The router prompt tested "reports current state -> fact" before "wants
information -> query", so a question naming a fact key was written as a fact.
Query now comes first, plus an explicit question test.
Reviewers: the prompt block in llmrouter.go, and the note about the
training-side copy of the prompt that needs the same edit.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CGeSZxh1DCtRxmFVSYVGvJ
Completes #319's comparison. Three configurations, because "the LLM router"
was ambiguous: the model alone, the cascade #320 would actually ship (stage-0
grammar → model → classifier floor), and a thinking-off diagnostic.
intent-only full RU missed-clarify p50
classifier+onnx 36.8% 36.8% 25/61 5/6 31ms
llm-only (0.8B) 48.7% 23.7% 13/61 6/6 850ms
cascade+llm (0.8B) 50.0% 32.9% 18/61 6/6 825ms
On the question asked — does the resident model route better? — yes, 50.0%
vs 36.8% intent accuracy. REARCH.md's premise holds. It costs 27x the
latency (p50 825ms vs 31ms, max 3.1s), on the same llama-server the phraser
needs, so it is a trade rather than a free win.
Three things the numbers surface that the headline hides:
query→fact x15 is the dominant failure, four times the classifier's x4 on
the same axis. routeSystem's decision order puts "сообщает или обновляет
состояние" (rule 3) above "хочет получить информацию" (rule 4), so any
utterance naming a fact key matches the earlier rule and a question about
past state reads as an assertion of it. A prompt fix, not a model limit.
The LLM router cannot clarify: llmrouter.go hardcodes Confidence 1.0, so
stage 3's gate can never fire on its decisions — 6/6 missed. With #359's
finding that the classifier's gate is miscalibrated under ONNX, neither path
currently refuses. Flipping #320 as-is removes the refusal lane.
The gap between 50.0% intent and 32.9% full accuracy is entirely slots: the
LLM path fills neither Fn nor Time (it returns Slots.Text for acts, and
Extract never runs on an LLM decision).
Also settles a hypothesis rather than leaving it in the air: thinking mode is
a non-issue under a grammar (identical score), and the grammar's unbounded
("," ws action)* repetition that ran away in an isolated smoke test does not
reproduce under the real prompt — 2 errors in 76, not 76. internal/llm
deliberately does not grow a chat_template_kwargs field.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
The onnxruntime .so was already vendored at deps/onnxruntime-linux-x64-1.26.0
— nothing to download. make eval-router now defaults MAVEN_ONNX_LIB there, so
both baselines run by default and only a fresh clone without deps/ falls back
to the hash ratchet alone.
Prod-representative result, deployed 0.55 gate: 28/76 (36.8%), RU 25/61,
EN 3/15, hard 0/11 → 4/11, p50 31ms / p95 71ms. Versus the hash floor's
13/76 at p50 9µs.
The finding is not the accuracy, it's the refusal lane: missed clarifies went
0 → 5 of 6. Better embeddings raise cosine everywhere, so the 0.55 threshold
that used to hold ambiguous utterances back stops holding — "сделай это"
routes to act at 0.847, "бэкап" to chat at 0.755. The gate was implicitly
tuned to the hash floor's low similarities. That is an argument about the
threshold, not about the embedder, and it lands before #320 rather than after.
Also fixes a fixture-model mismatch: ReminderGrammar deliberately skips the
extractor at stage 0 and the daemon's applyAction parses the time downstream
(stage0.go says so). Charging the router for that slot made 4 exact-match wins
read as misses; they are now counted as SlotsDeferred instead. Hash baseline
moves 13/76, ratchet to 0.15.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
#319 asks for a measurement before #320 flips the route decider from the
classifier cascade to the resident model. There was nothing to measure
against: the only routing tests assert single utterances, and the
classifier's seed corpus is its own training set — scoring it there
measures memorisation of frozen centroids, which is the illusion that hid
the weak RU query handling in the first place.
internal/router/eval is a separate package so both paths can be scored
from outside router (including cmd/mavend, where the real llama-server
client lives). The fixture is embedded; the scorer takes a Router
interface, so *router.Router and a bare LLM stage both go through the same
76 cases.
