- strata launch --stop now always runs before the card is read and the start's own port must
refuse connections; a left server that answers, still loads (503) or stays silent was skipped
before and its memory counted as other programs'
- a stop that is not confirmed reads and keeps nothing: the model starts with expert-vram=0
- the stop confirms on the port it is given (stop_server_on), not on one remembered from an
earlier start
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- a measurement whose short answer fails, or whose second card reading fails or shows nothing
more held, is not kept: the model starts with expert-vram=0 and the next start measures again
- measurements are kept per context and requests at once (<context>/<parallel>) and per card
(its NVIDIA UUID), not by the card's size
- a server on this app's port left from a run that ended badly is stopped before the card is
read, so its memory is not counted as other programs'
- the game placement falls back to 0 GiB of experts before pausing when 2 GiB leaves a game
nothing
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- with an NVIDIA card the app sets --expert-vram itself by the M0 kit's formula: what the card
holds before the start, the model without experts, the STT worker's share (2740 MiB), the
engine's margin (768 MiB) and room for programs opened later (game.otherRoomGb, default 1 GB)
are left free; Strata's own placement filled the card and left the STT no room
- the model without experts is measured once per context and card (a start with
expert-vram=0, one short answer, the card's memory and the engine log's staging), kept in
config/llm-probe.json; the real start replaces that server
- placement.rs also works out the game placement (the reserve, else the 2 GiB minimum and the
room a game really gets, else pause) for the switching that comes next
- with the game PC's M0 numbers the normal placement comes out at 1.5 GiB, as M0 measured
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Battle.net installs anywhere: its games come from the machine's uninstall entries (removed by
its Blizzard Uninstaller or published by Blizzard Entertainment, with an install folder, not
Battle.net itself), read when the watch starts; a program inside one is a game named after
the entry
- launcher folders are matched part by part with ASCII case ignored; the name was cut by byte
positions of a lower-cased copy, which panicked the detection thread on a path such as
D:\İ\steamapps\common\하데스\game.exe
- checked on the win-ci VM: a Steam game under a non-ASCII path and a Battle.net game (test
uninstall entry) turn game mode on with their names and off when they end
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- while the services run the app looks at the foreground window every 2 s (Windows): the
person's notGames and games lists first, never a browser, video player, the desktop, Discord,
a launcher's own window or this app; a program under a Steam, Epic, Riot or Battle.net game
folder, exclusive full screen, or a window covering a monitor without title bar or border whose
process uses at least 10% of the GPU's 3D engine (the M0 rule) is a game
- game mode stays until that game's process ends, also behind other windows; the dashboard shows
"게임 모드: 켜짐 (name, reason)" and the game log tab its changes
- a 게임 모드 settings screen (detection on/off, the two lists)
- checked on the win-ci VM: a program in a Steam game folder turned it on after 1.4 s and off
2.0 s after it ended
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- a failed install forgets the driver (registry value, settings folder) only when its clean-up
left no device and no package, and never when it was ours before; otherwise it stays ours and
the error gives both the install's reason and that something is left to remove
- the driver status separates a usable monitor (installed) from what this app left without one
(leftover) and names the button (action): uninstall for both, so a removal whose package step
failed can be retried from the features screen
- pure rules after_failed_install and driver_status with unit tests; checked again on the win-ci
VM: a declined install leaves nothing, then install and uninstall
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- installing or removing the driver is refused unless the services are stopped (or failed), and
a service start or restart (and a second driver task) is refused while one runs, so the
monitor never vanishes under the bot browser's window
- a failed or declined install undoes what that attempt made (device, and the registry value and
settings folder when the driver was not ours before); a declined install's
0xE0000243 reads as cancelled; installed means ours with its monitor there
- devices are found among all root devices whatever their class: a device whose driver was
declined drops out of the display class and was left behind
- uninstall keeps the registry value (what makes it ours) until the device and the driver
package are gone (a package not in the store, 1168, counts as gone), so a failed removal can
be retried
- checked on the win-ci VM: a declined install leaves nothing, the next install gives one working
device, the uninstall leaves nothing
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- features.browserShare (봇용 크롬창 화면공유 선택 표시, off by default) and a driver card on the
features screen: the Virtual Display Driver 25.7.23 (MIT, SignPath-signed) ships in
display-driver/ (hash-checked by package.sh); install and uninstall run jarvis.exe elevated
once (--display-driver), settings in %ProgramData%\JarvisBot\display-driver, another
program's copy of the driver is left alone
- the monitor is on the desktop only while the bot browser uses it, at the rightmost real
