Ninja-Steel Stage Triptychs, Amiga Half-Brite Portraits, and the Three-Voice Piper TTS Cast

# Context 1
/home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

  THE STADIUM STREAMER (HEAD)            THE HONEYBOT 24/7/365 STREAM (LONG TAIL)
  .---------------------------------.    .---------------------------------------.
  | Streams 2 hrs/week (104 hrs/yr) |    | Streams 24/7/365 (8,760 hrs/yr)       |
  | Chases algorithm dopamine spikes|    | 1-2 concurrent humans at any moment   |
  | Dead air 98.8% of the year      |    | = ~150-300 unique engineers/day       |
  |                                 |    | = ~50,000-100,000 deep technical      |
  |                                 |    |   sessions/yr watching live wire      |
  |                                 |    |   telemetry + testing RFC 9110!       |
  '---------------------------------'    '---------------------------------------'

  THE ASIDE-PUBLISH SAFETY NET (BUILT IN DEEDS 562-568, WITNESSED TODAY)
  ==========================================================================
  1. You ran `article --aside` -> stamped `storytime: false` in YAML frontmatter
     (verified at line 36 of 2026-11-09-ugly-curly-cues-...-piper-tts.md).
  2. You ran `publish` -> git push hit Honeybot's `post-receive` hook:
       - `git checkout -q -f main` updated the Jekyll repo in place.
       - Touched `.deploy_standby` to notify `stream.py`.
  3. `stream.py` woke up and called `content_loader.get_featured_article_id()`.
  4. Because the new post has `storytime: false`, the featured Storytime
     article ID DID NOT CHANGE!
  5. Instead of killing the show or restarting from paragraph 1, `stream.py`
     kept the active article's paragraph cursor intact and resumed right
     where your 2 live viewers were listening!

  BABY-STEP WIN #1 (NEXT TURN): THE 3-PANEL MIDDLE BAND IN LOGS.PY
  ================================================================================
  +----------------+--------------------------------------------+----------------+
  | #left_stage    | #ticker_bar (1fr center marquee)           | #right_stage   |
  | (14 cols x 5)  | (Auto-centers 5-row LED cards in middle)   | (14 cols x 5)  |
  +----------------+--------------------------------------------+----------------+
  |                |   ███  ███ █████  ██   █████  ███  ██      |                |
  |     Mike-E     |   ██ █ ███   █   ██    ██     ██   ██      |    Shipmunk    |
  |      the       |   ██  █ ██   █   ██    ████    ██ ██       |       O'       |
  |   Apprentice   |   ██    ██   █   ██ ██ ██   ██  ███   ██   |    Theseus!    |
  |                |   ██    ██ █████ █████ ████ ██   █    ██   |                |
  +----------------+--------------------------------------------+----------------+

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
context.txt
remotes/honeybot/scripts/logs.py
! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

Verify the LIVE COMMAND RECEIPTS for `rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py` and inspect `remotes/honeybot/scripts/logs.py` in raw source. Execute Baby-Step Win #1 of our theatrical build: partition the 5-row middle band in `logs.py` into a 3-panel `Horizontal` triptych (`#left_stage` displaying the 5-row `Mike-E / the / Apprentice` curtain label, `#ticker_bar` auto-centering the 5-row LED marquee in the remaining center columns, and `#right_stage` displaying the 5-row `Shipmunk / O' / Theseus!` curtain label) with a straddle probe that verifies all three widgets and curtain labels before we even touch the live stream.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
context.txt
remotes/honeybot/scripts/logs.py
! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

  THE 400-YEAR LINEAGE OF THE HONEYBOT 3-PANEL STAGE
  ==============================================================================
  1. SHAKESPEARE & BRECHT (1599 / 1928)
     - Elizabethan Prologue / Chorus ("Henry V") + Brechtian Stage Placards:
       "Don't hide the stage mechanics! Hang a wooden sign naming the character,
        have the Chorus announce the turn, and let the audience's mind fill in
        the rest."
                                      |
                                      v
  2. THE SNES / PS1 HUD COMMS STRIP (1993 - 1998)
     - DOOM Status Bar Face (1993): 5-row bottom HUD portrait reacting to state.
     - Star Fox (SNES, 1993) & Star Fox 64 (1997): Fox, Peppy, Falco & Slippy
       pop up in tiny pixel boxes with pitch-shifted voice bleeps over live 3D!
     - Metal Gear Solid Codec (PS1, 1998): Left portrait + Center waveform +
       Right portrait!
                                      |
                                      v
  3. THE NINTENDO DS DUAL-SCREEN GOLDEN AGE (2004 - 2010)
     - Why the DS perfected it: Screen 1 held the live city/map telemetry,
       forcing dialogue into a compact 3-part banner (Left Portrait | Text |
       Right Portrait) where the active speaker lights up at 100% brightness
       and the listener dims to 40% shadow!
     - Iconic DS examples:
       * Grand Theft Auto: Chinatown Wars (2009) -- between missions driving
         around Liberty City, 2D portraits pop up left/right of the street strip!
       * Advance Wars: Dual Strike (2005) & Fire Emblem: Shadow Dragon (2008)
       * Castlevania: Portrait of Ruin (2006) -- Jonathan & Charlotte dual boxes!
       * The World Ends With You (2007) & Chrono Trigger DS (2008)
       * Phoenix Wright: Ace Attorney & Animal Crossing ("Animalese" pitch!)
                                      |
                                      v
  4. HONEYBOT STORYTIME (2026)
     - Top Panel: Live Nginx Wire Stream | Bottom Panel: Dual Bot Intelligence
     - Middle 5-Row Band:
       [Mike-E the Apprentice] | [5-Row LED Marquee] | [Shipmunk O'Theseus!]
       + Stage Narrator Norma Centroid calling the curtain!