The fixture is a CONTRACT, not a snapshot: cases the cascade fails today
stay in the file and fail loudly. TestFixtureIsHeldOut enforces that no
utterance appears verbatim in models/seeds/*.txt.
Baseline, hash embedder at the deployed 0.55 gate: 9/76 (11.8%), 63 false
clarifies, 0 missed clarifies, p50 9µs. Almost everything falls to the
confidence gate — the documented floor behaviour, not a new bug. The
number worth comparing is TestONNXBaseline's (skipped without
MAVEN_ONNX_LIB); the assertions here are a regression ratchet plus a tight
bound on the dangerous direction: ambiguous utterances must not start
being routed confidently.
Seeding is order-fixed on purpose — a few phrases appear under two intents
and map iteration handed them to a different centroid each run, which made
the score jitter between 9 and 10.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
Backlog item #3 (20-07-2026-BACKLOG.md). A morning routine is a checklist for
a daily window: several items, each evidenced by a fact key, completed in any
order, checked once near the end of the window. Modelling it as four
independent reminder timers would stack into exactly the kind of noise Maven is
supposed not to produce, so the engine nags at most once per day per routine
and only for what is actually still missing.
internal/morning follows the established pure-engine pattern (loop, routine,
pattern): no store, no clock of its own. Evaluate answers "what's still
missing" at any point; Due decides whether to nag. The impurity — reading
facts under the store lock, holding the last-nudge map across ticks — stays in
the tick driver, which calls Due each tick exactly as it does for loop.Rule
and routine.Routine.
Completion evidence is a fact key's latest non-voided value timestamped inside
today's window, so manual ("выпил воды", voice-tapped) and inferred (another
daemon writing the same key) are indistinguishable and both count. Weekdays
scopes which days a routine applies to, so weekday/weekend variants are two
routine rows rather than a special case in the engine.
Exposed read-only: a MorningStatus RPC over ipc, and a /morning page in mavweb
built on the same server-rendered shape as /trace — no live-update loop, since
checklist state moves on the scale of minutes.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
The most load-bearing decision in the project was stated four incompatible
ways: the docs said Qwen3-1.7B, deploy/mavend.json said Qwen3.5-2B, the repo's
models/llm/ held an LFM2.5-1.2B gguf, and five code comments still said LFM.
Answering "which model is deployed" meant re-deriving it from scratch every
time.
Two facts the review missed, found while resolving it:
- /mnt/hdd1/llms is bind-mounted over /opt/maven/models/llm, which shadows the
repo's models/llm/. The LFM2.5 gguf sitting there was never loaded by
anything, so it was not evidence of the deployed model at all.
- That library holds Qwen3.5-0.8B, -2B and -4B, and no Qwen3-1.7B. The config
pointed at a file that does exist; the docs' Qwen3-1.7B was the stale claim,
the reverse of the assumed direction. Qwen3-1.7B is the CPT target, and that
training is still in flight (Vikunja #122), so no such gguf exists yet.
phraser.model_path moves to Qwen3.5-0.8B (Q4_K_M) — the smallest checkpoint on
disk, chosen for latency, and relevant to whether the LLM router is affordable
on this box. Docs and comments now say the same thing in one voice: 0.8B
resident now, CPT'd Qwen3-1.7B as the target, and the bind-mount shadowing
written down so the next reader does not mistake models/llm/ for ground truth.
Comments name the model, never a filename, so a swap stays a one-line config
change.
n_gpu_layers: 99 is correct and stays — compose passes /dev/dri and the render
gid for Vulkan offload to the Vega iGPU. CLAUDE.md's "CPU-only" was the stale
half of that contradiction and is corrected.
phraser.go also dropped a wrong "sub-1b, prompted not trained" size claim: the
target is trained end-to-end (RU CPT + joint persona/router SFT).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
15 root markdown files, ~4,900 lines against ~33,000 lines of Go, with at least
three pairs contradicting each other. When five documents describe the
architecture, the code becomes the only trustworthy one — which defeats the
point of having them. That drift is why the resident-model question had four
incompatible answers.