monitor's bottom-right corner; the stop closes Chrome first, then takes the monitor off
- Chrome starts with no window and the app opens it over CDP in the background, so it never
takes the focus; the window is a tool window (no Alt+Tab, no taskbar button), fills the
monitor's work area and keeps drawing when covered; without a monitor it runs headless with
the reason on the dashboard
- Chrome runs with --do-not-de-elevate on Windows: started by an app running as administrator
it restarted itself as the user and the app lost hold of it
- checked on the win-ci Windows VM from install to uninstall (focus kept while typing, no
taskbar button, no Chrome left, nothing left after uninstall) and on Linux with a real Chrome
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- each asker has a tab and connection of their own (at most 5, least recently used closes first)
- links, forms and window.open that would open a new tab load in the same tab
- one 15 s deadline covers connecting, the action and the page; events do not extend it
- a broken connection retries only open and read; click, type and back may have run and are not repeated
- <<< and >>> in page content are replaced so it cannot close the fence
- the description says how to search on a site (open, then type with submit=true); eval case 0/3 -> 3/3 on qwen3:8b
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
open, read, click (label or [n]), type (optionally Enter) and back; page content comes back fenced
as untrusted data; only http(s) addresses; no LLM call. Offered only when the app started the bot
browser (JARVIS_BROWSER_CDP) and the browser feature is on; meanwhile fetchWebPage, which has no
address guard, is not offered. Eval evals/test_bot_browser_tool.py.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- IPv6 counts as public only in 2000::/3 (without 2001::/23, documentation and 6to4), so site
local fec0::/10 and IPv4-compatible addresses are refused; the collector uses the same rule
- every proxy connection is a task of one set: a stop ends the open CONNECT tunnels and waits
until their sockets are gone
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- install step browser: Chrome for Testing 157 (pinned SHA-256) into runtime/chrome
- services start it headless with its own profile when features.browser is on; CDP on a free
loopback port passed to the worker as JARVIS_BROWSER_CDP; a failure leaves the services running
- every Chrome connection goes through the app's loopback egress proxy: one lookup, public
addresses only (403 otherwise, no host logged), QUIC off, WebRTC only through the proxy
- stop: Browser.close, then kill after 5 s; dashboard shows the browser's state
- the worker's runtime environment follows the settings step, not every step being installed
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The request line is ASCII: other characters and spaces in the path and query go as UTF-8 %HH,
while existing %HH and separators stay.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The name lookup runs on a daemon thread waited for only until the deadline, and a watchdog
shuts the connection down at the deadline, so a server that sends its status line or headers
byte by byte within the per-read limit is cut off too. A response cut short is a failure.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- the fetcher resolves once, requires every address to be public and connects only to those
(TLS still checks the name), so DNS rebinding cannot reach the local network; redirects too
- 15 s for the whole fetch also against servers that send slowly; no URL in any log line
- a site whose title alone changed is kept again
- target names that differ only in case are refused (one Windows folder)
- the worker always gets the collection file, so results older than the keep days go also
with no targets or the feature off
- the results screen lists every target with results, not only those in the newest 100
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Targets (none by default), interval and keep days on a new 백그라운드 수집 screen with the
results; the worker collects on its own thread through the web search's address check and
writes only to the data folder's collect/ (logs carry target names and counts).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Covers the admin paths (`mcp_env_set`, `settings_apply` removing a server): after the services
started with the old values, a late worker line carrying them is masked on the log screen and in
the error report. Fails with the in-use values dropped from the masks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each worker run's log sink keeps the values masked when it started (its secrets in use), so a
line still being logged after a stop gave up waiting, and after the next start replaced the values
in use, is still masked. The debug echo on stderr is masked the same way.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stopping (and restarting) the voice worker now waits, at most 5 s, for its output readers to log
what the exited worker wrote, so its last lines are masked with the secrets that run was started
with before the next start replaces them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The log hub masks the stored secrets and the values the services were started with (bot token,
MCP environment values), so a late worker or MCP line carrying a key replaced or removed while the
services run stays masked on the screen, in copies, files and exports; the next start replaces
the in-use values.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Discovery runs outside the MCP tool cache's lock (one at a time), so a turn reading the cached
tools never waits for a slow server; a new conversation refreshes in the background.
