So the supervisor sees a contender during that job's startup,
+and yields before the job loses the memory it asked for. A
+threshold reads the card too late. By the time free VRAM has dropped, the other
+job has already lost the allocation race. Free VRAM is still read, but only as a
+precondition for loading, never as the eviction signal. One blind spot is known.
+A job can take the card without registering on the KFD, as a Vulkan or a
+video-decode job would. `describe()` logs every contender's comm, and that log is
+how we find out whether the blind spot is real.
+
+`mavgpud` runs from a systemd unit on the workstation with
+`deploy/mavgpud.json` as its config, and `llama_args` is passed to llama-server
+untouched. The model, the context size, the layer count and the MTP flags are the
+owner's business and not this daemon's schema.
+
+## What stays on homesrv, permanently
+
+The **embedder** (multilingual-e5-small, ONNX, CPU). It backs the classifier, which
+must answer while the GPU is saturated. It is also cheap enough on CPU that moving
+it buys nothing. Four callers:
+
+| Caller | What for |
+|---|---|
+| `internal/router/classifier.go` | the routing floor |
+| `cmd/mavend/actions_query.go` (`queryEmbed`) | memory recall |
+| `cmd/mavend/feeds.go` | ingest embedding for every RSS item |
+| `internal/crawl/watch.go` | ingest embedding for every crawled page |
+
+`internal/speaker` becomes a fifth once it lands.
+
+## Inventory: what runs a model on homesrv today
+
+The **resident model** is one llama-server with seven callers, and 03-08-2026 is
+the date each of them stopped or did not stop being resident-only:
+
+| Caller | What for | Offloaded |
+|---|---|---|
+| `cmd/mavend/voicewire.go` | routing | silently, through `hot` |
+| `cmd/mavend/replier_llm.go` | replies | silently, through `hot` |
+| `cmd/mavend/tick.go` | digestion worker: `PhraseNudge`, `PhraseReminder` | silently, inside the phraser |
+| `cmd/mavend/actions_query.go` | world questions, and any fetched passage | names the gap |
+| `cmd/mavend/capture.go` | capture summarisation (unreachable, see #480) | no, holds its own client |
+| `cmd/mavend/mail.go` | mail extraction (off, no IMAP) | no, holds its own client |
+| `memoryeval.go`, `modelswap.go` | admin and evals | no, and deliberately |
+
+The last three rows are resident-only on purpose. `memoryeval.go` and
+`modelswap.go` measure and swap the resident model, so sending their work
+elsewhere would measure the wrong thing. `capture.go` and `mail.go` are
+background jobs that hold a gated background client (`llmBackgroundClientFor`),
+and that priority has no equivalent on the remote yet. Both are also unreachable
+on this deploy, so wiring them would ship an untestable path.
+
+The `tick.go` row needs one caveat. `phraser.llm_nudges` is `false` in deploy, so
+nudges come from templates and the seam under them changes nothing until that
+flips. It is wired anyway: `PhraseReminder` is on the same transport and is on.
+
+Then the embedder above, **whisper.cpp** in `mavsttd`, and **piper** in `mavttsd`.
+`mavwaked` uses no model at all: an energy-threshold VAD over 30ms frames.
+
+## Order
+
+1. **Transport** (#484). Nothing else is possible until a seam can cross a host.
+   `internal/netaddr` landed in PR #92. A seam address now carries its own scheme,
+   and a scheme-less one is still unix. A tcp seam requires a shared token, because
+   the filesystem permission that authenticated the unix socket is gone.
+2. **The resident model** (#485, #490). Wired. A `workstation` block builds an
+   `llm.Pair` in `modelSeam` (`cmd/mavend/voicewire.go`), routing and replies
+   complete through it, and the phraser holds the same pair (`UseRemote`). Both
+   halves of the rule are live: see the table above for which caller gets which.
+   Measured, `docs/evals/2026-08-02-workstation-gemma4-12b.md`: gemma-4-12b
+   through the cascade scores 84.4% full accuracy at p50 329ms. The resident
+   model scores 72.7% at p50 0.80-1.04s. On the talk fixture it is 25/27
+   against 20/27. Biggest quality delta. A 16GB card runs a 7-14B,
+   which fixes what the 1.7B gets wrong: world knowledge, and the persona the CPT
+   targets. The degradation path is already written and measured, since the
+   classifier scores 68.8% full accuracy at p50 16.6µs on its own.
+3. **Speech-to-text and text-to-speech** (#486). They gain a real margin, but on
+   quality alone, and both already work.
+4. **The wake word** (#487). Independent of all of the above.
+
+## Assumptions
+
+- The LAN is trusted enough that wireguard is supported but not required (owner's
+  call). What crosses the wire is still his utterances. That is why the tcp seam
+  carries its own token instead of assuming a network boundary.
+- The workstation is not expected to be up. Every child task must still serve a
+  turn while it is down.

docs/operations.md

@@ -1,5 +1,7 @@
 # Start Commands
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 All commands assume `ROOT=/home/kami/apps/Maven` and the local Go toolchain at `$ROOT/deps/go/go/bin/go`.
 
 ## Prerequisites

docs/plans/2026-07-10-router-lfm-foundation.md

@@ -6,7 +6,7 @@
 > use the embedded LFM model paths or old single-object examples as current ops
 > guidance; see `2026-07-18-qwen3-resident-training-eval.md`.
 
-> Scope from `REARCH.md`. Make Maven trustworthy: the LFM becomes the router
+> Scope from `docs/rearchitecture.md`. Make Maven trustworthy: the LFM becomes the router
 > (fixes "messes up queries" / "doesn't take notes"), the engine actually runs
 > (fixes stub replies), dates stop being read as "number dot number dot number",
 > and telegram becomes a reach channel. NOT in scope: on-demand 4B reasoner,
@@ -693,7 +693,7 @@ ssh kami@192.168.1.104 'curl -s localhost:9201/api/chat -d "{\"text\":\"запо
 4. `docker compose up -d mavend && docker logs -f maven-mavend-1` — confirm the
    phraser spawns and no `phraser: NewStub` path. Run the two verify curls.
 5. Update `AGENTS.md`: LFM model download + note that routing is now LFM-first
-   with classifier fallback (`REARCH.md` is the design of record).
+   with classifier fallback (`docs/rearchitecture.md` is the design of record).
 
 ---
 

docs/protocol.md

@@ -1,5 +1,7 @@
 # Maven Voice Protocol
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 > Auto-generated from `internal/voice/wire.go`, `internal/voice/errors.go`,
 > `internal/voice/frame.go`, `internal/voice/client.go`. If this file and
 > those files disagree, the code wins.