SPEC.md, maven.md and ROADMAP.md are deduped into DESIGN.md rather than
concatenated, with a "Superseded" section carrying eight retired decisions and
what replaced each: classifier-owns-the-route (the cascade is still the live
path, but as a stopgap, not a design to extend), faster-whisper/vosk/silero,
the small-model phrasing claim, sqlcipher, the Kotlin/Spring sketches,
obsidian->chroma, script deployment, and FloorEnrollment. Superseded material
is kept and marked rather than deleted, so it cannot read as current.
SESSION-05/06-07-2026.md and PLANS.md are removed outright — git history holds
them, and both were verified tracked before deletion.
Go doc comments citing the deleted files are repointed to the equivalent
DESIGN.md sections. Several asserted designs that were already retired, so the
claims are corrected and not just relinked: stt.go named faster-whisper as
production (it is whisper.cpp), tts.go named silero (it is piper), intent.go
still described the classifier as owning the route, and stale vosk/chroma
vocabulary is replaced. ECOSYSTEM-SPEC.md references are deliberately
untouched — that is a different document, and a naive grep for SPEC.md matches
it.
Root markdown drops from 4,880 to ~3,700 lines. The review's ~1,500 target is
not reachable while keeping the files it also said to keep — those alone are
2,553 lines — so trimming further needs a separate decision on
MAVEN_ECOSYSTEM_ARCHITECTURE.md and PROGRESS.md.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X5JApcrCRVGmqrxnhynSik
WriteFactReq gains an optional Subject field (empty = old behavior,
no CoreAPI signature change) and the IntentFact handler now passes the
fact's key as its resolution subject, so voice-tapped facts flow into
the Vikunja #279 enrichment queue automatically.
Also: 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, keeping GPU offload.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018ghELqYhZNLub2TXGMazqA
facts gain a Subject/EntityID/ResolutionState triple and an async
enrichment worker that resolves free-text subjects to canonical Nexus
entity_ids, mirroring Praxis's enrichment-worker pattern. Ambiguous or
unreachable Nexus never guesses an entity_id — the fact stays pending
or terminal-ambiguous instead.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018ghELqYhZNLub2TXGMazqA
Companion to the dispatcher-side outbox change: adds the
delivery_attempts table migration, Store.BeginDeliveryAttempt/
CompleteDeliveryAttempt/ReconcileStaleDeliveryAttempts, and wires
ReconcileStaleDeliveryAttempts + Config.Outbox into mavend startup
before the tick loop resumes.
Vikunja #270.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018ghELqYhZNLub2TXGMazqA
Closes the audit finding: a crash between 'sink accepted it' and 'we
recorded that' caused duplicate sends on the next tick with no trace.
BeginDeliveryAttempt now runs before Send, CompleteDeliveryAttempt
after — a stale 'pending' row found at startup reconciles to 'unknown'
(never silently resent, never silently dropped, same rule as the Hexis
execution engine's timeout handling). Wired into DispatchNudge,
DispatchReminder, and RepeatUnacked; ReconcileStaleDeliveryAttempts
runs once at mavend startup before the tick loop resumes.
9 new dispatcher tests cover begin-before-send ordering, failed-send
completion, the reminder path, and begin-failure not blocking send.
Vikunja #270.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018ghELqYhZNLub2TXGMazqA
Vikunja #269 (P0): Client.call() retried any connection-loss uniformly,
including the case where the request frame was already sent and the reply
never arrived — the server may have already committed the write before
dying, so a blind retry could double-apply it. This violates the ecosystem
rule against retrying an unknown mutation outcome.
- Split errConnLost into errWriteLost (request never sent — always safe to
retry) and errReadLost (request sent, reply lost — ambiguous).
- errReadLost is only auto-retried for read-only methods (replaying a read
can't double-apply). A mutation method instead returns ErrAmbiguousOutcome
so the caller can decide, rather than the boundary silently guessing.