- The log hub masks secrets before a line is kept or written, so a key replaced or removed later
stays hidden on the screen, in copies, files and exports, also after a restart.
- The worker takes JARVIS_MCP_ENV out of its environment at start, before MeloTTS runs, and hands
it to the brain.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Settings `features` (weather, webSearch, meals) and `mcpServers` (name, command, args, env names,
enabled), edited on the new 기능 screen. The tools of a feature that is off go into the brain's
`disabled_tools`; only enabled MCP servers go into its `mcps`.
- MCP environment values (API keys) live only in the encrypted secrets (`mcp:<server>:<NAME>`),
shown as stored or not, masked in logs, removed with their server, and reach the worker only in
`JARVIS_MCP_ENV`, which the brain takes out of its environment before giving each server its own.
- `enabled_builtin_tools(cfg)` is the one list the router, toolSearchTool and the executor use, so
a turned-off tool is neither offered nor run, also when the model names it.
- The v2 brain never offers v1's `localFiles` (home folder read/write/delete) and `screenshot`:
anyone in the voice or chat channels could ask for them. Screen viewing comes in M7.
- The brain connects MCP servers on its own thread and logs each outcome without debug logging.
- Stored secrets are masked in the log hub as soon as the data folder is found.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A rate-limited edit of a newly registered channel is retried with the
same "never clear Discord with an empty field" rule (it was retried as
an existing channel and cleared the topic).
- Channel edits go one at a time, and Discord's form of the name is
written back only while the settings still hold the name that was sent,
so a slow answer to an earlier save never overwrites a later one.
- The chat channel screen drops a Discord name/topic lookup that answers
after the ID was changed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Review follow-ups to M5.
- Chat channel names and topics (app/src/channel_sync.rs): a channel with
syncDiscord (default on) gets its name and description as its Discord
name and topic when saved; registering fills them from Discord;
Discord's form of the name is written back; a topic over 1,024
characters is refused before saving; a rate-limited edit is pending and
retried once by itself with the latest text; a missing permission fails
with the reason and the app keeps the text. Commands
discord_channel_info and channel_sync_status. Checked on real Discord
with a temporary channel (fill, edit, rate limit, automatic retry).
- voice.tts.speed (0.5-2.0) on the model screen, passed to the worker as
--melo-speed. MeloTTS's Korean model has one voice, so no voice choice.
- Closing the window runs AdminApi::shutdown first: an install in
progress is cancelled and waited for, the services stop, a failed run's
leftovers (Strata) are cleared (Controller::shutdown). If that fails the
window stays open with the reason; closing again quits.
- WebView2 data goes in the data folder (cache/webview), or a temp scratch
folder before one is chosen, removed at the next start: %APPDATA%\JarvisBot
holds only the location file.
- Command output that is not UTF-8 (CP949 on a PC without the UTF-8
option) is decoded in the OEM code page instead of ending the reading.