docs/qa.md

@@ -0,0 +1,566 @@
+# QA plan: checking Maven properly
+
+*Last verified: 2026-08-02 @ 20aa2d5. Living doc: correct it in place, do not append.*
+
+Written 2026-08-01, after the 35-PR stack landed and the box came back up.
+Refreshed 2026-08-02 against the live list, after PRs #85-#90.
+
+42 of the 50 open Vikunja tasks are `QA:` tasks. They are verification work, not
+build work. Most sat unverifiable while Maven was down for 11 days. That
+blocker is gone.
+
+The plan as written on 2026-08-01 named 40 task numbers. Ten open `QA:` tasks were
+missing and two of the named ones had closed. Every open task now appears below,
+the eight non-QA ones in the last two sections.
+
+This plan orders them by what unblocks what. Do sessions 1 and 2 first. Almost everything
+downstream assumes the voice loop works, and nobody has confirmed that since
+the redeploy.
+
+---
+
+## What the 02-08-2026 run found
+
+Sessions 1, 2 and 3 all ran. Read these five before picking anything up.
+
+- **470: a question writes invented knowledge into memory.** Recall then serves
+  it back. `что дальше?` lands on `IntentFact` and stores the model's answer as a
+  `self` fact at confidence 1.00. Two junk rows then claimed seven unrelated
+  world questions through recall, outranking the search leg. A question about the
+  capital of Australia was answered `какая последняя версия языка Go?`. Two bad
+  writes silently disabled world answering, with nothing logged.
+- **466: a pending clarify is global.** One unanswerable clarify swallowed the
+  next three utterances from three separate sessions. With ntfy, telegram and
+  voice all live, a clarify raised on web chat eats the next telegram message.
+- **467: spoken task capture is dead.** The router calls the capture marker an
+  `act`, and capture is reachable only from the `note` intent.
+- **The classifier baseline in this repo was wrong**, and it flattered the
+  router. See session 2 and **464**.
+- **477: the model swap and the self-update cannot be triggered on this box.**
+  Both are built and both are correct in test. The swap needs a passkey and
+  WebAuthn is unconfigured. `mavupdate` needs to reach a socket that only an
+  in-container uid can open.
+
+- **479: an unconfigured capability lets the question escape to web search.**
+  Netscan off, asked `какие устройства в сети?`. She answered from the live web
+  with a general article about network hardware. A question about his LAN went to
+  an upstream engine. The crawler fails the same way.
+
+Twenty-one defects were filed on 02-08-2026: 462 through 482. Six tasks this plan
+had written off as blocked turned out to be ready to check. All six ran. Every
+one of them is code-correct and stops at the deploy.
+
+Three of the five config blockers in **472** were then cleared. The morning
+routine, ambient ingest, feeds, the crawler and netscan are all live. Two remain,
+and both are the owner's call: a token for each ecosystem sibling, and seed data
+in Nexus and Praxis.
+
+---
+
+## Before you start
+
+Two things bite anyone running these checks on homesrv.
+
+**curl needs `--noproxy '*'`.** The shell exports `http_proxy=http://127.0.0.1:18080`.
+Without the flag, every local check returns 503 from the proxy and looks like a
+dead service. This cost me a false regression report today.
+
+**The database is not readable with sqlite3.** Four older QA steps say
+`docker compose exec mavend sqlite3 /data/maven.db "select ..."`. That cannot
+work: the container has no `sqlite3` binary, and the store is AES-256-GCM at
+rest with a tmpfs working copy. Read state through mavweb instead, at
+`/history`, `/trace`, `/routines` and `/dash`.
+
+---
+
+## Session 1: the voice loop (half a day)
+
+Nothing here has been confirmed since the redeploy, and everything else assumes
+it works. Do this first.
+
+Closes or advances: **44** (conversation), **45** (text chat), **287** (voice
+session quality), **321** steps 3-5 (quiet mode), **288** (STT golden audio).
+
+**288 is not blocked.** The fixtures are committed under `cmd/mavsttd/testdata/`
+and `make test-stt-golden` runs today. This plan said otherwise until 02-08-2026.
+
+Steps 1 and 3-6 were run on 02-08-2026 and pass. Steps 2 and 7-9 still need a
+person at the box, because they need a microphone or a nudge to arrive.
+
+Steps 1 and 3-6 do not need a browser. `POST /api/chat` takes a form-encoded
+`text=` field and a cookie jar, and answers with the rendered `/chat` page:
+
+```sh
+curl -s --noproxy '*' -c jar -b jar -L -X POST \
+  http://127.0.0.1:9201/api/chat --data-urlencode 'text=привет'
+```
+
+Parse the whole page, not the last text node. The page carries nav and footer
+text. A naive tail of the Cyrillic nodes returns the wrong string, which makes
+turns look misaligned when they are not.
+
+1. Open `http://127.0.0.1:9201/chat` and hold a short conversation in Russian.
+   Watch for three things: she answers in feminine forms (`рада`, `поняла`), she
+   says `ты` and never `вы`, and no pet names appear.
+   **Passes** (02-08-2026, five turns): `я рада`, `поняла`, `помогла`,
+   `проверила`, `записала`, `грустна`, `ты` throughout, no pet names.
+2. Press push-to-talk on `/dash`. Say `привет`. Confirm a spoken reply comes
+   back. This is the only check that covers mic to STT to core to TTS to
+   speaker as one path. It is also the path the eleven-day outage most likely
+   broke.
+3. Say `тихий режим`. Expect `тихий режим включён. буду реже напоминать.` **Passes.**
+4. Say `выключи тихий режим`. Expect `тихий режим выключен.` Negation must win. **Passes.**
+5. Say `в комнате тихо`. Quiet mode must NOT flip. Confirm on `/history` that no
+   `quiet_hours` fact was written. **Passes**: no row written. She answers `пока
+   не умею отвечать на этот вопрос.`, so it lands on `IntentSystem` with no arm.
+6. Say `включи режим тишины`, then `сделай потише`. Both must flip quiet mode
+   on. These are the noun form and the comparative, added 01-08-2026. **Both pass.**
+7. Wait for a nudge, then say `потом` within twenty minutes. Expect `хорошо,
+   вернусь к этому позже.` and the nudge row on `/notifications` reading
+   `snoozed`. Say `потом` again with nothing pending: it must route as an
+   ordinary utterance, not be swallowed.
+8. Wait for the water nudge, then say `выпил воды`. Expect the ordinary fact
+   reply and nothing extra. She must not congratulate you. Check
+   `/notifications`: the row reads `acted`. Then trigger another nudge and say
+   `готово`. Expect `отлично, отметила.` and the same outcome.
+9. Note anything where she is slow, cuts off, or talks over herself. That is
+   287's whole content and it has no written acceptance criteria yet.
+   **First evidence, in text** (02-08-2026): nothing breaks, but answers wander
+   and stitch unrelated topics. Asked whether he should move flats, she opened
+   with the weather. That is 287, and it is a phrasing problem, not a loop problem.
+
+**The wake path cannot be checked as deployed.** `mavwaked` and `mavenclient`
+appear in no compose file and run as no host process. Step 2 covers only
+push-to-talk, from `/dash` through mavsttd and mavttsd. Wake word and VAD
+are untested by construction. Decide whether they belong in compose or on a
+client machine, and say which in the deploy docs. Tracked as **463**.
+
+**319's single-token bug is fixed** (01-08-2026). Single-word Russian utterances no longer come
+back as `не совсем поняла — можешь переформулировать?`. `привет` and `поужинал`
+both pass now: `thinSingleToken` spares social singles and any token carrying a
+verb ending, and only thins a bare nominal like `вода`. A one-word utterance that
+still gets clarified in this session is a new case for the lexicon, not the old bug.
+
+---
+
+## Session 2: measurement (half a day, mostly waiting)
+
+Closes or advances: **320** items 2-4, **278** (make the eval lab routine).
+Also **248** (memory evaluation), **319** (the margin gate) and **323** (the
+startup timeout arm).
+
+The resident llama-server cannot be reached by the eval harness. It binds
+`--host 127.0.0.1 --port 0` inside the container, so the port is kernel-assigned
+and never published. Start a second one on a fixed port instead:
+
+```sh
+llama-server -m /mnt/hdd1/llms/qwen3/Qwen3-1.7B-UD-Q4_K_XL.gguf \
+  --host 127.0.0.1 --port 18100 -c 4096 -ngl 99 --no-webui
+```
+
+`-c 4096` matters. The recorded numbers were measured at that context size, and
+a mismatch invalidates the comparison.
+
+Then:
+
+```sh
+make eval-models MAVEN_LLM_URL=http://127.0.0.1:18100   # want ~72.7% cascade
+make eval-router                                         # classifier baseline
+MAVEN_LLM_URL=http://127.0.0.1:18100 make eval-phrasing  # persona checks, slow
+make eval-recall
+```
+
+A large miss against 72.7% means the deploy differs from the bench harness.
+
+**Run on 02-08-2026 @ af9d213. The deploy matches the bench.** `eval-models`
+scored 56 of 77: 72.7% full, 77.9% intent-only, 2 false clarifies and 1 missed.
+That is the recorded figure to the decimal, and calendar sat at 2 of 2, so the
+stage 0 agenda rules hold. `eval-phrasing` scored 21 of 27 on the talk fixture
+against a recorded 20, and the 15 nudge templates passed every check.
+
+Two numbers in this repo were wrong, and both flattered the resident model.
+
+- **The classifier is not 36.8% and not 31ms.** `make eval-router` reports
+  `classifier+onnx: 53/77 (68.8% full)` at p50 16.6µs. The figure repeated here
+  and in `CLAUDE.md` predates the stage 0 rules and the seed additions. Both now
+  score inside that baseline. The accuracy gap the router buys is
+  roughly 4 points, not 36. Re-argue the trade on the real numbers: **464**.
+- **Router latency was measured under contention again.** p50 1.126s, p95 1.58s,
+  max 3.24s, against a recorded p50 825ms. The resident model was serving the
+  daemon on the same iGPU throughout. Do not record this as a regression, and do
+  not record it as a measurement either. Stop the stack before timing the router.
+
+`classifier+hash` scores 19.5%, which is the no-ONNX degraded path and is not the
+failure floor the deploy uses. Do not quote it as the classifier baseline.
+
+Then three things to decide while the numbers are in front of you:
+
+- **319 is done.** 359 gave the LLM path a real confidence signal.
+  `thinSingleToken` was narrowed on 01-08-2026, and agenda questions moved to
+  stage 0. Missed clarify sits at 1 of 6 and false clarifies at 2. Item 2 point 2
+  closed on 02-08-2026: the `make eval-recall` margin sweep is the distribution
+  that was asked for, and `0.008` sits at the knee.
+
+  | delta | answered | false recall |
+  |---|---|---|
+  | 0.005 | 18/27 | 2/5 |
+  | **0.008** | **18/27** | **1/5** |
+  | 0.010 | 16/27 | 1/5 |
+
+  It removes four of five false recalls at no cost in answers, and the next step
+  costs two answers for nothing. The hand-picked value survives on evidence.
+- **278's real ask** is making the eval lab routine rather than building it. It
+  is built. Decide whether it runs on a timer, on every merge, or on demand, and
+  the task can close.
+- **248** is the memory evaluation loop. It ships, it writes notes, and it cannot
+  speak. `make eval-recall` covers the retrieval half. The open question is whether
+  a written evaluation nobody reads is worth the tick.
+
+**323 is down to one check.** PR #90 covered the spawn path and took phraser
+coverage to 76.9%. Only the 60s startup timeout arm is untested, because testing it
+needs a `StartupTimeout` field on `Config` rather than a test-only hack. While you
+are on the box, time a cold 1.7B load off spinning disk. If it runs near 60s, the
+default is too tight and the field earns itself twice.
+
+Warm, it is nowhere near. A second llama-server answered `/health` 1.8s after
+launch at `n_ctx 4096` on 02-08-2026. That is page cache, so it does not settle
+the question. A cold read needs a cache drop, which needs root.
+
+**`CheckFeminine` has a false positive.** On 02-08-2026 it failed
+`query-notes-do-not-answer` for `ты заплатил`, calling it masculine
+self-reference. Masculine second person is correct, because the owner is male.
+The check matches a masculine
+past-tense verb before `за` without confirming the subject is `я`. Fix it in
+`internal/phraser/eval/checks.go` before trusting a phrasing score to the case.
+The real talk-fixture score on that run is 22 of 27, not 21. Tracked as **462**.
+
+Item 4 of **320** needs a permission I do not have. Kill the `llama-server`
+pid under `maven-mavend-1`, post a turn, and confirm it still completes
+through the classifier. Either grant it or run it yourself. It is the only
+check that the failure floor catches a mid-session model death.
+
+---
+
+## Session 3: the interaction batch (a day, or five sittings)
+
+These need real use rather than a command, grouped by what one sitting covers.
+
+**Morning and delivery** (**280**, **281**, **128**, **282**, **283**, **285**):
+open `/morning`, walk the seven required behaviours, then check the four
+interruption outcomes and the digest gap. **282** needs the `desk_active` script
+enabled on the desk PC first, which is **15** and needs you at that machine.
+**283** is the event intake envelope every reach shares, so a delivery check
+exercises it whether you name it or not. **285** is not verification: the bridge
+framework works and the remaining ask is more adapters. Decide which reach comes
+next, or park it.
+
+Run 02-08-2026. **280 is blocked.** No morning routine is configured (**472**).
+`morning.Item` also has no required-versus-optional field, so behaviour 1 cannot
+hold whatever you configure (**473**). **281's digest gap is closed**, and
+its presence rule passes on inspection. Three of its five items need traffic the
+box has not had. **283 is blocked**: nothing feeds the intake journal. **128
+found the worst defect of the whole session, see below.**
+
+Three of 472's five blockers were cleared the same day, in `deploy/mavend.json`
+and `docker-compose.yml`.
+
+- A `morning_routines` block, one routine `утро` 08:00-11:00 with medicine,
+  water and pets. It is live: the dispatcher logged `dropped morning:утро (sev1,
+  presence=away)`, so the plan builds and the nudge is proposed. 280's
+  behaviours and 128 step 11 are checkable now. 473 still stands.
+- `-ambient-token` on mavweb, value in a gitignored `/.env` that docker compose
+  reads for interpolation. `/api/ambient` answers 401 without the token and 201
+  with it, storing `calendar_event_20260802_Standup`. 283 step 5 and 128 step 8
+  are unblocked. The token is a flag, so it shows in `ps` inside that container.
+  The zenmoney and IMAP secrets are read from files instead. Ingest also
+  reads the notification's wall clock as UTC and stores a 14:30 meeting at 18:30
+  (**482**).
+- `feeds` (two sources), `crawl.on_demand` and `netscan.enabled`. The intake
+  journal now fills: `/events` holds `scan:lan` and `ambient:notif` rows.
+
+Two are not mine to clear. No sibling has a `token` in `deploy/mavend.json`, so
+273 steps 6 and 8 need a credential decision. Nexus has no entities and Praxis no
+attention items, so 272 step 3 needs seed data whose content is the owner's call.
+
+For **285**, two facts bear on the choice. Synapse is already running on this box
+and healthy, so a Matrix reach has a live target and needs no new service. And
+mavweb is already a PWA with a service worker, which 285 itself calls the highest
+value adapter left. Today's reaches are ntfy, telegram and voice.
+
+**Query sources** (**258**, **286**): ask her something the RSS feeds answer and
+something only a ZIM answers, with the search block on. Live search leads and the
+ZIMs are the fallback since 02-08-2026. **286**'s remaining half is doc and
+git ingestion, which is build work, not a check.
+
+**Do not read `/trace` for this.** `/trace` is the nudge-rule trace: rule,
+severity, predicate, gate, selected. No query-source field exists anywhere in the
+codebase. The only evidence of which query source claimed a turn is the
+`voice: search:` and `voice: kiwix:` lines in `docker compose logs mavend`
+(`actions_query.go:589` and `:660`).
+
+Run 02-08-2026, 20 turns. **Search leads and the personal boundary holds.** Every
+world question that reached the boundary was claimed by search. All three
+personal questions produced no search and no kiwix line at all.
+
+The rest of this sitting went badly. **Kiwix has zero live coverage.** SearXNG
+returns four results for everything, including two invented nonsense terms. So
+`querySearch` always claims, and Kiwix is unreachable code as deployed. The ZIM
+half of the 02-08-2026 decision is unverified. A ZIM answer cannot signal a
+silent search failure, because a ZIM answer cannot happen.
+**Ordering defects** in feeds and calendar, plus 258 step 1's utterance not
+working: **474**. And the sitting independently found stage 2 of **470**.
+
+**Tasks and calendar** (**129**, **130**, **127**, **126**, **246**): capture a
+task by voice, confirm it lands, check prioritisation ordering is not nonsense.
+**246** (mail reader) also exercises the `IngestMail` rung that moved to
+`AuthWrite` this morning.
+
+Run 02-08-2026. **129 passes.** The page and the spoken answer agree on ordering.
+The undistinguished task carries no invented reason on either surface, which is
+the thing 129 asks for. **130 fails outright** and **127 half fails**:
+**467**, **469**. **246 cannot be run**: `mavmaild` is commented out in
+`docker-compose.yml` and there is no `email` block, so nothing in steps 4-13 is
+reachable. The `IngestMail` rung does sit at `AuthWrite`
+(`internal/auth/policy.go:96`, asserted in `auth_test.go:421`), verified by
+reading only.
+
+**Routines and patterns** (**43**, **46**, **247**, **254**): these need history
+to detect against. If the database is thin after the outage, they may have
+nothing to propose, which is not a failure. Check `/routines` before
+concluding anything.
+
+Run 02-08-2026. The answer is the middle case: **the detector ran and found
+nothing.** The tick loop is live, and `detectPatterns` is called unconditionally
+at `cmd/mavend/tick.go:227`. It has run about 25 times since the restart. It
+finds nothing because the events table is empty upstream of it. Rows land there
+only from `pattern.Extract` at fact-write time, and `Extract` requires the fact
+value to match a closed 7-action lexicon. All 200 facts on `/history` are
+`page_heartbeat`, `netdata_alarm`, `quiet_hours`, `name`, `service_down` and
+`рост`. Not one lexicon hit, so no event can exist, let alone the four one pair
+needs. **46 step 5 passes**: `/routines` renders `noticed 0` with the empty state
+and the hint string.
+
+Two things block this sitting, and both are build work. The seeding recipe on
+**43** goes through `sqlite3` and cannot work. And `pattern.Detect` has no
+minimum-interval floor, so seeding by hand mints a permanent false routine
+(**468**). Do not try to seed a pattern with four fast chat turns.
+
+**Ecosystem** (**272**, **273**, **276**): nexus, hexis and praxis are wired and
+logged clean at boot.
+
+Run 02-08-2026, read-only half. All three answer `/health` 200 and `/ecosystem`
+lists 18 Hexis capabilities with correct read-only and mutating badges. **272 and
+273 are blocked on empty data**, not on code. Nexus holds no entities, Praxis
+holds no attention items, and the Calls panel has never recorded a call. See
+**472**, and read its warning first. 273's trace fix has never been validated
+here. An empty Calls panel is exactly what the old bug looked like. The page is
+`/ecosystem`, not `/siblings`.
+
+**276 ran 02-08-2026 and the suite is sound.** 17 `TestEcosystem_` cases pass
+under `-race`, not the 10 the task describes. The mutation check bites: patching
+the Nexus-error branch of `handleHexisAct` to `return ""` fails
+`TestEcosystem_MalformedNexusResponseFailsClosed` on the expected line.
+
+Steps 4 and 6 could not be checked through chat, because no utterance reaches
+Praxis (**475**). «что требует внимания» routes to `intent=query` and is answered
+by the search leg, identically whether `ecosystem-praxis-1` is up or stopped. The
+degraded string never appears because its branch is never entered. Step 5 is
+blocked the same way: `перезапусти muzick indexer` clarifies on
+`HasFn:false`, and the router had already rewritten the entity name to
+`музик индексер` (**476**).
+
+Both steps were checked on `/ecosystem` instead, which reads Praxis directly.
+With Praxis stopped the card reads `praxis — unreachable` while Nexus and Hexis
+keep rendering. On `docker start` the card returns to `nothing needs attention.`
+with no mavend restart. Independent degradation and recovery both hold.
+
+**Operations** (**249**, **250**): both ran 02-08-2026. The code is correct and
+neither lever can be pulled on this box. See **477**.
+
+**250** passes steps 1, 2, 3, 9 and 10 on the deploy. The capability announces
+itself. `/models` names the model llama-server reports, not the config filename.
+Asking her to switch models does nothing. Removing `swap_models` renders `swap
+not configured`. Step 4's refusal half passes at HTTP 403, and the 403 comes from
+mavend rather than mavweb. WebAuthn is unconfigured, so the web gate fails open
+and the wire gate fails closed. Steps 5 to 8 need a passkey assertion nothing on
+this box can produce. They pass in test: 13 swap cases and 7 page cases covering
+drain, mid-swap refusal, rollback, failed rollback and the not-owned refusal.
+
+**249** passes steps 1 and 2. Step 3 stops it. `mavupdate` health-checks over
+`/run/maven/mavend.sock`, which is `srw------- 1 10001 999` inside a docker
+volume. The host owner cannot traverse `/var/lib/docker/volumes` and cannot
+connect to a socket owned by an in-container uid. `mavupdate` assumes a
+host-installed daemon and the deploy is containers. Do not sudo around this.
+
+---
+
+## Housekeeping (done 02-08-2026, and this section was mostly wrong)
+
+This section claimed eleven tasks were not verification work. **Three were not.
+The other eight are.** Every one of the eight has shipped, tested code behind it.
+The error ran one way: it wrote off work that is ready to check. Do not trust a
+"nothing is built" line in this plan without grepping for the package first.
+
+Relabelled to `Blocked:`, claim verified:
+
+- **125** zenmoney. `internal/zenmoney/` ships and is tested against a fixture.
+  `deploy/zenmoney.token` does not exist and the compose mount is commented out.
+  One token unblocks it.
+- **256** Home Assistant. `internal/smarthome/` ships, the `smarthome` block sits
+  in `deploy/mavend.json` at `enabled: false`, and 8123 and 1883 are closed.
+- **14** cold-start unlock. The seam is real at `cmd/mavend/main.go:128` and
+  `internal/webauthn/prf.go` is in place. `lockedAPI` is gone, replaced by
+  `Server.Check` in `internal/ipc/server.go`. Gated on an authenticator that
+  implements the WebAuthn PRF extension, which is hardware, not code.
+
+Left alone, because the claim here was false:
+
+- **284** simulator. `cmd/mavend/simulator_test.go`, three scenarios under
+  `cmd/mavend/testdata/scenarios/`, and a `simulate` target at `Makefile:98`.
+  **Run 02-08-2026: all three scenarios pass**, plus the determinism and
+  backwards-step guards. One defect found, see below.
+- **288** STT golden audio. Four WAVs and `golden_v1.json` are committed under
+  `cmd/mavsttd/testdata/`, the make targets exist, and `models/stt/ggml-small.bin`
+  is on the box. Session 1 lists 288 as blocked on fixtures, which is wrong.
+  **Run 02-08-2026: all four pass**, WER at or under ceiling with no drift.
+
+| fixture | transcript | WER | ceiling |
+|---|---|---|---|
+| ru_reminder | `Напомни мне через час позвонить маме.` | 0.00 | 0.10 |
+| ru_fact | `А отметь, что я выпил воды.` | 0.20 | 0.25 |
+| ru_query | `Что у меня сегодня по календарю?` | 0.00 | 0.10 |
+| en_act | `Restart the web server and check the disk space.` | 0.00 | 0.10 |
+
+That also settles a session 1 worry indirectly: whisper.cpp works on Vulkan
+after the redeploy. Only the mic and the wake path remain unproven.
+
+**The simulator routes with an empty seed set.** Every `make simulate` run logs
+`loaded 0 seed examples from models/seeds`, seven times per scenario. The test
+runs from `cmd/mavend`, and the seed path is relative to the repo root. The
+scenarios still pass, which means they pass without the classifier having any
+seeds to match against. Whatever 284 is proving, it is not proving the routing
+the deploy runs. Fix the path before trusting a green simulator.
+- **257** Bluetooth. The bluez half is genuinely absent. The LAN-scan half shipped
+  (`internal/netscan/`), and steps 1-9 run today. Only step 10 is Bluetooth, so
+  relabelling the whole task would bury real pending work.
+- **251** MCP, **253** hearing, **259** crawler. All three ship
+  (`internal/mcp/`, `internal/capture/`, `internal/crawl/`) with no external gate.
+  Fully checkable. `259`'s step 1 wants no `crawl` block in `deploy/mavend.json`,
+  and there is none, so it is already set up correctly.
+- **252** vision and **255** speaker recognition. Both ship. Each is blocked only
+  on a model download: a vision gguf with mmproj, and a speaker embedding model.
+  Neither is present under `/mnt/hdd1`. Their refusal-path steps run today.
+
+So the honest split is three blocked on a credential or hardware, two blocked on
+a download, and six ready to check. That is roughly a session of real QA this
+plan had written off as backlog.
+
+**All six ran on 02-08-2026.** Every one of them is code-correct and stops at the
+deploy. The pattern repeats often enough to be the headline: the packages pass,
+and the box cannot reach them.
+
+**251, MCP.** Steps 1, 2, 3, 4 and 13 pass. Package tests green under `-race`.
+Off-by-default is clean, and the SSRF refusal is exact: without `allow_private`
+the log reads `refusing to connect to a private address: 127.0.0.1` and `/tools`
+shows the server down with zero proposals. Steps 5 to 12 are blocked. `ss -lntp`
+shows the Vikunja MCP server on `127.0.0.1:9100` only, so no container reaches it
+at any address (**478**). `allow_private` does work, measured both ways.
+
+**253, hearing.** Steps 1, 2 and 17 pass. `internal/capture` covers 90.3%. Steps
+7 to 16 are blocked on something nobody can work around: no shipped client calls
+`CaptureStart`. There is no `cmd/mavheard`, no mavweb route, and `mavenclient`
+never calls it (**480**). Two of its QA steps are also stale.
+
+**257, netscan.** Steps 2, 3 and 9 pass at unit level. Step 1 fails. Steps 4 to 8
+need the block enabled. Step 10 is Bluetooth and stays skipped.
+
+**259, crawler.** Steps 1 and 15 pass. Step 2 fails. Steps 3 to 14 need a `crawl`
+block that nobody has written.
+
+Both were configured later the same day, and both work. `netscan.enabled: true`
+answers `какие устройства в сети?` with `нашла 3 устройства, из них 2 с вебом, 2 с
+ssh. список записала.` and the scan lands in the intake journal as `scan:lan`.
+`crawl.on_demand: true` answers `посмотри https://lwn.net — что там пишут?` from
+the real page. So **479** is one defect, not the routing defect it was filed as.
+An unconfigured capability declines its own turn instead of naming the gap.
+Nothing is wrong with the routing.
+
+257 step 1 and 259 step 2 fail the same way and share a task (**479**). An
+unconfigured capability does not name the gap, so the question escapes to web
+search. `какие устройства в сети?` was answered with a general article about
+network hardware. That is his LAN going to an upstream engine.
+
+**252 vision and 255 speaker.** Both confirmed blocked. The disk claim was
+re-verified rather than taken on trust: 16 text-only ggufs under `/mnt/hdd1`, no
+mmproj and no speaker embedding model. Everything not needing the model passes,
+including the two refusals that matter. `TestNewLocalRefusesNonPrivateEndpoints`
+rejects `https://api.openai.com`, and forget really deletes
+(`internal/store/memory.go:145` is a real `DELETE`, not a tombstone). Vision is
+19/19, speaker 22/22, media 16/16.
+
+**470 got worse, then closed.** Both poisoned facts showed `voided` on
+`/history` and the defect survived. Re-measured at 15:42, after four restarts:
+`почему небо синее?` still answered `какая последняя версия языка Go?` with no
+`search:` line. What came back was the question he typed, not the value the fact
+held. So the poison was a vector in the memory index, and `revert` did not
+remove it.
+
+Repaired in two parts. 470 stopped the writes: a question is never a fact, and a
+void drops the key's vectors. 493 fixed what the index holds. A fact is indexed
+as the fact and not as the utterance, and a correction drops its superseded
+vector too.
+
+A poisoned box now repairs itself on the next start. `RepairFactVectors`
+re-embeds every fact vector from the fact it names, and deletes the voided and
+superseded ones. It runs once, guarded by a marker, and logs what it did.
+
+---
+
+## Needs you specifically
+
+Not QA. These are blocked on a decision or a credential only you have.
+
+| # | what |
+|---|---|
+| 16 | Create the Kuma API key. `-kuma-key uk5_mavpoll-key` in `docker-compose.yml` is still the placeholder. |
+| 15 | Deploy `desk_active` on the desk PC. Blocks **282**. |
+| 122 | Finish the CPT run for Qwen3-1.7B. The persona fix depends on it. |
+| 355 | Deploy the Hexis auth change. Was blocked on Maven being under construction, which it no longer is. The client half is vendored and wired. |
+| 357 | Decide whether entity-existence validation is the permanent target guard or whether blessing lands in Nexus. |
+| 275 | Hexis native API and MCP parity. |
+| — | Decide on `-require-stepup`. Making it the default needs WebAuthn configured first, or it locks you out of your own admin surfaces. |
+
+317 and 354 closed on 01-08-2026. The step-up gate now covers `POST /api/chat` and
+`/routines`, and the nginx template is locked down with a `maven.<domain>` block for
+mavweb. The `-require-stepup` default is still your call.
+
+---
+
+## Not this repo
+
+Two open tasks sit on the Maven board and are not Maven work. Move them or note
+where they land, so the board stops reading as 50 things Maven owes.
+
+- **358** replace the rowid execution cursor with a real seq column. This is Hexis,
+  and it must land before any execution retention or pruning does.
+- **362** mirror the router prompt reorder into the relabelling prompt. This is the
+  training workspace, enforced by `llm/check_prompt_parity.py` there, not here.
+
+---
+
+## Suggested order
+
+1. Session 1. If the voice loop is broken, nothing else matters.
+2. The `-require-stepup` and Kuma decisions. Five minutes, and it unblocks **16**.
+3. Session 2. **Run on 02-08-2026.** The numbers came back worse for the router
+   than the docs claimed. The classifier is 68.8%, not 36.8%, and 16.6µs, not
+   31ms. The router buys about 4 points of accuracy for four orders of magnitude
+   of latency. Whether that still earns its place is now an open question.
+4. Housekeeping. Cheap, and it makes the remaining backlog honest.
+5. Session 3, split whichever way suits you. All five sittings ran on
+   02-08-2026. Read the per-sitting notes before repeating any of them.
+
+The next thing to fix is not in this plan. Four defects say the same sentence:
+a capability is built and no utterance reaches it. **466** (a clarify is global),
+**467** (capture is act-routed), **475** (attention is act-routed), **476** (the
+router rewrites entity names). Routing is where the work is.