- Added commit-then-disconnect regression tests: a mutation (WriteFact)
surfaces ErrAmbiguousOutcome and does not retry; a read (Presence) retries
transparently past the same disconnect timing.
Add a read-only /ecosystem page that consumes the sibling services'
JSON APIs (Nexus entities, Praxis attention, Hexis capabilities),
fetched concurrently with honest per-panel error states. Siblings stay
headless — mavweb is their human surface (arch §16). Wired via mavweb
-nexus/-praxis/-hexis flags; mavweb joins the ecosystem compose network.
Fix mobile horizontal overflow across all pages: .content is a flex
child with default min-width:auto, so it refused to shrink below the
tables' intrinsic width. min-width:0 lets wide tables pan inside .scroll
instead of dragging the page sideways. Verified via CDP geometry check
(scrollWidth === clientWidth at 430px).
Also includes in-progress Ethos UI redesign, ecosystem deploy compose,
and planning docs.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Bring the Nexus/Praxis/Hexis integration in line with
MAVEN_ECOSYSTEM_ARCHITECTURE.md:
- Praxis over HTTP: drop the in-process praxis.db open (praxisstore/
praxistools) and call praxisd's /api/v1/tools/* API via a new praxisClient.
Honors the "no component reads another's DB" invariant (AC#12).
PraxisConfig.DBPath -> URL.
- Hexis confirmation gate: mutating capabilities (ReadOnly=false) now park a
bound pendingHexis confirmation and require a spoken "да" before executing;
read-only run immediately (AC#7, no auto attention->action).
- Capability safety: >1 verb match is ambiguous -> ask instead of firing the
first; ambiguous Nexus resolution asks for clarification (AC#2).
- Correlation IDs on Hexis execute, recorded in the cross-service trace.
- Bug: importance arrives as JSON float64 over HTTP, not int.
- Tests: confirm-gate, decline, read-only, and ambiguity paths.
Build: vendor/ bakes in the hexis client (replace-directed at a sibling repo
outside the Docker context); Dockerfile builds from vendor and no longer
`go mod download`s the unreachable replace paths.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Route contract is now a JSON array of action objects (one per ask) so
compound utterances route all their intents, not just the first. Grammar
root emits `[{intent...},...]`; parseActions tolerates a bare object.
Cascade still returns one Decision — full N-action dispatch lands with the
engine turn-on (marked in-code).
Router prompt rewritten shorter + decision-ordered (prompt-guy feedback),
fact redefined as "implicit update" not "trackable state", kept in Russian
to match the CPT base + phraser. "интент" → "намерение".
CLAUDE.md: routing-architecture section + refreshed open items.
docs/plans: route-data generation plan.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017GMrVfuYN3nE4L1vEiFYC9
Daemon side of Decision B: parse {"response","mood"} across the 4 consumers
(replier, nudges, reminders, chat), fall back to legacy formats. Drop the
LLM router — the classifier handles routing; replier/phraser share one
llm.Client (timeout 20s->60s). llm.Client reads reasoning_content when
content is empty (thinking models).
Docs: TTS piper-student plan (OmniVoice teacher -> piper student, from
scratch, phoneme-first). CLAUDE.md training guide.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Add llmphraser: LFM-based phraser implementing Phraser interface with
PhraseChat, PhraseNudge, PhraseReactive, and PhraseReminder methods.
- Add shared internal/llm/client: llama-server completion client used by
both the phraser (talking back) and router (routing), sharing one model.
- Add LLMReplier in mavend: replaces StubReplier for chat/nudge/reactive
replies, falls back to stub on model errors.
- Update Phraser interface: add PhraseChat method, update stub to match.
- Wire LLM phaser into mavend voice init, plumb LLM config from JSON.
- Add LLMRouter: grammar-constrained LFM call for intent classification
after stage-0, before classifier cascade. Errors fall through gracefully.