Verified on the win-ci VM: closing during the install leaves no process,
no --location keeps only location.json under %APPDATA%, the unit tests on
Windows 168/168.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Found running the app's admin commands on the game PC (Windows, RTX 5070
Ti): the full install, Strata started by the app downloading the 22.9 GB
model, the worker and Discord up, five spoken questions all answered.
- Windows refuses a connection to a closed local port only after about
2 s, so the 2 s request timeout made every stop "not confirmed". The
check now waits up to 10 s to connect and still counts only a refusal
(not a timeout) as stopped.
- Strata 0.1.59 reports the model download as "X of Y GB"; percent()
reads that, and the dashboard shows the current line's progress only
(it had stayed on the engine download's 80 %).
- Strata and the NVIDIA shader cache write under %APPDATA% whatever
STRATA_* says: the layout sets APPDATA inside the data folder on
Windows, next to LOCALAPPDATA.
Roadmap: M5 step 5 done.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
M5 step 5, first part. app/shell is a Tauri 2 crate (jarvis.exe): one
window over app/ui whose invoke("admin") goes to AdminApi::call. Its
WebView2 data sits next to the location file; closing the window stops
the services and waits for a cancelled start (AdminApi::shutdown).
app/shell/package.sh cross-builds it on Linux (x86_64-pc-windows-gnu)
and lays out Jarvis/ with WebView2Loader.dll and python-src/
(voice_worker, jarvis). The worker's default sources are python-src next
to the executable when present, else the repository's src/.
The bot screen now shows the masked token right after saving.
Checked on the win-ci VM: the window and wizard, PC check, uv and worker
venv installed, a cancel leaving no uv/Python process, the token stored
DPAPI-encrypted, a start failing with its reason, logs, error copy, zip
export, closing the window ending the process, nothing under %APPDATA% or
%LOCALAPPDATA%; the core's unit tests on Windows (156/157, the one
Linux-path assumption fixed).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A stop right after a background start set the LLM's cancel flag, but the
start task then cleared it in Llm::start and went on to download and load
the model. The cancel is now armed in the controller's begin_start, under
its lock (Host::prepare_start -> Llm::arm), and Llm::start never clears
it; a start that finds it set returns before running anything.
wait_gone counted a timeout, a 503 (a server still loading) or an
unreadable answer as stopped, so a failed `launch --stop` could still mark
a live engine off. Only a refused connection confirms the stop now; the
rest keeps checking until the deadline and then reports failed.
Tests: a stop right after a background start runs no launch (and the next
start is not cancelled); a failed stop with a silent or 503 server leaves
the LLM and the services failed through the controller. Fixed-port tests
use an unused port.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
service_start now moves the controller to starting before it returns
(Controller::start_in_background), so a stop right after it cancels the
start and a second start is refused instead of the start running late.
service_restart stops, then starts the same way.
Stopping Strata now confirms the server is gone: `launch --stop` is ended
after 2 minutes with everything it started, then the port must stop
listing a model. A server still answering makes the LLM failed with the
reason, and Services::stop / Running::stop return it, so the controller
shows failed instead of stopped (also for a cancelled start that could
not clear the server). A stop command that fails while nothing answers
counts as stopped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
M5 step 4, second part. The services start the LLM first: the installed
Strata with the M0 kit's model and normal placement (models in
models/llm, parallel 1, or 2 when a chat channel answers every message,
pictures off, --restart), wait until /v1/models lists the model, and stop
it last. The first start downloads the model; the dashboard shows
Strata's last line and percentage, and stop also cancels a start (the
server is cleared). llm.url uses a server the app does not start.
On every start the app writes config/brain.json (Strata's /v1, the
listed model id, memory/jarvis.db) and hands it to the worker as
JARVIS_CONFIG_PATH. Loading a config with its own db_path no longer
creates ~/.local/share/jarvis.
service_start/service_restart run in the background. Settings gain
llm.context and llm.port (model screen) and llm.url (file only).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A cancelled command ends with every process it started (process group on
Unix, taskkill /T on Windows).