docs/rearchitecture.md

@@ -1,5 +1,7 @@
 # Maven — Re-architecture (Qwen3 resident model, revised 2026-07-18)
 
+*Last verified: 2026-08-02 @ 7079a24. Living doc: correct it in place, do not append.*
+
 > Supersedes the classifier-first routing model. Agreed in a design session
 > after diagnosing that homesrv deploys with a **stub phraser** (no LLM
 > running) and an embedder-classifier that routes by nearest-neighbor between

internal/auth/tier.go

@@ -1,7 +1,7 @@
 // Package auth is maven's authority layer — the 4-layer cascade and the
 // "surface caps authority" invariant.
 //
-// Spec contract (from DESIGN.md § Auth):
+// Spec contract (from docs/design.md § Auth):
 //
 //	a cascade, not a pick-one — each layer answers a different question:
 //

internal/config/config.go

@@ -224,6 +224,11 @@ type Config struct {
 	// See SearchConfig.
 	Search *SearchConfig `json:"search,omitempty"`
 
+	// Workstation — the big model on the owner's desktop, preferred over the
+	// resident one when its GPU is free. nil / absent / url empty ⇒ homesrv
+	// behaves exactly as it does today. See WorkstationConfig.
+	Workstation *WorkstationConfig `json:"workstation,omitempty"`
+
 	// Praxis — the ecosystem attention-state service. When configured, maven
 	// calls the Praxis HTTP tools API for attention listing and item lifecycle.
 	// Maven never touches Praxis's database directly (ecosystem invariant: no
@@ -649,7 +654,7 @@ type VoiceConfig struct {
 	// LLMRouter — route with the resident model instead of the embedding
 	// classifier. On by default since Vikunja #320.
 	//
-	// Measured on the held-out fixture (ROUTING-EVAL-31-07-2026.md): 63.2% of
+	// Measured on the held-out fixture (docs/evals/2026-07-31-routing.md): 63.2% of
 	// intents right against the classifier's 50.0%, and no route errors. It
 	// costs about 1s per turn instead of 30ms.
 	//
@@ -1105,6 +1110,44 @@ const (
 	DefaultKiwixSnippetRunes = 1500
 )
 
+// WorkstationConfig — the big model on the owner's desktop (workpc, a
+// 7900 GRE with 16GB), fronted by mavgpud.
+//
+// homesrv cannot grow a GPU, so the resident Qwen3-1.7B is the floor and this
+// is the preferred model above it (owner's call, 2026-08-02, docs/offload.md).
+// The workstation is never assumed up: its card is often held by a CPT run and
+// the machine sleeps. No block, or an empty URL, and homesrv behaves exactly as
+// it does today.
+//
+// Only the prompt crosses the LAN, and the workstation is not "the box". The
+// rules in CLAUDE.md about what may leave still apply.
+type WorkstationConfig struct {
+	// URL — where mavgpud listens, e.g. "http://192.168.1.105:8080". Empty ⇒
+	// the whole block is normalised to nil and nothing probes anything.
+	URL string `json:"url,omitempty"`
+
+	// Health — the admission endpoint. Empty ⇒ URL + "/health", which is what
+	// mavgpud serves. It answers 503 while the card is held, and that is the
+	// signal, so it must be the supervisor's endpoint and not llama-server's.
+	Health string `json:"health,omitempty"`
+
+	// Probe — how often admission is re-checked. 0 ⇒ DefaultWorkstationProbe.
+	// Nothing on the hot path waits for it: the answer is cached and read
+	// atomically, so this only sets how late Maven notices the card came back.
+	Probe Duration `json:"probe,omitempty"`
+
+	// Timeout — the per-request budget for a completion on the workstation.
+	// 0 ⇒ DefaultWorkstationTimeout. A big model on a LAN host is slower than
+	// the resident one, and a request that overruns falls back to the floor.
+	Timeout Duration `json:"timeout,omitempty"`
+}
+
+// Workstation defaults, applied in Normalise.
+const (
+	DefaultWorkstationProbe   = 15 * time.Second
+	DefaultWorkstationTimeout = 90 * time.Second
+)
+
 // SearchConfig — the self-hosted SearXNG instance she searches with.
 //
 // External search is allowed and off unless configured (CLAUDE.md). Configuring
@@ -1500,6 +1543,24 @@ func (c *Config) applyDefaults() {
 		}
 	}
 
+	// No address, no preferred model. An unconfigured workstation is the
+	// default deploy and must be indistinguishable from today.
+	if c.Workstation != nil && strings.TrimSpace(c.Workstation.URL) == "" {
+		c.Workstation = nil
+	}
+	if c.Workstation != nil {
+		w := c.Workstation
+		if strings.TrimSpace(w.Health) == "" {
+			w.Health = strings.TrimRight(w.URL, "/") + "/health"
+		}
+		if w.Probe <= 0 {
+			w.Probe = Duration(DefaultWorkstationProbe)
+		}
+		if w.Timeout <= 0 {
+			w.Timeout = Duration(DefaultWorkstationTimeout)
+		}
+	}
+
 	if c.Voice != nil {
 		if c.Voice.RouterThreshold <= 0 {
 			c.Voice.RouterThreshold = DefaultRouterThreshold

internal/config/config_test.go

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

internal/delivery/channel.go

@@ -1,6 +1,6 @@
 // Package delivery is maven's channel-routing + dispatch layer.
 //
-// Spec contract (from DESIGN.md § Delivery / channel routing):
+// Spec contract (from docs/design.md § Delivery / channel routing):
 //
 //   - routing = f(severity, presence). presence decides REACHABILITY; severity
 //     decides INSISTENCE. need both.

internal/delivery/routing_table_test.go

@@ -9,7 +9,7 @@ import (
 	"github.com/kami/maven/internal/store"
 )
 
-// This file walks every cell of the DESIGN.md § "Delivery / channel routing"
+// This file walks every cell of the docs/design.md § "Delivery / channel routing"
 // table, once as the pure table and once through the dispatcher, so a change
 // to either side has to break a named cell.
 //
@@ -205,7 +205,7 @@ func TestAwayChannelsGetMinimalBody(t *testing.T) {
 // An empty Summary no longer means "send the whole body" — it means a short
 // generic line — so the old expectation here was wrong as well as duplicated.
 
-// TestCareAwayDropIsRecorded — DESIGN.md's drop is a decision ("a missed water
+// TestCareAwayDropIsRecorded — docs/design.md's drop is a decision ("a missed water
 // nudge is noise, a missed backup failure isn't"), so it should be visible
 // rather than vanish. Today drop is a bare `continue`: no nudge row, no outbox
 // attempt, no log — nothing an operator can see afterwards. now it leaves a

internal/delivery/voicesink/voicesink.go

@@ -19,7 +19,7 @@
 //  3. if no live session exists, Send returns voice.ErrNoSession
 //     (wrapped). The daemon logs the partial dispatch; an OPEN deferred
 //     question is whether the dispatcher should reroute to away-channels
-//     instead of returning partial — listed in PROGRESS.md.
+//     instead of returning partial.
 //
 // Import direction: voicesink imports internal/tts (synth seam) and
 // internal/voice (Sessions registry). Both are siblings of delivery; the

internal/event/event.go

@@ -1,5 +1,4 @@
-// Package event is the unified intake envelope (Vikunja #283,
-// 20-07-2026-BACKLOG.md item 1).
+// Package event is the unified intake envelope (Vikunja #283).
 //
 // # The problem it solves
 //

internal/ipc/client.go

@@ -8,13 +8,15 @@ import (
 	"net"
 	"sync"
 	"time"
+
+	"github.com/kami/maven/internal/netaddr"
 )
 
-// Client — the module side of the boundary. Wraps a unix-socket connection
-// and satisfies CoreAPI, so a module imports ipc, holds a CoreAPI, and is
-// agnostic to whether it's been wired in-process (tests / daemon-embedded)
-// or over this socket (full topology). The swappability is the seam auth
-// will insert into without touching module code.
+// Client — the module side of the boundary. Wraps a connection to core and
+// satisfies CoreAPI, so a module imports ipc, holds a CoreAPI, and is
+// agnostic to whether it's been wired in-process (tests / daemon-embedded),
+// over a local unix socket, or over tcp to another host. The swappability is
+// the seam auth will insert into without touching module code.
 //
 // One Client ⇒ one conn ⇒ one concurrent request at a time. A module that
 // wants parallel requests opens one Client per goroutine; the store is the
@@ -22,7 +24,8 @@ import (
 // per-Client lock keeps frame interleaving impossible by construction.
 type Client struct {
 	conn net.Conn
-	path string // kept so a dropped conn can be re-dialed (core restart)
+	path string       // the address as configured, kept for errors and logs
+	addr netaddr.Addr // parsed, so a dropped conn can be re-dialed (core restart)
 	mu   sync.Mutex
 }
 
@@ -81,14 +84,22 @@ var readOnlyMethods = map[Method]bool{
 	MethodPing:                  true,
 }
 
-// Dial connects to a core socket at path and returns a Client. The module
-// owns its Client lifecycle; Close on shutdown.
+// Dial connects to core at path and returns a Client. The module owns its
+// Client lifecycle; Close on shutdown.
+//
+// path is a netaddr seam address: a bare path is the unix socket it has
+// always been, and "tcp://host:port?token=..." reaches a core on another
+// host. See internal/netaddr.
 func Dial(path string) (*Client, error) {
-	c, err := net.Dial("unix", path)
+	addr, err := netaddr.Parse(path)
 	if err != nil {
-		return nil, fmt.Errorf("ipc: dial %s: %w", path, err)
+		return nil, err
 	}
-	return &Client{conn: c, path: path}, nil
+	c, err := netaddr.Dial(addr)
+	if err != nil {
+		return nil, fmt.Errorf("ipc: dial %s: %w", addr, err)
+	}
+	return &Client{conn: c, path: path, addr: addr}, nil
 }
 
 func (c *Client) Close() error {
@@ -189,9 +200,9 @@ func (c *Client) call(ctx context.Context, m Method, params, result any) error {
 // re-dials clean. Caller holds c.mu.
 func (c *Client) roundtrip(m Method, raw json.RawMessage, resp *Response) error {
 	if c.conn == nil {
-		conn, err := net.Dial("unix", c.path)
+		conn, err := netaddr.Dial(c.addr)
 		if err != nil {
-			return fmt.Errorf("%w: dial %s: %v", errWriteLost, c.path, err)
+			return fmt.Errorf("%w: dial %s: %v", errWriteLost, c.addr, err)
 		}
 		c.conn = conn
 	}

internal/ipc/server.go

@@ -8,11 +8,11 @@ import (
 	"fmt"
 	"log"
 	"net"
-	"os"
 	"sync"
 	"sync/atomic"
 	"time"
 