- Add IntentChat: conversational intent with no store side-effect, routed
through LLM -> phraser chat endpoint.
- Extract slots for Chat: no structured slots, full utterance is payload.
- Extend stage-0 grammars to fire through Cyrillic wake-word spellings
(Мэйвен/Мейвен/Майвен/etc.) produced by Russian STT model.
- StripWakeToken helper strips leading wake in any script so time/date
grammars still match when wake is present.
- Add classifier examples for chat utterances (EN + RU).
- Wire LLMRouter into Router.Config; optional, nil-safe.
Store layer: ListReminders returns the n most recent reminders
(newest first). IPC: new MethodListReminders wired through server,
client, and lockedAPI. Web: /reminders page with table of created
time, fire time, status badge, and payload text; empty state with
prompt to ask maven for a reminder. Sidebar entry under Automation.
PythonDateParser shells out to python3 with the dateparser library for
full natural-language date/time extraction. Two-step approach:
1. search_dates() finds the date substring in surrounding text
2. parse() re-parses the substring for correct time resolution
Russian time qualifiers (утра/вечера/дня/ночи) are pre-processed to
AM/PM before parsing — dateparser drops them during substring extraction.
Falls back to StubDateTimeParser when python3 or dateparser isn't
available (graceful degradation, no hard runtime dependency).
Dockerfile updated: python3 + dateparser==1.4.1 in runtime stage.
five fixes spotted during routing investigation:
- gitignore: replace blanket models/ ignore with per-dir exceptions
(/models/embedder/, /models/stt/, /models/tts/) so the seed text
files under models/seeds/ are tracked in version control
- query.txt: fix merged line — 'сколько стоит свет в этом месяце' and
'найди заметку про сервер' were fused with no separator
- reminder.txt: add 11 pure-verb reminder seeds without time expressions
to shift centroid toward the reminding intent rather than time-lexicon
- StubDateTimeParser: add Russian 'через <N> <unit>'/'через час'/'через
полчаса', 'сегодня'/'завтра'/'послезавтра' with optional clock, and
'в <clock>' scan. Add Russian word numbers (один-десять) and unit
inflections (час/часа/часов, минута/минуты/минут, день/дня/дней,
неделя/недели/недель). Also adds missing English day/week units.
- replySystem: guard time branch against duration queries ('сколько
времени прошло') reaching it via the classifier path after the
stage-0 grammar's build filter rejects them. Mirrors stage0.go
duration keywords.
three bugs causing time queries to land on reminder or fact intent:
- seed collision: query.txt and system.txt shared identical time/date
seeds (который час, сколько времени), making system intent
indistinguishable from query intent in centroid space
- threshold (0.35) too low for ONNX embedder — cosine similarities
cluster 0.5-0.7 for related intents, so Clarify never fired
- reminder centroid contaminated by time-lexicon (every seed has a time
expression), pulling any time-word utterance toward reminder intent
fixes:
- remove 3 duplicate time/date seeds from query.txt (keep in system.txt)
- DefaultRouterThreshold 0.35 -> 0.55
- stage-0 grammar for напомни/remind me -> IntentReminder, bypasses
classifier (fixes 'напомни через час' being misrouted to fact)
- stage-0 grammars for time/date system queries (сколько времени,
который час, какой сегодня день) -> IntentSystem, with Build filter
to exclude elapsed/duration queries (сколько времени прошло)
- time parser fallback in applyAction for stage-0 reminder matches
(extractor doesn't run on stage-0 decisions)
- session.go: add History []Turn + Turn type for multi-turn context
- slots.go: add AnaphoraResolver with Resolve() for RU pronoun detection
(это/он/она/оно/тот/мой and inflected forms)
- followup.go: extend followUpMerge with cross-intent inheritance:
Query/Fact/Reminder after a Fact with anaphora inherits the key.
Same-intent path unchanged. Anaphora detection from utterance.
- voice.go: add fact-by-key lookup path in applyAction for IntentQuery
when dialogue resolved an anaphoric reference (calls LatestFact,
formats with formatTime helper). History tracked in Session.History
capped at 4 most recent turns.