- Waiting for a response or the next chunk gives way to a cancel, so a
silent server cannot hold the install.
- Pointing the settings at the runtime is the last install step: a failure
shows as failed and can be retried; done means written.
- Strata's installer gets its folder as the working folder, not inside the
script text.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- download: resumed downloads checked against a pinned SHA-256.
- install: uv 0.11.20, the worker/brain venv (Python 3.12, the new
src/voice_worker/requirements-brain.txt), the MeloTTS venv (Python 3.11,
PyTorch cu128), Whisper large-v3-turbo, the Korean MeloTTS voice and
Strata, the M0 kit's recipe; markers per step and version, cancel, retry,
everything (caches, temp, HF and NLTK data) inside the data folder, the uv
cache cleaned and the worker settings pointed at the venvs when done.
- The services give the worker the same environment once installed.
- The wizard's model step shows and drives the install.
Installed from an empty folder on Linux and started the services on it:
worker connected, STT and TTS ready, nothing written outside the folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- app_status reports firstRun until wizard_finish; wizard_step saves the
step reached in config/setup.json, and the wizard resumes there (a stored
token lets the bot step go on).
- show() and refresh() drop answers that arrive after the screen changed.
- folder_choose points the log hub at the folder's logs/ at once.
- Choosing another server clears the voice channel picker.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- admin_api: the one entry point of the screens (app status, data folder,
PC check, Discord token check and pickers, settings, service control and
the dashboard, logs query, error report and zip export).
- discord_setup: token, Message Content and invite link checks over REST.
- app/ui: first-run wizard, dashboard, five settings screens and the log
screen in plain HTML/CSS/JS (dark theme after DESIGN.md, Korean text).
- jarvis-admin-dev: a loopback, token-guarded development server so the
screens run in a browser against the real commands.
Checked in a browser with the real test bot and voice worker: wizard
steps, start/restart/stop, settings save, chat channels, logs and export.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each Joined carries a new session number; the core records, when an
utterance's speech_start arrives, the session Jarvis is in, and its Partial
and Final carry that number (Heard.session) even when they arrive after a
leave or a move. The dashboard shows only the current session's captions.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Each start gets a new dashboard; stop ends the event feeding and waits for
it before forgetting, so queued or earlier-run events never reach the
screen.
- AnswerAsked carries the request's speech end and final times, so a second
waiting question does not change the first answer's timing.
- Captions are shown only while Jarvis is in a voice channel.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- control: one Controller starts, stops and restarts the services (stopped,
starting, running, stopping, failed); a failed start leaves nothing
running and keeps its error with secrets masked. The app binary uses it.
- dashboard: built from the core's events alone: the bot's tag, the voice
channel, live captions (screen only, cleared on leaving or stopping), the
last 20 voice answers split into stages with medians (matched by the new
AnswerAsked event), and missed requests without text.
- system: memory in use and per NVIDIA GPU its use and VRAM (nvidia-smi).
- DiscordCore::voice_members, WorkerClient::hello, the conversation's missed
requests shared with the dashboard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- redact: a token with sentence dots around it is masked too.
- Worker error messages are always logged (code and id), so the log screen
and the error report show them without debug logging.
- The voice worker logs with stt, llm and tts categories, so its lines land
in those tabs.
- The zip leaves out the oldest lines by the time they were logged, across
every tab's file, not by file name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- datadir: the data folder layout, the location file (JARVIS_DATA_DIR wins),
free space for the wizard.
- settings_edit: the fields the five settings screens edit; apply writes only
non-defaults, keeps unknown keys and writes nothing unless the app's own
check passes. A memory section reaches the worker's hello.