+	"github.com/kami/maven/internal/netaddr"
 	"github.com/kami/maven/internal/store"
 	"golang.org/x/sys/unix"
 )
@@ -446,6 +446,7 @@ func mapErr(err error) error {
 type Server struct {
 	api  atomic.Value // stores CoreAPI
 	path string
+	addr netaddr.Addr
 
 	ln     net.Listener
 	wg     sync.WaitGroup
@@ -610,31 +611,29 @@ type CheckFunc func(ctx context.Context, m Method, params json.RawMessage) error
 // MethodAssertStepUp dispatch calls this instead of going through CoreAPI.
 type StepUpFunc func(ctx context.Context) error
 
-// Listen creates a Server bound to path. path's parent dir must exist and be
-// 0700 (we chmod it if we own it); the socket file itself is created 0600 so
-// only the same unix user can connect — the current "auth floor", same radius
-// as wg at the network boundary. Removing a stale socket at path first lets
-// the daemon restart cleanly.
+// Listen creates a Server bound to path.
+//
+// A bare path is a unix socket, unchanged: its parent dir is 0700 and the
+// socket file itself is 0600, so only the same unix user can connect — the
+// current "auth floor", same radius as wg at the network boundary. A stale
+// socket is removed first so the daemon restarts cleanly.
+//
+// A "tcp://host:port?token=..." address binds a network listener instead, for
+// a module that lives on another host. There is no filesystem there to be the
+// auth floor, so netaddr checks the shared token before this package sees the
+// connection and a token is mandatory. See internal/netaddr.
 func Listen(path string, api CoreAPI) (*Server, error) {
-	_ = os.Remove(path) // stale socket from a crashed daemon; ignore missing
-	if err := os.MkdirAll(parentDir(path), 0o700); err != nil {
-		return nil, fmt.Errorf("ipc: mkdir socket dir: %w", err)
-	}
-	// umask could widen the perms on socket creation; tighten then chmod to
-	// be explicit. 0600 ⇒ read+write by owner only.
-	oldMask := unix.Umask(0o077)
-	ln, err := net.Listen("unix", path)
-	unix.Umask(oldMask)
+	addr, err := netaddr.Parse(path)
 	if err != nil {
-		return nil, fmt.Errorf("ipc: listen %s: %w", path, err)
+		return nil, err
 	}
-	if err := os.Chmod(path, 0o600); err != nil {
-		_ = ln.Close()
-		_ = os.Remove(path)
-		return nil, fmt.Errorf("ipc: chmod socket: %w", err)
+	ln, err := netaddr.Listen(addr)
+	if err != nil {
+		return nil, err
 	}
 	s := &Server{
 		path: path,
+		addr: addr,
 		ln:   ln,
 		done: make(chan struct{}),
 	}
@@ -1268,7 +1267,7 @@ func (s *Server) Close() error {
 		// missing the seal costs every write since the last clean shutdown.
 		log.Printf("ipc: %d connection(s) still busy after %s, closing anyway", s.liveConns(), closeGrace)
 	}
-	_ = os.Remove(s.path)
+	netaddr.Cleanup(s.addr)
 	return err
 }
 
@@ -1338,25 +1337,6 @@ func (s *Server) Path() string { return s.path }
 // while the server is serving (dispatch loads api once per request via atomic).
 func (s *Server) SetAPI(api CoreAPI) { s.api.Store(api) }
 
-func parentDir(p string) string {
-	if i := lastIndexByte(p, '/'); i >= 0 {
-		if i == 0 {
-			return "/"
-		}
-		return p[:i]
-	}
-	return "."
-}
-
-func lastIndexByte(s string, b byte) int {
-	for i := len(s) - 1; i >= 0; i-- {
-		if s[i] == b {
-			return i
-		}
-	}
-	return -1
-}
-
 // peerCaller — read SO_PEERCRED off a unix conn to identify the connecting
 // process. Returns ok=false on a non-unix conn or a platform without
 // SO_PEERCRED; the caller then proceeds without a Caller (the socket perms

internal/kiwix/rewrite.go

@@ -31,7 +31,7 @@ type Completer interface {
 // or reply in Russian.
 //
 // Why the JSON wrapper: this model always thinks out loud and this llama-server
-// build ignores the thinking switch (see ROUTING-EVAL-31-07-2026.md). A bare
+// build ignores the thinking switch (see docs/evals/2026-07-31-routing.md). A bare
 // word-list grammar just captured the reasoning — every case came back as
 // "Let me analyze this request carefully". Demanding JSON, like routeGrammar and
 // responseGrammar already do, gives the reasoning nowhere to go.

internal/llm/client.go

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

internal/llm/remote.go

@@ -0,0 +1,193 @@
+package llm
+
+import (
+	"context"
+	"errors"
+	"log"
+	"net/http"
+	"sync/atomic"
+	"time"
+)
+
+// Pair — a preferred model on another host, with the resident one as the floor.
+//
+// homesrv cannot grow a GPU and the workstation has 16GB of VRAM, so the big
+// model runs there and the resident Qwen3-1.7B stays here. See docs/offload.md.
+// The workstation is never assumed up: its GPU is often busy with CPT runs and
+// the manga-recap pipeline, and the machine sleeps. So the remote is preferred,
+// never required, and Pair is what makes "preferred" mean something precise.
+//
+// This is admission control, not a scheduler. There is no arbiter deciding who
+// gets the card. A prober asks the remote whether it will take work, caches the
+// answer, and every request reads that cached answer in nanoseconds. Routing
+// sits on the hot path at p50 825ms and must never wait on a machine that may
+// be asleep, so no request ever pays for a health check itself.
+//
+// Pair satisfies nothing by itself. Callers pick a method by which half of the
+// degradation rule they live under:
+//
+//   - Complete falls back silently. For routing, replies, and nudge phrasing,
+//     where the big model is only better and the 1.7B is today's shipping
+//     quality. He is not told which model phrased his reply.
+//   - CompleteRemote returns ErrRemoteUnavailable instead of falling back. For
+//     a world question, or a long Kiwix or search passage, where a 1.7B
+//     confabulates rather than summarises. A named gap beats an invented
+//     answer.
+type Pair struct {
+	remote *Client
+	floor  *Client
+
+	// up — the cached admission answer, written only by the prober goroutine
+	// and read by every request. Atomic so the read costs nanoseconds and no
+	// request ever contends with the prober.
+	up atomic.Bool
+
+	health   string
+	interval time.Duration
+	http     *http.Client
+	stop     chan struct{}
+}
+
+// ErrRemoteUnavailable — the workstation model was required and is not
+// answering. Callers on the naming half of the degradation rule turn this into
+// a gap in the reply ("не могу сейчас"), never into a guess from the floor.
+var ErrRemoteUnavailable = errors.New("llm: workstation model unavailable")
+
+// ErrNoFloor — a Pair was built with no resident model to fall back to. A
+// configuration mistake: the floor is the whole point.
+var ErrNoFloor = errors.New("llm: no floor client")
+
+// NewPair builds the two-model arrangement. remote may be nil, which is the
+// unconfigured deploy and must behave exactly as the box behaves today: every
+// call goes to the floor and nothing probes anything.
+//
+// health is the URL the prober asks. llama-server's /health answers "is a model
+// loaded and ready", which is the useful signal here, because llama-server
+// refuses to load at all when VRAM is short. That makes a busy card detectable
+// without any cooperation from the owner's other jobs.
+func NewPair(remote, floor *Client, health string, interval time.Duration) *Pair {
+	p := &Pair{
+		remote:   remote,
+		floor:    floor,
+		health:   health,
+		interval: interval,
+		http:     &http.Client{Timeout: probeTimeout},
+		stop:     make(chan struct{}),
+	}
+	return p
+}
+
+// probeTimeout — a remote that cannot answer /health this fast is not going to
+// serve a turn either. Short on purpose: the prober runs on its own goroutine,
+// but a slow probe still delays the moment Maven notices the card came back.
+const probeTimeout = 2 * time.Second
+
+// Start begins probing. It returns immediately, and the first probe runs before
+// the first tick so a remote that is already up is used on the first turn
+// rather than after one interval of falling back. Safe to call with a nil
+// remote; it does nothing.
+func (p *Pair) Start(ctx context.Context) {
+	if p.remote == nil || p.health == "" {
+		return
+	}
+	go func() {
+		p.probe(ctx)
+		t := time.NewTicker(p.interval)
+		defer t.Stop()
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case <-p.stop:
+				return
+			case <-t.C:
+				p.probe(ctx)
+			}
+		}
+	}()
+}
+
+// Stop ends the prober. Idempotent.
+func (p *Pair) Stop() {
+	select {
+	case <-p.stop:
+	default:
+		close(p.stop)
+	}
+}
+
+// Available reports whether the workstation will take work right now. It reads
+// a cached flag, so it is safe to call per turn on the hot path. A false answer
+// is never stale in the direction that matters: the worst case is that Maven
+// falls back for up to one probe interval after the card frees up.
+func (p *Pair) Available() bool {
+	return p.remote != nil && p.up.Load()
+}
+
+func (p *Pair) probe(ctx context.Context) {
+	ctx, cancel := context.WithTimeout(ctx, probeTimeout)
+	defer cancel()
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, p.health, nil)
+	if err != nil {
+		p.set(false)
+		return
+	}
+	resp, err := p.http.Do(req)
+	if err != nil {
+		p.set(false)
+		return
+	}
+	defer resp.Body.Close()
+	p.set(resp.StatusCode == http.StatusOK)
+}
+
+// set records the admission answer and logs only the transitions. A machine
+// that sleeps every night would otherwise write one line per interval forever.
+func (p *Pair) set(up bool) {
+	if p.up.Swap(up) == up {
+		return
+	}
+	if up {
+		log.Printf("llm: workstation model available at %s", p.health)
+	} else {
+		log.Printf("llm: workstation model unavailable, falling back to the resident model")
+	}
+}
+
+// Complete runs r on the workstation when it will take work, and on the
+// resident model otherwise. A remote that fails mid-request falls back too: the
+// admission answer is a cache and can be one interval out of date, so an error
+// here is expected rather than exceptional.
+//
+// This is the silent half of the degradation rule. It must be indistinguishable
+// from today's behaviour when the workstation is down.
+func (p *Pair) Complete(ctx context.Context, r Req) (string, error) {
+	if p.floor == nil {
+		return "", ErrNoFloor
+	}
+	if p.Available() {
+		out, err := p.remote.Complete(ctx, r)
+		if err == nil {
+			return out, nil
+		}
+		// The cached answer was wrong. Correct it now rather than sending the
+		// next request into the same hole, then fall back.
+		p.set(false)
+	}
+	return p.floor.Complete(ctx, r)
+}
+
+// CompleteRemote runs r on the workstation or refuses. It never falls back,
+// because for a world question the resident 1.7B does not answer worse, it
+// invents. Callers turn ErrRemoteUnavailable into a named gap.
+func (p *Pair) CompleteRemote(ctx context.Context, r Req) (string, error) {
+	if !p.Available() {
+		return "", ErrRemoteUnavailable
+	}
+	out, err := p.remote.Complete(ctx, r)
+	if err != nil {
+		p.set(false)
+		return "", errors.Join(ErrRemoteUnavailable, err)
+	}
+	return out, nil
+}

internal/llm/remote_test.go

@@ -0,0 +1,210 @@
+package llm
+
+import (
+	"context"
+	"errors"
+	"net/http"
+	"net/http/httptest"
+	"sync/atomic"
+	"testing"
+	"time"
+)
+
+// completionServer stands in for a llama-server. It counts what reached it, so
+// a test can say which of the two models answered.
+func completionServer(t *testing.T, reply string, hits *atomic.Int64) *httptest.Server {
+	t.Helper()
+	s := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		hits.Add(1)
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"choices":[{"message":{"content":"` + reply + `"}}]}`))
+	}))
+	t.Cleanup(s.Close)
+	return s
+}
+
+func healthServer(t *testing.T, ok *atomic.Bool) *httptest.Server {
+	t.Helper()
+	s := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if !ok.Load() {
+			w.WriteHeader(http.StatusServiceUnavailable)
+			return
+		}
+		w.WriteHeader(http.StatusOK)
+	}))
+	t.Cleanup(s.Close)
+	return s
+}
+
+// waitFor polls until cond holds or the deadline passes. The prober runs on its
+// own goroutine, so a test has to wait for it rather than assume it has run.
+func waitFor(t *testing.T, cond func() bool) bool {
+	t.Helper()
+	deadline := time.Now().Add(2 * time.Second)
+	for time.Now().Before(deadline) {
+		if cond() {
+			return true
+		}
+		time.Sleep(5 * time.Millisecond)
+	}
+	return false
+}
+
+// The unconfigured deploy. No remote, no probing, every call to the floor —
+// exactly what the box does today.
+func TestNoRemoteGoesToTheFloor(t *testing.T) {
+	var floorHits atomic.Int64
+	floor := completionServer(t, "floor", &floorHits)
+
+	p := NewPair(nil, New(floor.URL, time.Second), "", time.Second)
+	p.Start(context.Background())
+	defer p.Stop()
+
+	if p.Available() {
+		t.Fatal("a Pair with no remote reports available")
+	}
+	out, err := p.Complete(context.Background(), Req{User: "привет"})
+	if err != nil {
+		t.Fatalf("complete: %v", err)
+	}
+	if out != "floor" || floorHits.Load() != 1 {
+		t.Fatalf("out = %q, floor hits = %d", out, floorHits.Load())
+	}
+}
+
+// The workstation is up, so it answers and the resident model is not touched.
+func TestAvailableRemoteAnswers(t *testing.T) {
+	var remoteHits, floorHits atomic.Int64
+	remote := completionServer(t, "remote", &remoteHits)
+	floor := completionServer(t, "floor", &floorHits)
+	up := &atomic.Bool{}
+	up.Store(true)
+	health := healthServer(t, up)
+
+	p := NewPair(New(remote.URL, time.Second), New(floor.URL, time.Second), health.URL, 20*time.Millisecond)
+	p.Start(context.Background())
+	defer p.Stop()
+	if !waitFor(t, p.Available) {
+		t.Fatal("prober never saw the remote come up")
+	}
+
+	out, err := p.Complete(context.Background(), Req{User: "привет"})
+	if err != nil {
+		t.Fatalf("complete: %v", err)
+	}
+	if out != "remote" || floorHits.Load() != 0 {
+		t.Fatalf("out = %q, floor hits = %d", out, floorHits.Load())
+	}
+}
+
+// The card is busy, so /health refuses and Complete degrades silently. This is
+// the constraint from 483: the workstation being down is indistinguishable from
+// today's behaviour.
+func TestBusyCardFallsBackSilently(t *testing.T) {
+	var remoteHits, floorHits atomic.Int64
+	remote := completionServer(t, "remote", &remoteHits)
+	floor := completionServer(t, "floor", &floorHits)
+	health := healthServer(t, &atomic.Bool{}) // never ok
+
+	p := NewPair(New(remote.URL, time.Second), New(floor.URL, time.Second), health.URL, 20*time.Millisecond)
+	p.Start(context.Background())
+	defer p.Stop()
+	time.Sleep(60 * time.Millisecond)
+
+	out, err := p.Complete(context.Background(), Req{User: "привет"})
+	if err != nil {
+		t.Fatalf("complete: %v", err)
+	}
+	if out != "floor" || remoteHits.Load() != 0 {
+		t.Fatalf("out = %q, remote hits = %d", out, remoteHits.Load())
+	}
+}
+
+// The cached admission answer can be one interval out of date, so a remote that
+// dies between probes must still not break the turn.
+func TestRemoteErrorMidRequestFallsBack(t *testing.T) {
+	var floorHits atomic.Int64
+	dead := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.WriteHeader(http.StatusInternalServerError)
+	}))
+	defer dead.Close()
+	floor := completionServer(t, "floor", &floorHits)
+	up := &atomic.Bool{}
+	up.Store(true)
+	health := healthServer(t, up)
+
+	p := NewPair(New(dead.URL, time.Second), New(floor.URL, time.Second), health.URL, time.Hour)
+	p.Start(context.Background())
+	defer p.Stop()
+	if !waitFor(t, p.Available) {
+		t.Fatal("prober never saw the remote come up")
+	}
+
+	out, err := p.Complete(context.Background(), Req{User: "привет"})
+	if err != nil {
+		t.Fatalf("complete: %v", err)
+	}
+	if out != "floor" || floorHits.Load() != 1 {
+		t.Fatalf("out = %q, floor hits = %d", out, floorHits.Load())
+	}
+	// The failed request must have corrected the cached answer, so the next
+	// one does not walk into the same hole.
+	if p.Available() {
+		t.Fatal("a failed remote request left the admission answer up")
+	}
+}
+
+// The naming half of the degradation rule. A world question must not be handed
+// to the resident model, because it answers by inventing.
+func TestCompleteRemoteNamesTheGap(t *testing.T) {
+	var floorHits atomic.Int64
+	floor := completionServer(t, "floor", &floorHits)
+	health := healthServer(t, &atomic.Bool{}) // never ok
+
+	p := NewPair(New("http://127.0.0.1:1", time.Second), New(floor.URL, time.Second), health.URL, 20*time.Millisecond)
+	p.Start(context.Background())
+	defer p.Stop()
+	time.Sleep(60 * time.Millisecond)
+
+	if _, err := p.CompleteRemote(context.Background(), Req{User: "почему небо голубое"}); !errors.Is(err, ErrRemoteUnavailable) {
+		t.Fatalf("err = %v, want ErrRemoteUnavailable", err)
+	}
+	if floorHits.Load() != 0 {
+		t.Fatalf("CompleteRemote fell back to the floor %d times", floorHits.Load())
+	}
+}
+
+// Routing sits on the hot path and must never pay for a health check. Available
+// reads a cached flag, so it costs no network at all.
+func TestAvailableDoesNotProbe(t *testing.T) {
+	var probes atomic.Int64
+	health := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		probes.Add(1)
+		w.WriteHeader(http.StatusOK)
+	}))
+	defer health.Close()
+
+	p := NewPair(New("http://127.0.0.1:1", time.Second), New("http://127.0.0.1:1", time.Second), health.URL, time.Hour)
+	p.Start(context.Background())
+	defer p.Stop()
+	if !waitFor(t, p.Available) {
+		t.Fatal("prober never ran")
+	}
+
+	before := probes.Load()
+	for range 1000 {
+		p.Available()
+	}
+	if got := probes.Load(); got != before {
+		t.Fatalf("1000 Available calls made %d probes", got-before)
+	}
+}
+
+// A Pair with no floor is a configuration mistake, and it must say so rather
+// than silently having nowhere to degrade to.
+func TestNoFloorIsAnError(t *testing.T) {
+	p := NewPair(nil, nil, "", time.Second)
+	if _, err := p.Complete(context.Background(), Req{User: "привет"}); !errors.Is(err, ErrNoFloor) {
+		t.Fatalf("err = %v, want ErrNoFloor", err)
+	}
+}