- followup_test.go: 7 new test cases: anaphora query-after-fact,
no-inheritance-without-anaphora, three-turn break, anaphora in
reminder, anaphora in fact, explicit key wins, time inheritance.
make test green (303+, -race, all 29 packages).
- VoiceConfig: add QueryMinScore (default 0.55) + Persona config fields
- voice.go: remove queryMinScore const, wire from cfg.Voice.QueryMinScore
as reactiveHandler field
- llmphraser.go: add Persona to Config, prepend to system prompts in
chat and query paths (systemPrompt/querySystemPrompt methods)
- main.go: pass personaFromCfg into both phraser config blocks
- Makefile: add download-embedder target (Xenova/paraphrase-multilingual-
MiniLM-L12-v2, ~90MB ONNX)
- AGENTS.md: document embedder model download + libonnxruntime setup
- server.go: fix pre-existing wg.Add vs wg.Wait data race using accept
mutex. make test green, zero races across all 29 packages.
- webauthn.go: keyIPC interface for StoreEncryptionKey/Unlock, wired
through PasskeyHandle. RegisterFinish calls StoreEncryptionKey with
the credential's public key after successful enrollment. AssertFinish
calls Unlock with the stored public key after assertion (alongside
existing AssertStepUp call).
- server.go: fix data race on s.api by switching from bare CoreAPI field
to atomic.Value. SetAPI uses Store(), dispatch uses Load(). No more
race-flagged tests.
- make test green (303+, -race)
Two additive proactive/recall features.
Routines (internal/routine): a third proactive class beside reminders
(user-stated) and care rules (world-state) — operator-declared clockwork.
config.routines[] (cron + literal RU body + severity) fire through the
normal dispatcher on schedule. Bodies are literal, not LLM-phrased (can't
hallucinate); rule name routine:<name> keeps them out of the care
autotuner; a cold-start guard seeds on first sight so a restart never
replays a missed schedule. Pure routine.Due + config validation, unit-
tested; the tick driver holds the last-fired map and calls fireRoutines.
Persistent memory (internal/store/memory.go): store.MemoryStore backs the
memory.Store interface with the SAME encrypted sqlite db — survives
restarts and recall text inherits at-rest encryption (no plaintext
sidecar). float32-blob vectors, brute-force cosine (ANN is a later swap
behind the interface), upsert-by-id. The daemon wires st.VectorMemory()
into wireVoice; the in-memory impl stays the test/no-store floor. Closes
the "in-memory only, lost on restart" gap (PROGRESS #8).
Gate green: gofmt/vet clean, -race across routine/config/store/mavend.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01U2PNdwDj2Gt8YW294J7oSc
Task 4 created and tested router.KnowledgePrompt() but the live path in
LLMPhraser.PhraseQuery used a separate hardcoded copy of the same RU
anti-hallucination prompt, leaving KnowledgePrompt() as dead code and two
strings that could drift. Point the phraser at the tested helper so there
is a single source of truth.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Add CalendarEvents method to recordingAPI in auth_test.go
- Add CalendarEvents method to fakeCore in handlers_test.go
Co-Authored-By: opencode <opencode@anthropic.com>
- New internal/weather/ package: Provider interface, Weather struct, StubProvider
- OpenMeteoProvider with geocoding + current weather (keyless, free API)
- Config: WeatherConfig in VoiceConfig (provider, default_location)
- Wire in voice.go as weatherProvider on reactiveHandler
- Handle weather queries in IntentQuery (before notes RAG)
- Helper: isWeatherQuery / extractWeatherLocation
- Tests: mocked HTTP round-trip for OpenMeteo, stub ErrNotConfigured, config tests
- No real network calls in any test
Co-Authored-By: opencode <opencode@anthropic.com>
- Stub.PhraseQuery: empty notes → "не знаю." (was: "no notes")
- LLMPhraser.PhraseQuery: empty notes → general knowledge prompt to LLM
- Voice handler: on notes RAG failure, try phraser before giving up
- New router.KnowledgePrompt() pure function with test
Co-Authored-By: opencode <opencode@anthropic.com>
Store: CalendarEvents(ctx, from, to) — filters caldav facts by key date prefix.