- secrets: config/secrets.dat, DPAPI for the Windows user; the services take
the bot token from it before JARVIS_DISCORD_TOKEN.
- logs: the log hub behind the log screen (tabs, daily files kept 14 days,
the recent error report, a zip under 10 MB, secrets and token-shaped text
masked); the log macros and the worker's output feed it.
Spec: app/src/admin.spec.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every <parameter> value was passed as text, so a delete confirmed with
confirm=true reached chatChannels as "true" and asked again. Values now take
the type the tool's input schema declares (boolean, integer, number, array,
object); a value that does not fit stays text. The roadmap's eval summary now
names the reply-length failure apart from the later passing runs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A chatChannels pick from the router counts as a tool turn for the intent
judge ("post this in the schedule channel" needs the planner).
- The default-language directive names an explicit language request as its
one exception; the history block says plans and appointments people
mentioned need no calendar tool; why/how answers give only the core reason.
Evals (Qwen3.6, v2 options): reply length 3/3 (v2 185-205 chars, v1 315-356),
language, history, chat channels and tool-skip 22/22; full suite 219 passed,
6 failed (paris/london, graph branch and four v1 single-user memory cases).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A partial transcript that call detection already reads as a request is sent
to the worker as prepare, so the intent judge runs while the speaker talks.
The live round trip now times the earcon and the answer's own sound apart.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- compact_prompt: the v2 persona in a tenth of v1's words, answers as short as
possible, the full tool guidance only on turns with a data tool. A greeting's
reply prompt went from about 4,500 to 1,400 tokens on Strata.
- intent_judge: the tool router judges whether the turn needs tools; the
planner runs only when it does. prepare_intent runs the pick ahead on a
partial transcript (worker message prepare), and the reply reuses it.
- Without stream_browser the screen-share tools are not offered.
- The time context carries the 12-hour form (10:37 was read as pm).
- A <tool_call> block written as the reply is parsed as a call, not spoken.
Evals (Qwen3.6 on Strata, EVAL_COMPACT_PROMPT=1 EVAL_INTENT_JUDGE=1):
221 passed, 4 failed (known failures); new evals/test_voice_reply_length.py
(v2 answers 234 chars against v1's 416) and test_reply_language_default.py.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The brain passes default_language (worker --reply-language, default Korean) to
the reply engine; unlike the output_language lock it gives way when the user
explicitly asks for a language. Tools get the ko code (--tool-language).
Eval evals/test_reply_language_default.py (Qwen3.6 on Strata): 6/6, English
questions answered in Korean, an explicit English request in English, and the
control without the default answers English in English.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A short two-tone chime, made in code, plays right before each voice answer
starts and at once for a wake word said alone while Jarvis is free; the
question that joins it gets no second one. call.earcon turns it off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A wake word said alone waited without a time limit once the speaker's
next utterance started. If the worker connection ended meanwhile, that
final never came and the restarted worker's utterances (numbered from 1
again) were joined to it, so a later sentence without a wake word was
answered. The core now reports WorkerLost; the conversation drops
waiting wake words and forgets utterances in progress then and on leave.
An utterance in progress is tracked until its own final arrives instead
of by comparing utterance numbers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
live_round_trip.rs runs the app's services with the real worker and brain,
plays recorded questions from a second bot and times the final, the
answer text, playback start and the first audible frame after the speech
ends; a second worker session streams nine more voices for the load run.
Roadmap: first M4 step 4 numbers (median 3.5 s alone, 4.4 s under load).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A pause after the name ("자비스, ... 지금 몇 시야?") is longer than the
worker's 300 ms end silence, so the wake word ended its own utterance:
Jarvis answered the bare name and ignored the question that followed (no
wake word). Found on real Discord. A voice utterance that is only a wake
word is now joined to the same speaker's next final when that utterance
is running or starts within call.followUpMs (default 1500); otherwise the
wake word alone is the request after that time. Stops drop it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>