internal/loop/gate_test.go

@@ -10,12 +10,12 @@ import (
 
 // Tests for the universal restraint gate.
 //
-// DESIGN.md § Trigger model: "the gate is universal, applied by the loop, never
+// docs/design.md § Trigger model: "the gate is universal, applied by the loop, never
 // per-rule — quiet-hours, presence, cooldown, snooze, calendar-busy all live in
 // one fires()." These tests pin the CONSERVATIVE side of that: the cases where
 // Maven must stay quiet. They exist so nobody loosens the gate by accident.
 //
-// Where the code does not yet do what DESIGN.md promises, the test is written to
+// Where the code does not yet do what docs/design.md promises, the test is written to
 // show the gap and then skipped, with the file and line to fix. Behaviour is not
 // changed to make a test pass.
 
@@ -54,7 +54,7 @@ func TestGateQuietHoursSuppressesCareOnly(t *testing.T) {
 
 // ---------------------------- presence ---------------------------------------
 
-// DESIGN.md § Delivery: "sev <= 2 drops on away, sev >= 3 holds: a missed water
+// docs/design.md § Delivery: "sev <= 2 drops on away, sev >= 3 holds: a missed water
 // nudge is noise, a missed backup failure isn't."
 func TestGateAwayDropsCareHoldsOps(t *testing.T) {
 	cases := []struct {
@@ -250,7 +250,7 @@ func TestTickOrderOfRulesDoesNotMatter(t *testing.T) {
 
 // ---------------------------- reminders bypass the gate ----------------------
 
-// DESIGN.md § User reminders: "bypasses the restraint gate — 'wake me 7' fires
+// docs/design.md § User reminders: "bypasses the restraint gate — 'wake me 7' fires
 // in quiet hours; that's the point." Every suppressor set at once, and the
 // reminder still comes through.
 func TestRemindersBypassEverySuppressor(t *testing.T) {
@@ -269,7 +269,7 @@ func TestRemindersBypassEverySuppressor(t *testing.T) {
 	}
 }
 
-// GAP — DESIGN.md § User reminders ends "Snooze still applies." RemindDecisions
+// GAP — docs/design.md § User reminders ends "Snooze still applies." RemindDecisions
 // passes every due reminder straight through with no snooze check, so a snoozed
 // reminder fires anyway. The test below is what the contract asks for.
 func TestRemindersStillHonourSnooze(t *testing.T) {

internal/loop/rules_test.go

@@ -18,7 +18,7 @@ import (
 //   - does it stay quiet when it should?
 //   - is it silent when the key it needs has no data at all?
 //
-// The last one is load-bearing. DESIGN.md: "since(key)==null → don't fire.
+// The last one is load-bearing. docs/design.md: "since(key)==null → don't fire.
 // Silence on no-data is 'shuts up when uncertain'."
 
 // stateWith builds a snapshot at refTime() holding just the given facts.
@@ -179,7 +179,7 @@ func TestCareRulePredicates(t *testing.T) {
 
 // ---------------------------- ops rules --------------------------------------
 
-// The two ops rules match on a value AND on which poller wrote it. DESIGN.md:
+// The two ops rules match on a value AND on which poller wrote it. docs/design.md:
 // "a compromised poller must not be able to forge a trigger." Half of this
 // table is forgery attempts; all of them must be refused.
 func TestOpsRulePredicates(t *testing.T) {
@@ -331,7 +331,7 @@ func TestNoDefaultRuleFiresOnEmptyState(t *testing.T) {
 	}
 }
 
-// Severities are the delivery contract (DESIGN.md § Delivery / channel
+// Severities are the delivery contract (docs/design.md § Delivery / channel
 // routing): care is sev1-2 and drops when away, ops is sev3-4 and holds. Pin
 // them so a change to a rule's insistence has to be deliberate.
 func TestDefaultRuleSeverities(t *testing.T) {
@@ -356,7 +356,7 @@ func TestDefaultRuleSeverities(t *testing.T) {
 	}
 }
 
-// Cooldown bounds keep the feedback tuner honest — DESIGN.md wants
+// Cooldown bounds keep the feedback tuner honest — docs/design.md wants
 // `cooldown in [min,max]` "so a weird week can't mutate Maven silent or
 // stalker". A base outside its own envelope would make that meaningless.
 func TestDefaultRuleCooldownsAreBounded(t *testing.T) {

internal/memory/lexical.go

@@ -0,0 +1,126 @@
+package memory
+
+import (
+	"strings"
+	"unicode"
+)
+
+// stopwords — words that carry no topic. A question and a note that share only
+// these share nothing: "почему небо синее" and "сеть какая-то медленная" both
+// contain "какая"-shaped filler and are about different worlds.
+var stopwords = map[string]bool{
+	// interrogatives and demonstratives
+	"что": true, "чего": true, "какой": true, "какая": true, "какое": true,
+	"какие": true, "каких": true, "кто": true, "кого": true, "кому": true,
+	"почему": true, "зачем": true, "где": true, "куда": true, "откуда": true,
+	"когда": true, "сколько": true, "как": true, "то": true, "это": true,
+	"этот": true, "тот": true, "там": true, "тут": true, "такой": true,
+	// pronouns — every sentence he says is about him, so "я" is not a topic
+	"я": true, "меня": true, "мне": true, "мой": true, "моя": true, "мои": true,
+	"ты": true, "тебя": true, "тебе": true, "твой": true, "он": true, "она": true,
+	"они": true, "мы": true, "себя": true, "свой": true,
+	// prepositions, conjunctions, particles, copulas
+	"в": true, "во": true, "на": true, "с": true, "со": true, "у": true,
+	"о": true, "об": true, "про": true, "за": true, "из": true, "по": true,
+	"до": true, "от": true, "для": true, "над": true, "под": true, "при": true,
+	"и": true, "а": true, "но": true, "или": true, "же": true, "ли": true,
+	"не": true, "ни": true, "бы": true, "был": true, "была": true, "было": true,
+	"быть": true, "есть": true, "был-ли": true, "уже": true, "ещё": true,
+	"еще": true, "так": true, "вот": true, "там-же": true,
+	// English filler, for the mixed utterances he does say
+	"the": true, "a": true, "an": true, "is": true, "are": true, "was": true,
+	"were": true, "be": true, "of": true, "in": true, "on": true, "at": true,
+	"to": true, "for": true, "about": true, "and": true, "or": true, "not": true,
+	"what": true, "who": true, "why": true, "when": true, "where": true,
+	"which": true, "how": true, "i": true, "my": true, "me": true, "it": true,
+	"this": true, "that": true,
+}
+
+// firstPerson — the words that make an utterance a question about his own
+// life. Not possession only: "как я восстановил конфиги" owns nothing and is
+// still about him.
+var firstPerson = map[string]bool{
+	"я": true, "меня": true, "мне": true, "мной": true, "мой": true,
+	"моя": true, "моё": true, "мое": true, "мои": true, "моего": true,
+	"моей": true, "моих": true, "моим": true, "себя": true, "свой": true,
+	"своя": true, "свои": true, "своего": true, "мною": true,
+	"i": true, "me": true, "my": true, "mine": true, "myself": true,
+}
+
+// RecallAllowed is the second half of the recall gate (#470). A hit that
+// cleared the score and margin gate may still be about something else
+// entirely: the held-out fixture puts the right note at 0.791-0.890 and the
+// must-be-silent cases at 0.795-0.835, so no threshold sits between them, and
+// a note about his slow network answered "почему небо синее?".
+//
+// The veto applies only to a question that mentions nothing of his. That
+// restriction is what keeps the fix from costing more than it saves: recall
+// exists to find the note whose words he no longer remembers, and demanding a
+// shared word of every recall silenced four true recalls on the fixture to
+// kill one false one. A question about his own life keeps the embedder alone
+// as its judge. A question about the world has to name something the memory
+// actually mentions.
+func RecallAllowed(query, text string) bool {
+	if mentionsHim(query) {
+		return true
+	}
+	return SharesContentWord(query, text)
+}
+
+func mentionsHim(query string) bool {
+	for _, w := range strings.FieldsFunc(strings.ToLower(query), func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	}) {
+		if firstPerson[w] {
+			return true
+		}
+	}
+	return false
+}
+
+// SharesContentWord reports whether query and text have at least one topic
+// word in common, after dropping the words that carry no topic. Stems are
+// compared, so the note and the question do not have to inflect alike.
+func SharesContentWord(query, text string) bool {
+	q := contentWords(query)
+	if len(q) == 0 {
+		// Nothing to compare — a question made entirely of filler. The score
+		// gate is then the only judge it can have.
+		return true
+	}
+	t := contentWords(text)
+	for _, a := range q {
+		for _, b := range t {
+			if a == b || sameStem(a, b) {
+				return true
+			}
+		}
+	}
+	return false
+}
+
+func contentWords(s string) []string {
+	var out []string
+	for _, w := range strings.FieldsFunc(strings.ToLower(s), func(r rune) bool {
+		return !unicode.IsLetter(r) && !unicode.IsDigit(r)
+	}) {
+		if !stopwords[w] {
+			out = append(out, w)
+		}
+	}
+	return out
+}
+
+// sameStem is inflection and derivation tolerance: Russian marks case and
+// tense on the ending, and the note and the question rarely use the same form.
+// "воду" and "вода" are the same water, "кормить" and "корм" the same feeding.
+// All but the last rune of the shorter word must match, and never fewer than
+// three, which is what keeps "сеть" clear of "сеанс".
+func sameStem(a, b string) bool {
+	ar, br := []rune(a), []rune(b)
+	n := min(len(ar), len(br)) - 1
+	if n < 3 {
+		return false
+	}
+	return string(ar[:n]) == string(br[:n])
+}

internal/memory/lexical_test.go

@@ -0,0 +1,40 @@
+package memory
+
+import "testing"
+
+func TestRecallAllowed(t *testing.T) {
+	cases := []struct {
+		name        string
+		query, text string
+		want        bool
+	}{
+		// The #470 shape: a world question and a note about his box.
+		{"world question, unrelated note", "почему небо синее", "сеть какая-то медленная", false},
+		{"world question, unrelated fact", "какая столица Франции", "какая последняя версия языка Go", false},
+		{"silent fixture case", "во сколько отходит поезд", "бэкап запускается в три ночи", false},
+
+		// A world question that does name the topic keeps its answer.
+		{"world question, same topic", "какой поезд идёт в Минск", "поезда в Минск ходят утром", true},
+
+		// A question about his own life is judged by the embedder alone,
+		// because recall exists for words he no longer remembers.
+		{"about him, no shared word", "во сколько я обычно засыпаю", "ложусь около одиннадцати", true},
+		{"about him, english", "which colour scheme do i like", "тёмная тема везде", true},
+
+		// Inflection must not break a match.
+		{"inflected", "чем кормить кота", "корм для кота в шкафу", true},
+	}
+	for _, c := range cases {
+		if got := RecallAllowed(c.query, c.text); got != c.want {
+			t.Errorf("%s: RecallAllowed(%q, %q) = %v, want %v", c.name, c.query, c.text, got, c.want)
+		}
+	}
+}
+
+// A question made only of filler has no topic word to match on, and the score
+// gate is then the only judge it can have.
+func TestRecallAllowedFallsBackWhenNothingToCompare(t *testing.T) {
+	if !RecallAllowed("что это", "сеть какая-то медленная") {
+		t.Error("a question with no content word must not be vetoed")
+	}
+}

internal/memory/recalleval/recalleval.go

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

internal/memory/recalleval/recalleval_test.go

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

internal/memory/recalleval/ru_recall_v1.json

@@ -41,7 +41,7 @@
       "tags": ["preference", "homelab", "paraphrase", "hard"],
       "query": "когда запускать резервное копирование",
       "want": "n1",
-      "note": "The DESIGN.md preference-seam example, phrased as the operator would ask it later.",
+      "note": "The docs/design.md preference-seam example, phrased as the operator would ask it later.",
       "notes": [
         {"id": "n1", "text": "бэкапы лучше делать ночью в три часа", "kind": "note"},
         {"id": "n2", "text": "обновления ставлю по субботам", "kind": "note"},

internal/morning/morning.go

@@ -1,5 +1,5 @@
 // Package morning is maven's morning routine engine — item #3 off the
-// 2026-07-20 backlog (see Maven/20-07-2026-BACKLOG.md).
+// 2026-07-20 backlog (Vikunja #280).
 //
 // A Routine is NOT four independent reminder timers. It's a checklist for a
 // daily window: several Items, each evidenced by a fact key, completed in