IPC: full wiring through interface, server dispatch, and client.
Router: ParseCalendarDate (сегодня/завтра), CalendarEventFormatter (RU reply).
Voice: calendar detection before notes RAG in IntentQuery handler.
Tests: store integration test, date parser tests, formatter tests.
Co-Authored-By: opencode <opencode@anthropic.com>
- digest queue no longer dropped on failed dispatch (retry next tick)
- collapsed reminders marked fired/rescheduled only after digest delivers
- /tools step-up gate skipped when WebAuthn is not configured (was 403 forever)
- passkey credential store rolls back memory on persist failure
- auth_test fake updated for TickTrace (branch build break)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ASstMtsZWLSRcD1Tq8T68Q
ExplainTick and ExplainGate produce a full TickTrace for every tick,
recording per-rule: predicate result, gate result, first gate blocker,
gate detail (snooze/cooldown/presence etc.), and win/loss info.
IPC layer: new MethodTickTrace, DTOs (TickTrace, RuleTrace, GateDetail),
dispatch, client proxy, and CoreAPI interface method.
Daemon: tickLoop caches the latest trace; daemonAPI wraps storeAPI
with a TickTrace override that returns the cached trace.
Tests: 15 new tests for ExplainGate (all blockers, bypass conditions,
ordering) and ExplainTick (nothing fires, one fires, tiebreak,
winner/lost_to recording, gate-blocked recording).
When enabled, eligible nudges (sev ≤ ceiling) are queued in memory
instead of sent immediately. Every 'window' duration (or when
'max_items' reached), the queue is flushed as a single digest
notification with concatenated bodies.
Config:
digest:
enabled: true # default false
window: 30m # flush window (default 30m)
max_items: 5 # flush at this count (default 5)
severity_ceiling: 2 # max sev batched (default 2; sev3+ bypass)
Changes:
- config.go: add DigestConfig struct with defaults
- tick.go: QueuedNudge type, digest queue/flush in TickLoop,
shouldQueue/maybeFlush/flushDigest helpers
- main.go: pass cfg.Digest to newTickLoop
- tick_test.go: 6 new tests covering queue, flush, bypass, dedup
Add a local PasskeySession that handleTools checks before processing
any POST action (enable/disable). If the session hasn't been asserted
within the 5-minute TTL, return 403 Forbidden.
Changes:
- webauthn/session.go: add IsStepUp() convenience method (nil-safe)
- webauthn.go: PasskeyHandle holds a *PasskeySession; AssertFinish
calls session.Assert() after IPC step-up
- main.go: create stepUpSession, pass to handleTools and
newPasskeyHandle; handleTools returns 403 if !session.IsStepUp()
- handlers_test.go: update TestEnableTool_NoInProcessAuthGate to
expect 403; add TestEnableTool_WithAuthGate_RequiresStepUp for
the happy path with asserted session; update all 10 call sites
Add cron expression support for recurring reminders using robfig/cron/v3.
Changes:
- Migration #2: ALTER TABLE reminders ADD COLUMN cron TEXT + next_fire_ts INTEGER
- Reminder struct: add Cron and NextFireTs fields
- scanReminder helper extracts full row including nullable cron
- CreateReminder: accept optional cron param, store next_fire_ts = fire_ts
- DueReminders: query on next_fire_ts instead of fire_ts
- RescheduleReminder: new method — parse cron, compute next fire, update
next_fire_ts or mark fired if no more valid times
- Dispatcher: call RescheduleReminder for cron reminders, MarkReminder for
one-shots (preserving existing behavior for ID=0 digest skip)
- ReminderCompleter interface: add RescheduleReminder method
- storeAPI adapter: forward RescheduleReminder
- All callers updated: CreateReminder signature includes cron param
- Tests: TestRecurringReminder (store), TestDispatchRecurringReminderReschedules
- Existing tests updated for new signature
Add a 'scope' TEXT column (default 'homelab') to the tools table so tools
can be namespaced by scope (e.g. "homelab:restart", "datacenter:reboot").
