Latency: replies are spoken sentence-by-sentence off the desk API's NDJSON
endpoint, so the wait is time-to-first-sentence rather than the whole model
call, and Deepgram's live websocket transcribes while you're still talking
instead of uploading the WAV afterwards. Both fall back invisibly — a stream
that fails before anything was said drops to converse(), and a socket that
never opens just means the old one-shot path.
Speaking lived in four near-copies in the controller (a reply, a holding line,
a streamed sentence, a dialogue scene) that had already drifted: one didn't arm
barge-in, another skipped the follow-up rule. It's now speech.Speaker plus an
Utterance describing the policy differences, with collaborators injected so the
whole of it tests without Qt or audio.
The mouth follows the audio rather than a timer: tts.level_of reduces each PCM
frame to a 0..1 loudness on a sqrt curve (speech sits well below peak, and a
linear map leaves the mouth barely open during normal talking) and that indexes
the talking frames, which the sprite script now draws as an openness ramp.
Offline pyttsx3 has no waveform, so stale levels hand control back to the timed
loop instead of freezing the mouth mid-syllable.
Also: the pet starts where you left it (ignoring positions on monitors that are
no longer connected, since restoring those faithfully is how it ends up
somewhere unreachable), and `python -m bolt_pet --doctor` is a preflight that
says what to do about each problem rather than only what's wrong.
tests/test_pipeline_smoke.py breaks the pure-logic rule on purpose. Every unit
test passed all week while notifications sat unspoken for minutes, the pet said
things twice and [laughing] got read aloud — each an interaction between two
individually-correct units. It drives whole turns against a real HTTP server on
a loopback port, faking only the mic and the speakers. It found a NameError in
the paint path that would have fired on every repaint while talking.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Multi-monitor jumps (`petctl jump`/`monitors`), pull-only screen OCR
(`petctl read`), generated sprite art with a distance-stepped walk cycle,
plus the filectl file ops and server file delivery merged back in.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
petctl gains screen verbs: `jump` (1-based number, name, next/prev/
primary/other, or a direction resolved from real geometry), `monitors`,
and `read` for OCR of a monitor's contents.
- monitors.py: pure layout model + jump-target resolution. The monitor
list is published by PetWindow from QGuiApplication.screens() over a
queued signal, so the controller and window agree on what "monitor 2"
means; xrandr and Qt order screens differently on the same machine.
- screen_text.py: pull-only OCR (mss capture + Tesseract/RapidOCR).
Nothing captures unless the server asks, and the text rides back up
the tool-result relay so Bolt can read a screen mid-turn. Both deps
optional, soft-failing with a reason. SCREEN_TEXT=false removes it.
- Query verbs are answered in controller._handle_command rather than
pet_actions.describe(), because their output is the point.
- scripts/generate_bolt_sprites.py draws every frame; walk/ is a
side-view cycle stepped by distance travelled, not by the animation
timer, so the planted paw tracks the window exactly. sprite.py loads
it via EXTRA_ANIMATIONS keyed by name, with has() so callers can
decline a placeholder blob.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
themajesticmagician