internal/netaddr/netaddr.go

@@ -0,0 +1,277 @@
+// Package netaddr parses a daemon seam address and dials or binds it.
+//
+// Every seam between Maven's daemons used to be a unix socket with the
+// network hardcoded at the call site — two dials in internal/ipc, one listen,
+// and the same pair again in internal/worker. That is correct for co-located
+// daemons and it is the reason a module cannot live on another host. This
+// package moves the choice into the address string so a deploy picks the
+// transport, not a recompile:
+//
+//	/run/maven/stt.sock                   unix   (the default, unchanged)
+//	unix:///run/maven/stt.sock            unix   (explicit, same thing)
+//	tcp://workstation:9310?token=hunter2  tcp
+//
+// A scheme-less address is unix and behaves exactly as it did before this
+// package existed: same 0700 parent dir, same 0600 socket, same bytes on the
+// wire with no handshake in front of them.
+//
+// Over TCP the filesystem permission that authenticated the unix socket is
+// gone, and what crosses this seam is audio of the owner speaking and the
+// text of his turns. So a TCP seam carries a shared token, checked before the
+// first protocol frame is read. Wireguard is supported underneath and is not
+// required.
+package netaddr
+
+import (
+	"crypto/subtle"
+	"errors"
+	"fmt"
+	"net"
+	"net/url"
+	"os"
+	"path/filepath"
+	"strings"
+	"time"
+
+	"golang.org/x/sys/unix"
+)
+
+// ErrUnauthorized — the peer presented a token the listener does not accept,
+// or presented none when one is required.
+var ErrUnauthorized = errors.New("netaddr: unauthorized")
+
+// Addr is a parsed seam endpoint.
+type Addr struct {
+	// Network is "unix" or "tcp".
+	Network string
+	// Address is the socket path (unix) or host:port (tcp).
+	Address string
+	// Token is the shared secret for a tcp seam. Empty for unix, where the
+	// filesystem does the same job.
+	Token string
+}
+
+// String renders the address for logs and errors. The token is never included.
+func (a Addr) String() string {
+	if a.Network == "unix" {
+		return a.Address
+	}
+	return a.Network + "://" + a.Address
+}
+
+// IsUnix reports whether this seam is a unix socket, and so is local, is
+// authenticated by file permissions, and needs no handshake.
+func (a Addr) IsUnix() bool { return a.Network == "unix" }
+
+// Parse reads a seam address. Anything without a "scheme://" prefix is a unix
+// socket path, which keeps every existing config and every default working
+// untouched.
+func Parse(s string) (Addr, error) {
+	if !strings.Contains(s, "://") {
+		return Addr{Network: "unix", Address: s}, nil
+	}
+	u, err := url.Parse(s)
+	if err != nil {
+		return Addr{}, fmt.Errorf("netaddr: parse %q: %w", s, err)
+	}
+	switch u.Scheme {
+	case "unix":
+		return Addr{Network: "unix", Address: u.Path}, nil
+	case "tcp":
+		if u.Host == "" {
+			return Addr{}, fmt.Errorf("netaddr: %q has no host:port", s)
+		}
+		return Addr{Network: "tcp", Address: u.Host, Token: u.Query().Get("token")}, nil
+	default:
+		return Addr{}, fmt.Errorf("netaddr: unsupported scheme %q", u.Scheme)
+	}
+}
+
+// MustParse is Parse for a literal known good at compile time. It panics on a
+// bad address, so use it in tests and constants, never on config input.
+func MustParse(s string) Addr {
+	a, err := Parse(s)
+	if err != nil {
+		panic(err)
+	}
+	return a
+}
+
+// handshakeTimeout bounds the token exchange. A peer that cannot write one
+// short line in this long is not going to serve a turn either.
+const handshakeTimeout = 5 * time.Second
+
+// greeting prefixes the token line. Versioned so a later mTLS seam can be
+// told apart from this one on the wire.
+const greeting = "MAVEN1 "
+
+// Dial connects to a. On a tcp seam it sends the token and waits for the
+// listener to accept it, so a returned conn is already authorized and the
+// caller can write its first protocol frame.
+func Dial(a Addr) (net.Conn, error) {
+	return DialTimeout(a, 0)
+}
+
+// DialTimeout is Dial with a bound on the connect. Zero means the operating
+// system default. The token exchange gets its own timeout either way.
+func DialTimeout(a Addr, timeout time.Duration) (net.Conn, error) {
+	var c net.Conn
+	var err error
+	if timeout > 0 {
+		c, err = net.DialTimeout(a.Network, a.Address, timeout)
+	} else {
+		c, err = net.Dial(a.Network, a.Address)
+	}
+	if err != nil {
+		return nil, err
+	}
+	if a.IsUnix() {
+		return c, nil
+	}
+	if err := clientHandshake(c, a.Token); err != nil {
+		_ = c.Close()
+		return nil, err
+	}
+	return c, nil
+}
+
+func clientHandshake(c net.Conn, token string) error {
+	_ = c.SetDeadline(time.Now().Add(handshakeTimeout))
+	defer c.SetDeadline(time.Time{})
+	if _, err := c.Write([]byte(greeting + token + "\n")); err != nil {
+		return fmt.Errorf("netaddr: send token: %w", err)
+	}
+	var reply [3]byte
+	if _, err := readFull(c, reply[:]); err != nil {
+		return fmt.Errorf("%w: %v", ErrUnauthorized, err)
+	}
+	if string(reply[:]) != "ok\n" {
+		return ErrUnauthorized
+	}
+	return nil
+}
+
+// Listener wraps a net.Listener so Accept performs the token check for a tcp
+// seam. A connection that fails the check is closed and never surfaces, so
+// the protocol above this layer only ever sees authorized peers.
+type Listener struct {
+	net.Listener
+	addr Addr
+}
+
+// Accept returns the next authorized connection. Unauthorized peers are
+// dropped and Accept keeps waiting: a bad token is a rejected stranger, not a
+// reason to stop serving.
+func (l *Listener) Accept() (net.Conn, error) {
+	for {
+		c, err := l.Listener.Accept()
+		if err != nil {
+			return nil, err
+		}
+		if l.addr.IsUnix() {
+			return c, nil
+		}
+		if err := serverHandshake(c, l.addr.Token); err != nil {
+			_ = c.Close()
+			continue
+		}
+		return c, nil
+	}
+}
+
+// Addr reports the parsed seam address this listener was built from.
+func (l *Listener) SeamAddr() Addr { return l.addr }
+
+func serverHandshake(c net.Conn, want string) error {
+	_ = c.SetDeadline(time.Now().Add(handshakeTimeout))
+	defer c.SetDeadline(time.Time{})
+	// The line is bounded: greeting, token, newline. Read a byte at a time so
+	// nothing of the first protocol frame is consumed when the token is short.
+	line := make([]byte, 0, 128)
+	var b [1]byte
+	for {
+		if _, err := readFull(c, b[:]); err != nil {
+			return err
+		}
+		if b[0] == '\n' {
+			break
+		}
+		line = append(line, b[0])
+		if len(line) > 512 {
+			return ErrUnauthorized
+		}
+	}
+	got, ok := strings.CutPrefix(string(line), greeting)
+	if !ok {
+		return ErrUnauthorized
+	}
+	if subtle.ConstantTimeCompare([]byte(got), []byte(want)) != 1 {
+		return ErrUnauthorized
+	}
+	if _, err := c.Write([]byte("ok\n")); err != nil {
+		return err
+	}
+	return nil
+}
+
+func readFull(c net.Conn, p []byte) (int, error) {
+	n := 0
+	for n < len(p) {
+		m, err := c.Read(p[n:])
+		n += m
+		if err != nil {
+			return n, err
+		}
+	}
+	return n, nil
+}
+
+// Listen binds a. A unix seam gets the perms it has always had: parent dir
+// 0700, socket 0600, and any stale socket from a crashed daemon removed
+// first. A tcp seam must carry a token, because there is no filesystem to
+// stand in for one.
+func Listen(a Addr) (*Listener, error) {
+	if a.IsUnix() {
+		ln, err := listenUnix(a.Address)
+		if err != nil {
+			return nil, err
+		}
+		return &Listener{Listener: ln, addr: a}, nil
+	}
+	if a.Token == "" {
+		return nil, fmt.Errorf("netaddr: listen %s: tcp seam requires a token", a)
+	}
+	ln, err := net.Listen("tcp", a.Address)
+	if err != nil {
+		return nil, fmt.Errorf("netaddr: listen %s: %w", a, err)
+	}
+	return &Listener{Listener: ln, addr: a}, nil
+}
+
+func listenUnix(path string) (net.Listener, error) {
+	_ = os.Remove(path) // stale socket from a crashed daemon; ignore missing
+	if err := os.MkdirAll(filepath.Dir(path), 0o700); err != nil {
+		return nil, fmt.Errorf("netaddr: mkdir socket dir: %w", err)
+	}
+	// umask could widen the perms on socket creation; tighten then chmod to
+	// be explicit. 0600 ⇒ read+write by owner only.
+	oldMask := unix.Umask(0o077)
+	ln, err := net.Listen("unix", path)
+	unix.Umask(oldMask)
+	if err != nil {
+		return nil, fmt.Errorf("netaddr: listen %s: %w", path, err)
+	}
+	if err := os.Chmod(path, 0o600); err != nil {
+		_ = ln.Close()
+		_ = os.Remove(path)
+		return nil, fmt.Errorf("netaddr: chmod socket: %w", err)
+	}
+	return ln, nil
+}
+
+// Cleanup removes the socket file behind a unix seam. It is a no-op for tcp.
+func Cleanup(a Addr) {
+	if a.IsUnix() && a.Address != "" {
+		_ = os.Remove(a.Address)
+	}
+}

internal/netaddr/netaddr_test.go

@@ -0,0 +1,185 @@
+package netaddr
+
+import (
+	"errors"
+	"net"
+	"path/filepath"
+	"testing"
+)
+
+// A scheme-less address must stay unix. Every deploy in the tree writes a bare
+// path, so this is the test that says the transport change costs them nothing.
+func TestParseSchemelessIsUnix(t *testing.T) {
+	a, err := Parse("/run/maven/stt.sock")
+	if err != nil {
+		t.Fatalf("parse: %v", err)
+	}
+	if !a.IsUnix() {
+		t.Fatalf("want unix, got %q", a.Network)
+	}
+	if a.Address != "/run/maven/stt.sock" {
+		t.Fatalf("address = %q", a.Address)
+	}
+	if a.Token != "" {
+		t.Fatalf("unix seam carries a token: %q", a.Token)
+	}
+}
+
+func TestParse(t *testing.T) {
+	cases := []struct {
+		in            string
+		net, addr, tk string
+		wantErr       bool
+	}{
+		{in: "", net: "unix", addr: ""},
+		{in: "unix:///run/maven/core.sock", net: "unix", addr: "/run/maven/core.sock"},
+		{in: "tcp://workstation:9310", net: "tcp", addr: "workstation:9310"},
+		{in: "tcp://workstation:9310?token=hunter2", net: "tcp", addr: "workstation:9310", tk: "hunter2"},
+		{in: "tcp://", wantErr: true},
+		{in: "udp://workstation:9310", wantErr: true},
+	}
+	for _, c := range cases {
+		a, err := Parse(c.in)
+		if c.wantErr {
+			if err == nil {
+				t.Errorf("Parse(%q) = %v, want error", c.in, a)
+			}
+			continue
+		}
+		if err != nil {
+			t.Errorf("Parse(%q): %v", c.in, err)
+			continue
+		}
+		if a.Network != c.net || a.Address != c.addr || a.Token != c.tk {
+			t.Errorf("Parse(%q) = %+v, want %s/%s/%s", c.in, a, c.net, c.addr, c.tk)
+		}
+	}
+}
+
+// The token must never reach a log line.
+func TestStringHidesToken(t *testing.T) {
+	a := MustParse("tcp://workstation:9310?token=hunter2")
+	if got := a.String(); got != "tcp://workstation:9310" {
+		t.Fatalf("String() = %q", got)
+	}
+}
+
+// A unix seam must round-trip with no handshake in front of the payload: the
+// first bytes the listener sees are the caller's, exactly as before.
+func TestUnixRoundTripHasNoHandshake(t *testing.T) {
+	a := MustParse(filepath.Join(t.TempDir(), "s.sock"))
+	ln, err := Listen(a)
+	if err != nil {
+		t.Fatalf("listen: %v", err)
+	}
+	defer ln.Close()
+	go echoOnce(ln)
+
+	c, err := Dial(a)
+	if err != nil {
+		t.Fatalf("dial: %v", err)
+	}
+	defer c.Close()
+	if got := roundTrip(t, c, "hello"); got != "hello" {
+		t.Fatalf("got %q", got)
+	}
+}
+
+func TestTCPRoundTripWithToken(t *testing.T) {
+	ln, addr := listenLoopback(t, "s3cret")
+	defer ln.Close()
+	go echoOnce(ln)
+
+	c, err := Dial(addr)
+	if err != nil {
+		t.Fatalf("dial: %v", err)
+	}
+	defer c.Close()
+	if got := roundTrip(t, c, "hello"); got != "hello" {
+		t.Fatalf("got %q", got)
+	}
+}
+
+func TestTCPWrongTokenIsRejected(t *testing.T) {
+	ln, addr := listenLoopback(t, "s3cret")
+	defer ln.Close()
+	// Accept keeps waiting past the bad peer, so nothing here should ever
+	// reach the echo. A conn that does means the token was not checked.
+	go echoOnce(ln)
+
+	bad := addr
+	bad.Token = "wrong"
+	if _, err := Dial(bad); !errors.Is(err, ErrUnauthorized) {
+		t.Fatalf("dial with wrong token: err = %v, want ErrUnauthorized", err)
+	}
+}
+
+// A stranger that speaks the protocol instead of the greeting is dropped, and
+// the listener stays up for the peer that follows it.
+func TestTCPUngreetedPeerDoesNotKillTheListener(t *testing.T) {
+	ln, addr := listenLoopback(t, "s3cret")
+	defer ln.Close()
+	go echoOnce(ln)
+
+	raw, err := net.Dial("tcp", addr.Address)
+	if err != nil {
+		t.Fatalf("raw dial: %v", err)
+	}
+	if _, err := raw.Write([]byte("GET / HTTP/1.1\n")); err != nil {
+		t.Fatalf("raw write: %v", err)
+	}
+	raw.Close()
+
+	c, err := Dial(addr)
+	if err != nil {
+		t.Fatalf("dial after stranger: %v", err)
+	}
+	defer c.Close()
+	if got := roundTrip(t, c, "still here"); got != "still here" {
+		t.Fatalf("got %q", got)
+	}
+}
+
+// A tcp seam with no token is a misconfiguration, and it must fail at bind
+// rather than serve the owner's turns to anyone who connects.
+func TestTCPListenRequiresToken(t *testing.T) {
+	if _, err := Listen(MustParse("tcp://127.0.0.1:0")); err == nil {
+		t.Fatal("listen on a tokenless tcp seam succeeded")
+	}
+}
+
+func listenLoopback(t *testing.T, token string) (*Listener, Addr) {
+	t.Helper()
+	ln, err := Listen(Addr{Network: "tcp", Address: "127.0.0.1:0", Token: token})
+	if err != nil {
+		t.Fatalf("listen: %v", err)
+	}
+	return ln, Addr{Network: "tcp", Address: ln.Addr().String(), Token: token}
+}
+
+func echoOnce(ln *Listener) {
+	c, err := ln.Accept()
+	if err != nil {
+		return
+	}
+	defer c.Close()
+	buf := make([]byte, 256)
+	n, err := c.Read(buf)
+	if err != nil {
+		return
+	}
+	_, _ = c.Write(buf[:n])
+}
+
+func roundTrip(t *testing.T, c net.Conn, msg string) string {
+	t.Helper()
+	if _, err := c.Write([]byte(msg)); err != nil {
+		t.Fatalf("write: %v", err)
+	}
+	buf := make([]byte, 256)
+	n, err := c.Read(buf)
+	if err != nil {
+		t.Fatalf("read: %v", err)
+	}
+	return string(buf[:n])
+}

internal/phraser/eval/checks.go

@@ -15,7 +15,7 @@ const (
 	CheckLang     = "lang"     // the operator's language, not the prompt's
 	CheckLength   = "length"   // a nudge is one sentence, not a paragraph
 	CheckFeminine = "feminine" // her self-reference is feminine (hard constraint)
-	CheckCringe   = "cringe"   // DESIGN.md § Non-goals, "not a relationship"
+	CheckCringe   = "cringe"   // docs/design.md § Non-goals, "not a relationship"
 	CheckOnTopic  = "ontopic"  // says the thing the rule is about
 