Backward-compat: bare name defaults to "homelab" scope.
Changes:
- Migration #1: ALTER TABLE tools ADD COLUMN scope
- store.Tool: add Scope field, update all SQL and scanTool()
- ipc.Tool DTO and request types: add Scope field
- CoreAPI interface: pass scope in ProposeTool/EnableTool
- storeAPI adapters: forward scope
- cmd/mavend/voice: pass scope (empty → homelab)
- cmd/mavweb/tools: show scope column in UI tables, hidden fields
- All tests updated for scope field
- Migration test made dynamic (startVer = len(migrations))
recordingAPI did not implement the new CoreAPI.RevertFact method
added in a02e10f, causing a build failure in auth tests.
Also adds coverage.out to .gitignore.
Tests cover all Send() code paths:
- Nil TTS synthesizer returns tts-not-wired error
- Nil sessions registry returns sessions-not-wired error
- No live session returns delivery.ErrVoiceNoSession (dispatcher reroutes)
- With session: pushes audio_nudge frame with correct kind, rule_name,
text, PCM16kMono format, non-empty audio bytes
- Empty Body falls back to Summary text
- TTS synthesize error propagates with 'synthesize:' prefix
- Invalid audio format rejects with 'refuse to ship' error
Uses net.Pipe() for real voice.Sessions integration, tts.Stub for
deterministic synthesis, and fake synthesizers for error paths.
- New store.VoidLatestFact() method finds latest non-voided fact for
a key and writes a void-marker row pointing at it (transactional).
- New IPC method MethodRevertFact with CoreAPI.RevertFact interface,
storeAPI adapter, server dispatch, and client proxy.
- New HTTP endpoint POST /api/revert?key=<key> in mavweb.
- History page adds a 'revert' button per non-voided fact row with
JS confirmation and optimistic UI (marks row voided on success).
- All existing store, IPC, and mavweb tests pass.
When the daemon starts after being offline, multiple due reminders
would fire simultaneously as separate notifications. Now a single
digest reminder is dispatched instead, summarizing all pending items.
- New collapseReminders() helper in gather.go: if 2+ reminders are
due, marks originals as fired and returns one synthetic reminder
(ID=0) with a combined JSON payload.
- Dispatcher skips MarkReminder for ID=0 (synthetic digest).
- All loop and delivery tests pass.
The cold-start crash-loop wasn't mavweb-specific — mavpoll and mavcaldav also
ipc.Dial + exit on failure, so they crash-looped until core booted too. Moved
the retry into ipc.DialWait (capped backoff, bounded) and switched mavweb,
mavpoll, mavcaldav to it. mavweb's local dialCoreWithRetry is gone.
Test: server appears after DialWait starts → it waits and connects.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
mavweb (and every ipc.Client) held one net.Conn from Dial and reused it for the
life of the process. When mavend restarted, the socket got a new inode, the
cached conn went dead, and every call failed forever with "broken pipe" — the
dash and page-heartbeat 502'd until mavweb was manually restarted.
Fix in the one place all 25 methods route through (call): on a lost connection
— write failure OR read EOF, since a peer restart can surface on either phase
depending on socket-buffer timing — drop the conn, re-dial the remembered path,
and retry once. Safe for the case that happens (core restarted, request never
processed); the rare committed-then-died window can double-apply a write, but
the store is append-only so a duplicate is a superseding row, not corruption.
ponytail: retry-once, not request-ids — revisit if double-apply ever bites.
Test reproduces the exact incident: server restart on the same socket path, and
asserts the next call transparently reconnects.
Note (not fixed here): Server.Close waits on its handler goroutines, which park
reading live client conns — so a graceful core shutdown with a client attached
blocks until the client disconnects. Minor; surfaces as a slow SIGTERM.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
kami