 	// CheckHisGender — the other half of the persona rule: SHE is feminine, HE
@@ -105,7 +105,7 @@ func checkLength(body string) Result {
 
 // --- feminine self-reference ---------------------------------------------
 //
-// The hard constraint (CLAUDE.md, DESIGN.md § Identity): Maven's Russian
+// The hard constraint (CLAUDE.md, docs/design.md § Identity): Maven's Russian
 // self-reference is feminine. The operator is male, so second-person forms
 // addressed to him are MASCULINE and must not be flagged — "ты не пил воду" is
 // correct, "я напомнил" is not. Both directions matter, which is why this is a
@@ -497,7 +497,7 @@ func isLatinWord(w string) bool {
 
 // --- the cringe checks ---------------------------------------------------
 //
-// "Think Jarvis without the cringe part". DESIGN.md § Non-goals: "Not a
+// "Think Jarvis without the cringe part". docs/design.md § Non-goals: "Not a
 // relationship — mom-tone is a function that makes nudges land, not emotional
 // company. Names the drift a warm small model falls into." Each pattern below
 // is one shape of that drift. They are deliberately specific: a check that

internal/phraser/eval/eval.go

@@ -11,7 +11,7 @@
 // length test that a human can read and disagree with. A score here is a claim
 // about measurable properties, not about whether a sentence is good.
 //
-// DESIGN.md § "Rules decide, LLM phrases" is why there is no send/veto signal
+// docs/design.md § "Rules decide, LLM phrases" is why there is no send/veto signal
 // anywhere in this package: the rule already decided she speaks. The phraser
 // only words it, so a nudge the model refuses to write is a failure, never a
 // legitimate outcome.
@@ -38,7 +38,7 @@ var fixtureJSON []byte
 const SchemaVersion = 1
 
 // Case — one nudge situation, as a real tick would present it. The fields are
-// the (rule, severity, context) input DESIGN.md names, flattened to JSON.
+// the (rule, severity, context) input docs/design.md names, flattened to JSON.
 //
 // WantAny is the on-topic contract: at least one of these lowercased fragments
 // must appear in the message. A water nudge that never mentions water is a

internal/phraser/llmphraser.go

@@ -46,6 +46,12 @@ type LLMPhraser struct {
 	launch func(ctx context.Context, cfg Config) (backend, error)
 	probe  func(ctx context.Context, base string) (string, error)
 
+	// remote — the workstation model, when one is configured. Set once at wiring
+	// time by UseRemote and read on every phrasing call. nil ⇒ every call goes to
+	// the resident llama-server this phraser owns, which is the whole deploy
+	// before a `workstation` block exists. See world.go.
+	remote Remote
+
 	// swapMu — single-flight around Swap. Held for the whole swap, including the
 	// model load, so two concurrent swap requests can never both be loading.
 	swapMu sync.Mutex
@@ -363,10 +369,7 @@ func (p *LLMPhraser) PhraseQuery(ctx context.Context, utterance string, notes []
 	// prompt guaranteed to make a small model fill the gap from memory.
 	notes = nonEmpty(notes)
 	if len(notes) == 0 {
-		// General knowledge — no notes to ground the answer. The system
-		// prompt is the single tested source in router.KnowledgePrompt.
-		sys := persona.Prepend(p.cfg.ContextBlock, router.KnowledgePrompt())
-		prompt := fmt.Sprintf("Пользователь спрашивает: \"%s\".", utterance)
+		sys, prompt := p.knowledgePrompt(utterance)
 		resp, err := p.chatWithSystem(ctx, sys, prompt, 768)
 		if err != nil || resp == "" {
 			return "не знаю.", nil
@@ -381,11 +384,7 @@ func (p *LLMPhraser) PhraseQuery(ctx context.Context, utterance string, notes []
 		}
 		return resp, nil
 	}
-	sys := p.querySystemPrompt()
-	prompt := fmt.Sprintf(
-		"Он спрашивает: \"%s\"\n\nИсточники:\n%s\nОтветь ему коротко и своими словами, опираясь только на эти источники. Если ответа в них нет — так и скажи.",
-		utterance, evidenceBlock(notes),
-	)
+	sys, prompt := p.evidencePrompt(utterance, notes)
 	resp, err := p.chatWithSystem(ctx, sys, prompt, 768)
 	text, _, perr := parseResponseMood(resp)
 	if err != nil || perr != nil {
@@ -481,6 +480,17 @@ func chatSystemPrompt(block func() string) string {
 // the LLM completion endpoint. Like chatWithSystem but for an arbitrary message
 // slice — the caller owns the system prompt placement.
 func (p *LLMPhraser) chatWithMessages(ctx context.Context, msgs []chatMsg, maxTokens int) (string, error) {
+	// Same silent preference as chatWithSystem, when the array is the shape
+	// llm.Req can carry: one system turn and one user turn. PhraseChat already
+	// folds the history into a single user message (some chat templates reject
+	// consecutive user turns), so today that is every call. A longer array goes
+	// to the resident model rather than get flattened here, because flattening a
+	// conversation is a decision its owner should make.
+	if len(msgs) == 2 && msgs[0].Role == "system" && msgs[1].Role == "user" {
+		if out, ok := p.remoteChat(ctx, msgs[0].Content, msgs[1].Content, maxTokens); ok {
+			return out, nil
+		}
+	}
 	base, release, err := p.acquire()
 	if err != nil {
 		return "", err
@@ -534,8 +544,14 @@ func (p *LLMPhraser) PhraseReminder(ctx context.Context, d loop.ReminderDecision
 		text = "reminder"
 	}
 
+	// Russian, like the other two prompts (Vikunja #404). Asking a model for a
+	// Russian reply in English is asking it to switch languages mid-prompt,
+	// and a 1.7B sometimes answers in the language it was asked in. The
+	// persona rules and the JSON contract are not repeated here: this call
+	// goes through chat(), so nudgeSystem already states both, and a second
+	// statement of the same contract is one more thing that can drift.
 	prompt := fmt.Sprintf(
-		`The user set a reminder: "%s". Rephrase it briefly as a gentle nudge. Respond as JSON: {"response": "...", "mood": "..."}`,
+		`Он поставил напоминание: "%s". Скажи это своими словами, коротко и мягко — одно предложение.`,
 		text,
 	)
 	resp, err := p.chat(ctx, prompt)
@@ -634,6 +650,12 @@ func (p *LLMPhraser) chat(ctx context.Context, userPrompt string) (string, error
 }
 
 func (p *LLMPhraser) chatWithSystem(ctx context.Context, system, user string, maxTokens int) (string, error) {
+	// The workstation model first when it will take work, and silently: every
+	// caller of this helper is on the silent half of the degradation rule. It
+	// answering is not news, and it being asleep is not news either.
+	if out, ok := p.remoteChat(ctx, system, user, maxTokens); ok {
+		return out, nil
+	}
 	base, release, err := p.acquire()
 	if err != nil {
 		return "", err
@@ -692,7 +714,7 @@ func (p *LLMPhraser) chatWithSystem(ctx context.Context, system, user string, ma
 // Written as filled-in examples, not as a schema with "..." in it. A 0.8B
 // copies whatever sits in the response slot, so a literal placeholder there
 // teaches it to answer with the placeholder. Measured: 7/15 nudges came back
-// as "..." before this. See PHRASING-EVAL-31-07-2026.md.
+// as "..." before this. See docs/evals/2026-07-31-phrasing.md.
 //
 // Russian only, feminine self-reference, second person masculine (the owner is
 // a man). She talks TO him, informally, singular — never "вы", never "он".
@@ -727,6 +749,27 @@ func (p *LLMPhraser) systemPrompt() string {
 	return persona.Prepend(p.cfg.ContextBlock, nudgeSystem)
 }
 
+// knowledgePrompt — the no-sources branch: a world question, answered from
+// weights alone. The system prompt is the single tested source in
+// router.KnowledgePrompt.
+//
+// Split out of PhraseQuery so PhraseWorld sends the workstation model the same
+// bytes the resident model gets. Prompt parity across two models is a stated
+// constraint (CLAUDE.md), and two copies of a prompt is how it stops holding.
+func (p *LLMPhraser) knowledgePrompt(utterance string) (sys, user string) {
+	return persona.Prepend(p.cfg.ContextBlock, router.KnowledgePrompt()),
+		fmt.Sprintf("Пользователь спрашивает: \"%s\".", utterance)
+}
+
+// evidencePrompt — the sources branch: read these, add nothing. Shared with
+// PhraseWorld for the same reason as knowledgePrompt.
+func (p *LLMPhraser) evidencePrompt(utterance string, notes []string) (sys, user string) {
+	return p.querySystemPrompt(), fmt.Sprintf(
+		"Он спрашивает: \"%s\"\n\nИсточники:\n%s\nОтветь ему коротко и своими словами, опираясь только на эти источники. Если ответа в них нет — так и скажи.",
+		utterance, evidenceBlock(notes),
+	)
+}
+
 // querySystemPrompt returns the system prompt for the evidence branch of
 // PhraseQuery. Prepends the configured persona when set.
 //
@@ -754,7 +797,7 @@ func (p *LLMPhraser) querySystemPrompt() string {
 	base := "Ты отвечаешь ему по источникам, которые тебе дали. Отвечай ТОЛЬКО по ним: всё, что ты говоришь, должно быть написано в источниках. " +
 		"Если ответа в них нет — так и скажи и на этом остановись; не добавляй ничего из своих знаний и не догадывайся. " +
 		"Не приплетай прошлые реплики разговора. " +
-		"Отвечай по-русски, коротко и своими словами, начинай с \"вот что я нашла: \". О себе — в женском роде, глаголы в прошедшем времени с окончанием -ла. Он мужчина, обращайся к нему на \"ты\". Respond ONLY with valid JSON: {\"response\": \"...\", \"mood\": \"neutral\"}."
+		"Отвечай по-русски, коротко и своими словами, начинай с \"вот что я нашла: \". О себе — в женском роде, глаголы в прошедшем времени с окончанием -ла. Он мужчина, обращайся к нему на \"ты\". Отвечай ТОЛЬКО одним объектом JSON: {\"response\": \"...\", \"mood\": \"neutral\"}."
 	return persona.Prepend(p.cfg.ContextBlock, base)
 }
 

internal/phraser/phraser.go

@@ -1,9 +1,9 @@
 // Package phraser is maven's "rules decide, llm phrases" seam — the layer
 // that turns a loop decision into the body + summary the delivery module ships.
 //
-// Per DESIGN.md § Resident language model: the phraser is the resident model
+// Per docs/design.md § Resident language model: the phraser is the resident model
 // (Qwen3-1.7B — RU continued pretraining plus joint persona/router SFT, not a
-// sub-1b prompted-only model as the retired spec claimed; see DESIGN.md
+// sub-1b prompted-only model as the retired spec claimed; see docs/design.md
 // § Superseded, "small-model phrasing claim"). It takes
 // (rule, severity, context) and produces Body (full voice message, local — no
 // shoulder-surf concern beyond who's in the room) + Summary (minimal body for

internal/phraser/spawn_test.go

@@ -0,0 +1,256 @@
+package phraser
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"regexp"
+	"strconv"
+	"strings"
+	"syscall"
+	"testing"
+	"time"
+)
+
+// The spawn path (NewLLMPhraser, spawnLlamaServer, startLlamaProc, llamaProc.Close)
+// was at 0% coverage: every test built the phraser with NewLLMPhraserAt, which
+// starts no process. These tests drive the real spawn code against a fake
+// llama-server script, so the startup race arms and the reaping are exercised
+// without a model or a GPU.
+
+// fakeLlama writes an executable script standing in for llama-server and returns
+// its path. body runs after the script has recorded its own pid.
+func fakeLlama(t *testing.T, body string) string {
+	t.Helper()
+	dir := t.TempDir()
+	path := filepath.Join(dir, "fake-llama-server")
+	script := "#!/bin/sh\n" + body + "\n"
+	if err := os.WriteFile(path, []byte(script), 0o755); err != nil {
+		t.Fatalf("write fake server: %v", err)
+	}
+	return path
+}
+
+// listensThenSleeps prints the line startLlamaProc scrapes, then stays alive
+// until killed — the shape of a real llama-server that came up.
+const listensThenSleeps = `echo "srv    load_model: listening on http://127.0.0.1:18081" >&2
+while : ; do sleep 1 ; done`
+
+func testCfg(bin string) Config {
+	cfg := DefaultConfig("/nonexistent/model.gguf")
+	cfg.BinPath = bin
+	return cfg
+}
+
+func TestExtractPort(t *testing.T) {
+	for _, tc := range []struct{ in, want string }{
+		{"127.0.0.1:0", "0"},
+		{"127.0.0.1:8080", "8080"},
+		{"127.0.0.1:", "0"},
+		{"", "0"},
+		{"8080", "0"}, // no colon: Cut yields no port, so the caller gets the "any port" default
+	} {
+		if got := extractPort(tc.in); got != tc.want {
+			t.Errorf("extractPort(%q) = %q, want %q", tc.in, got, tc.want)
+		}
+	}
+}
+
+func TestStartLlamaProcScrapesPortAndReaps(t *testing.T) {
+	bin := fakeLlama(t, listensThenSleeps)
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	p, err := startLlamaProc(ctx, testCfg(bin))
+	if err != nil {
+		t.Fatalf("startLlamaProc: %v", err)
+	}
+	if p.BaseURL() != "http://127.0.0.1:18081" {
+		t.Fatalf("BaseURL = %q, want the scraped address", p.BaseURL())
+	}
+	pid := p.cmd.Process.Pid
+
+	p.cancel = cancel
+	if err := p.Close(); err != nil {
+		t.Fatalf("Close: %v", err)
+	}
+	// Close must Wait, otherwise the child lingers as a zombie.
+	if p.cmd.ProcessState == nil {
+		t.Fatal("Close did not reap the child: ProcessState is nil")
+	}
+	if err := syscall.Kill(pid, 0); err == nil {
+		t.Fatalf("child %d still exists after Close", pid)
+	}
+}
+
+func TestStartLlamaProcFailureArms(t *testing.T) {
+	t.Run("binary missing", func(t *testing.T) {
+		cfg := testCfg(filepath.Join(t.TempDir(), "does-not-exist"))
+		_, err := startLlamaProc(context.Background(), cfg)
+		if err == nil || !strings.Contains(err.Error(), "llm: start") {
+			t.Fatalf("err = %v, want a start failure", err)
+		}
+	})
+
+	t.Run("server exits without listening", func(t *testing.T) {
+		// stderr closes, so the reader goroutine reports EOF on errCh.
+		bin := fakeLlama(t, `echo "ggml_vulkan: no device" >&2
+exit 1`)
+		_, err := startLlamaProc(context.Background(), testCfg(bin))
+		if err == nil || !strings.Contains(err.Error(), "llm: server output") {
+			t.Fatalf("err = %v, want the server-output arm", err)
+		}
+	})
+
+	t.Run("context cancelled during startup", func(t *testing.T) {
+		// Never prints the listen line and never exits: only ctx can end this.
+		bin := fakeLlama(t, `while : ; do sleep 1 ; done`)
+		ctx, cancel := context.WithCancel(context.Background())
+		go func() {
+			time.Sleep(150 * time.Millisecond)
+			cancel()
+		}()
+		defer cancel()
+		_, err := startLlamaProc(ctx, testCfg(bin))
+		if !errors.Is(err, context.Canceled) {
+			t.Fatalf("err = %v, want context.Canceled", err)
+		}
+	})
+}
+
+func TestNewLLMPhraserSpawns(t *testing.T) {
+	bin := fakeLlama(t, listensThenSleeps)
+	p, err := NewLLMPhraser(context.Background(), testCfg(bin))
+	if err != nil {
+		t.Fatalf("NewLLMPhraser: %v", err)
+	}
+	if p.BaseURL() != "http://127.0.0.1:18081" {
+		t.Fatalf("BaseURL = %q", p.BaseURL())
+	}
+	pid := p.be.(*llamaProc).cmd.Process.Pid
+	if err := p.Close(); err != nil {
+		t.Fatalf("Close: %v", err)
+	}
+	if p.BaseURL() != "" {
+		t.Fatalf("BaseURL after Close = %q, want empty", p.BaseURL())
+	}
+	if err := syscall.Kill(pid, 0); err == nil {
+		t.Fatalf("llama-server %d survived Close", pid)
+	}
+}
+
+func TestNewLLMPhraserSpawnFailure(t *testing.T) {
+	cfg := testCfg(filepath.Join(t.TempDir(), "does-not-exist"))
+	p, err := NewLLMPhraser(context.Background(), cfg)
+	if err == nil {
+		p.Close()
+		t.Fatal("want an error when the server cannot start")
+	}
+	if p != nil {
+		t.Fatalf("want a nil phraser on failure, got %#v", p)
+	}
+}
+
+// TestPdeathsigKillsOrphan is the orphan test the task asked for. A SIGKILLed
+// mavend never runs Close, so nothing but the kernel's Pdeathsig can stop its
+// llama-server. Re-exec this test binary as the "daemon", let it spawn the fake
+// server, SIGKILL the daemon, and assert the grandchild died with it.
+func TestPdeathsigKillsOrphan(t *testing.T) {
+	bin := fakeLlama(t, listensThenSleeps)
+
+	cmd := exec.Command(os.Args[0], "-test.run=TestSpawnHelperProcess", "-test.v=false")
+	cmd.Env = append(os.Environ(), "MAVEN_SPAWN_HELPER=1", "MAVEN_FAKE_LLAMA="+bin)
+	out, err := cmd.StdoutPipe()
+	if err != nil {
+		t.Fatalf("stdout pipe: %v", err)
+	}
+	if err := cmd.Start(); err != nil {
+		t.Fatalf("start helper: %v", err)
+	}
+	defer func() { _ = cmd.Process.Kill(); _ = cmd.Wait() }()
+
+	buf := make([]byte, 256)
+	n, err := out.Read(buf)
+	if err != nil {
+		t.Fatalf("read child pid: %v", err)
+	}
+	childPID, err := strconv.Atoi(strings.TrimSpace(string(buf[:n])))
+	if err != nil {
+		t.Fatalf("helper printed %q, want a pid: %v", string(buf[:n]), err)
+	}
+	if err := syscall.Kill(childPID, 0); err != nil {
+		t.Fatalf("llama-server %d not running before the kill: %v", childPID, err)
+	}
+
+	// SIGKILL: the helper gets no chance to clean up, exactly like an OOM kill.
+	if err := cmd.Process.Signal(syscall.SIGKILL); err != nil {
+		t.Fatalf("kill helper: %v", err)
+	}
+	_, _ = cmd.Process.Wait()
+
+	deadline := time.Now().Add(5 * time.Second)
+	for time.Now().Before(deadline) {
+		if err := syscall.Kill(childPID, 0); err != nil {
+			return // gone: Pdeathsig did its job
+		}
+		time.Sleep(20 * time.Millisecond)
+	}
+	_ = syscall.Kill(childPID, syscall.SIGKILL)
+	t.Fatalf("llama-server %d outlived the SIGKILLed parent", childPID)
+}
+
+// TestSpawnHelperProcess is not a test. It is the child half of
+// TestPdeathsigKillsOrphan: spawn a llama-server, print its pid, then block.
+func TestSpawnHelperProcess(t *testing.T) {
+	if os.Getenv("MAVEN_SPAWN_HELPER") != "1" {
+		t.Skip("helper for TestPdeathsigKillsOrphan")
+	}
+	cfg := testCfg(os.Getenv("MAVEN_FAKE_LLAMA"))
+	p, err := startLlamaProc(context.Background(), cfg)
+	if err != nil {
+		fmt.Println("spawn failed:", err)
+		os.Exit(1)
+	}
+	fmt.Println(p.cmd.Process.Pid)
+	os.Stdout.Sync()
+	select {} // wait to be killed
+}
+
+// TestKillMavenScriptMatchesRealCommandLine pins kill-maven.sh's fallback
+// pattern to the command line startLlamaProc actually builds. The script leaked
+// orphans twice already, both times because the pattern stopped matching: first
+// `llama-server.*maven`, then a hardcoded model name after the model was swapped.
+func TestKillMavenScriptMatchesRealCommandLine(t *testing.T) {
+	src, err := os.ReadFile("../../kill-maven.sh")
+	if err != nil {
+		t.Fatalf("read kill-maven.sh: %v", err)
+	}
+	m := regexp.MustCompile(`(?m)^\s*LLM='([^']+)'`).FindSubmatch(src)
+	if m == nil {
+		t.Fatal("no default LLM='...' pattern in kill-maven.sh")
+	}
+	pat, err := regexp.Compile(string(m[1]))
+	if err != nil {
+		t.Fatalf("LLM pattern %q does not compile: %v", m[1], err)
+	}
+
+	// Rebuild the command line from the production arg list, so a change to
+	// startLlamaProc that breaks the sweep fails here instead of on the box.
+	cfg := DefaultConfig("/opt/maven/models/llm/Qwen3-1.7B-UD-Q4_K_XL.gguf")
+	cfg.NCtx, cfg.NGpuLayers = 4096, 99
+	cmdline := strings.Join([]string{
+		cfg.BinPath,
+		"-m", cfg.ModelPath,
+		"--host", "127.0.0.1",
+		"--port", extractPort(cfg.Listen),
+		"-c", fmt.Sprintf("%d", cfg.NCtx),
+		"-ngl", fmt.Sprintf("%d", cfg.NGpuLayers),
+		"--no-webui",
+	}, " ")
+	if !pat.MatchString(cmdline) {
+		t.Fatalf("kill-maven.sh pattern %q does not match %q — orphans would leak", m[1], cmdline)
+	}
+}

internal/phraser/world.go

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

internal/phraser/world_test.go

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

internal/router/eval/llmrouter_test.go

@@ -39,7 +39,7 @@ import (
 // Re-measured with everything else held equal, thinking off scores exactly the
 // same, case for case — and a direct probe shows this llama-server build ignores
 // enable_thinking / reasoning_budget for this model anyway, so there was nothing
-// to turn off. Full write-up in ROUTING-EVAL-31-07-2026.md (Vikunja #376).
+// to turn off. Full write-up in docs/evals/2026-07-31-routing.md (Vikunja #376).
 func TestLLMRouterBaseline(t *testing.T) {
 	base := os.Getenv("MAVEN_LLM_URL")
 	if base == "" {

internal/router/intent.go

@@ -1,13 +1,13 @@
 // Package router is maven's reactive path — the cascade that turns a free-form
 // utterance into a deterministic Decision.
 //
-// Spec contract (from DESIGN.md § Reactive path — routing):
+// Spec contract (from docs/design.md § Reactive path — routing):
 //
 //   - the TARGET design is LLM-as-router: the resident model (Qwen3-1.7B)
 //     emits GBNF-constrained structured JSON for the route, and the same
 //     model phrases replies; the embedder is a RAG hint, not a routing gate.
 //     the classifier/embedder cascade below is the committed default today,
-//     but it is an interim stopgap (DESIGN.md § Superseded, "classifier-owns-
+//     but it is an interim stopgap (docs/design.md § Superseded, "classifier-owns-
 //     the-route") and the known cause of weak RU query handling — not a
 //     design to extend.
 //   - a CASCADE, not one decider — layers:
@@ -35,7 +35,7 @@ package router
 
 import "time"
 
-// Intent — the seven save-where labels from DESIGN.md's routing table. The
+// Intent — the seven save-where labels from docs/design.md's routing table. The
 // discriminator is "does the loop evaluate a predicate against it?":
 //
 //   - act:      command now, not stored (function call into the allowlist)

internal/router/knowledge.go

@@ -6,5 +6,5 @@ func KnowledgePrompt() string {
 	// No self-introduction here: the shared persona block already says who she
 	// is, and this line used to disagree with it — a different name ("Мавена")
 	// and a masculine noun ("ассистент") in front of a feminine persona.
-	return `Ответь кратко из своих знаний. Если не знаешь — скажи "не знаю". Не выдумывай. Respond ONLY with valid JSON: {"response": "...", "mood": "neutral"}.`
+	return `Ответь кратко из своих знаний. Если не знаешь — скажи "не знаю". Не выдумывай. Отвечай ТОЛЬКО одним объектом JSON: {"response": "...", "mood": "neutral"}.`
 }

internal/router/question.go

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

internal/router/question_test.go

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

internal/store/facts.go

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

internal/store/factvectors.go

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

internal/store/factvectors_test.go

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

internal/store/memory.go

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

internal/store/voidvectors_test.go

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

internal/stt/stt.go

@@ -18,9 +18,9 @@
 // returns a canned string the router + action path operate on); with a
 // worker socket configured, it wires Remote.
 //
-// Per DESIGN.md § Voice pipeline (STT / TTS): whisper.cpp (CGo, Vulkan) in
+// Per docs/design.md § Voice pipeline (STT / TTS): whisper.cpp (CGo, Vulkan) in
 // cmd/mavsttd is the production stt — the older faster-whisper/vosk picks are
-// retired (DESIGN.md § Superseded, "named STT/TTS model picks"). The
+// retired (docs/design.md § Superseded, "named STT/TTS model picks"). The
 // server-side stt module is the heavy multilingual path; the client's
 // wake-word + stage-0 command grammar (cmd/mavwaked) hits the router directly
 // and never crosses this seam. Today's Remote + Stub both return plain text

internal/tool/tool.go

@@ -2,7 +2,7 @@
 // store's allowlist, and drafts 'proposed' scaffolds for acts that aren't on
 // it yet.
 //
-// Boundary discipline (DESIGN.md § "Tool registration — drafting is suggest,
+// Boundary discipline (docs/design.md § "Tool registration — drafting is suggest,
 // enabling is act"):
 //
 //   - The store is the allowlist. Only status='enabled' rows run. A verb not
@@ -26,7 +26,7 @@
 //   - Destructive tools don't run on first hearing: Exec returns ErrNeedsConfirm
 //     and the handler runs a confirm turn ("выполнить X? да/нет"); only a
 //     confirmed re-Exec runs them. A gate assumes a fully-formed action, which
-//     an enabled+matched act is (DESIGN.md § "Confirmation is not one
+//     an enabled+matched act is (docs/design.md § "Confirmation is not one
 //     mechanism").
 package tool
 

internal/tts/tts.go

@@ -5,9 +5,9 @@
 // PCM, headerless per the audio package; the voice sink + reference client
 // wrap it in a WAV at the disk edge.
 //
-// Per DESIGN.md § Voice pipeline (STT / TTS): piper is the production tts
+// Per docs/design.md § Voice pipeline (STT / TTS): piper is the production tts
 // (subprocess + espeak-ng, CPU-only on the ryzen box, driven by cmd/mavttsd);
-// the older silero pick is retired (DESIGN.md § Superseded, "named STT/TTS
+// the older silero pick is retired (docs/design.md § Superseded, "named STT/TTS
 // model picks"). A different voice is a model-file swap, not a code change.
 // The daemon wires one impl — Remote pointing at the worker socket if
 // configured, Stub otherwise.

internal/worker/client.go

@@ -24,6 +24,8 @@ import (
 	"net"
 	"sync"
 	"time"
+
+	"github.com/kami/maven/internal/netaddr"
 )
 
 // Client — one connection to one worker module. NOT goroutine-safe for
@@ -38,15 +40,25 @@ type Client struct {
 	dial func() (net.Conn, error)
 }
 
-// Dial opens a Client to the worker socket at path. The first call lazily
+// Dial opens a Client to the worker module at path. The first call lazily
 // dials; subsequent calls reuse the conn (a fresh dial happens on next call
 // after a teardown). Lazy dial keeps a worker that's restarting from
 // blocking core's startup; core attempts the dial on first use.
+//
+// path is a netaddr seam address. A bare path is the unix socket it has
+// always been; "tcp://workstation:9310?token=..." reaches a module on another
+// host, which is how stt and tts move to the machine with the GPU and the
+// microphone. A bad address surfaces on the first call, not here, because
+// Dial does not fail — see internal/netaddr.
 func Dial(path string) *Client {
+	addr, err := netaddr.Parse(path)
 	return &Client{
 		path: path,
 		dial: func() (net.Conn, error) {
-			return net.Dial("unix", path)
+			if err != nil {
+				return nil, err
+			}
+			return netaddr.Dial(addr)
 		},
 	}
 }

internal/worker/server.go

@@ -7,10 +7,11 @@
 // which module to dial; mixing the two is a config error caught cleanly by
 // the wire, not a runtime goroutine panic). One Server per module process.
 //
-// Socket perms mirror ipc.Server: dir 0700, socket 0600 ⇒ same unix user.
-// The module has no key, so the floor is "same user"; the wg/mTLS layers
-// are out of scope here (this socket never crosses the network radius —
-// it's local-only, point-to-point between two processes on the box).
+// The seam address decides the transport. On the default unix socket the
+// perms mirror ipc.Server — dir 0700, socket 0600 ⇒ same unix user — and that
+// is the whole auth floor, because the seam never leaves the box. A tcp
+// address moves the module to another host and takes that floor away, so
+// netaddr checks a shared token before the first frame. See internal/netaddr.
 package worker
 
 import (
@@ -18,11 +19,10 @@ import (
 	"encoding/json"
 	"fmt"
 	"net"
-	"os"
 	"sync"
 	"sync/atomic"
 
-	"golang.org/x/sys/unix"
+	"github.com/kami/maven/internal/netaddr"
 )
 
 // Server — a worker module process's listener. Wires either a Transcriber,
@@ -34,6 +34,7 @@ type Server struct {
 	s Synthesizer
 
 	path string
+	addr netaddr.Addr
 	ln   net.Listener
 
 	wg   sync.WaitGroup
@@ -61,25 +62,24 @@ func NewSynthesizerServer(path string, s Synthesizer) *Server {
 // two separate processes per the restart-free / fail-independent invariant).
 func (srv *Server) SetSynthesizer(s Synthesizer) { srv.s = s }
 
-// Listen binds the unix socket with 0700 dir + 0600 socket perms (same floor
-// as internal/ipc). A stale socket at path is removed first so the worker
-// process restarts cleanly after a crash, no manual cleanup needed.
+// Listen binds the seam address the Server was built with.
+//
+// A bare path is a unix socket with 0700 dir + 0600 socket perms, the same
+// floor as internal/ipc, and a stale socket is removed first so the worker
+// process restarts cleanly after a crash. A "tcp://host:port?token=..."
+// address binds a network listener instead, so this module can run on the
+// workstation while core stays on homesrv; the token is mandatory there,
+// because there is no filesystem to be the auth floor. See internal/netaddr.
 func (srv *Server) Listen() error {
-	_ = os.Remove(srv.path)
-	if err := os.MkdirAll(parentDir(srv.path), 0o700); err != nil {
-		return fmt.Errorf("worker: mkdir socket dir: %w", err)
-	}
-	oldMask := unix.Umask(0o077)
-	ln, err := net.Listen("unix", srv.path)
-	unix.Umask(oldMask)
+	addr, err := netaddr.Parse(srv.path)
 	if err != nil {
-		return fmt.Errorf("worker: listen %s: %w", srv.path, err)
+		return err
 	}
-	if err := os.Chmod(srv.path, 0o600); err != nil {
-		_ = ln.Close()
-		_ = os.Remove(srv.path)
-		return fmt.Errorf("worker: chmod socket: %w", err)
+	ln, err := netaddr.Listen(addr)
+	if err != nil {
+		return err
 	}
+	srv.addr = addr
 	srv.ln = ln
 	return nil
 }
@@ -193,7 +193,7 @@ func (srv *Server) Close() error {
 	}
 	err := srv.ln.Close()
 	srv.wg.Wait()
-	_ = os.Remove(srv.path)
+	netaddr.Cleanup(srv.addr)
 	return err
 }
 
@@ -214,15 +214,3 @@ func marshalResult(v any) json.RawMessage {
 	b, _ := json.Marshal(v)
 	return b
 }
-
-func parentDir(p string) string {
-	for i := len(p) - 1; i >= 0; i-- {
-		if p[i] == '/' {
-			if i == 0 {
-				return "/"
-			}
-			return p[:i]
-		}
-	}
-	return "."
-}