  +----------------+--------------------------------------------+----------------+
  | #left_stage    | #ticker_bar (width: 1fr)                   | #right_stage   |
  | (14 cols x 5)  |                                            | (14 cols x 5)  |
  +----------------+--------------------------------------------+----------------+
  |                |   ███  ███ █████  ██   █████  ███  ██      |                |
  |     Mike-E     |   ██ █ ███   █   ██    ██     ██   ██      |    Shipmunk    |
  |      the       |   ██  █ ██   █   ██    ████    ██ ██       |       O'       |
  |   Apprentice   |   ██    ██   █   ██ ██ ██   ██  ███   ██   |    Theseus!    |
  |                |   ██    ██ █████ █████ ████ ██   █    ██   |                |
  +----------------+--------------------------------------------+----------------+

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
context.txt
remotes/honeybot/scripts/logs.py
! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
context.txt
remotes/honeybot/scripts/logs.py
! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
DEFAULT_TICKER_MESSAGE = (
    "JUST GIVE 30 SECS... | "
    "HONEYBOT STORYTIME | "
    "JOIN DISCORD 4-SEO! | "
    "1ST 1000 -> AI MEMORY | "
    "VISIT SUNPUG.IO | "
    "CHAT SUMMONS MIKE | "
    "AI'S @ MIKELEV.IN"
)
[[[DIVIDER]]]
DEFAULT_TICKER_MESSAGE = (
    "JUST GIVE 30 SECS... | "
    "HONEYBOT STORYTIME | "
    "JOIN DISCORD 4-SEO! | "
    "1ST 1000 -> AI MEMORY | "
    "VISIT SUNPUG.IO | "
    "CHAT SUMMONS MIKE | "
    "AI'S @ MIKELEV.IN"
)
STAGE_BOX_WIDTH = 14
STAGE_LEFT_CURTAIN_LINES = ("", "Mike-E", "the", "Apprentice", "")
STAGE_RIGHT_CURTAIN_LINES = ("", "Shipmunk", "O'", "Theseus!", "")

def render_stage_curtain(side: str, width: int = STAGE_BOX_WIDTH) -> Text:
    """Render the 5-row x 14-col theatrical curtain label for #left_stage or #right_stage."""
    is_left = (side or "").strip().lower() != "right"
    lines = STAGE_LEFT_CURTAIN_LINES if is_left else STAGE_RIGHT_CURTAIN_LINES
    primary_style = "bold #ffd700 on #0d061a" if is_left else "bold #00ffff on #0d061a"
    mid_style = "italic #ffae42 on #0d061a" if is_left else "bold #ff79c6 on #0d061a"
    txt = Text()
    for idx, raw_line in enumerate(lines):
        if idx > 0:
            txt.append("\n")
        padded = raw_line.center(width)[:width]
        row_style = mid_style if idx == 2 else primary_style
        txt.append(padded, style=row_style)
    return txt
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    /* MIDDLE SECTION: Eye-Level Giant 5-Row LED Marquee Ticker (borderless, 100% signage) */
    #ticker_bar {
        height: 5;
        width: 100%;
        background: #140826;
        color: #ffd700;
        text-style: bold;
        padding: 0 1;
        overflow: hidden;
    }
[[[DIVIDER]]]
    /* MIDDLE SECTION: 5-Row Theatrical Triptych (#left_stage + #ticker_bar + #right_stage) */
    #middle_stage {
        height: 5;
        width: 100%;
        background: #140826;
        layout: horizontal;
        overflow: hidden;
    }

    .stage_box {
        width: 14;
        height: 5;
        background: #0d061a;
        color: #ffd700;
        text-style: bold;
        padding: 0;
        overflow: hidden;
    }

    #left_stage {
        color: #ffd700;
    }

    #right_stage {
        color: #00ffff;
    }

    #ticker_bar {
        height: 5;
        width: 1fr;
        background: #140826;
        color: #ffd700;
        text-style: bold;
        padding: 0 1;
        overflow: hidden;
    }
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def compose(self) -> ComposeResult:
        yield Header()
        # RichLog handles wrapping and styling natively. 
        # wrap=True ensures lines don't get cut off.
        yield RichLog(id="log_stream", highlight=False, wrap=True)
        yield Static("", id="ticker_bar")
[[[DIVIDER]]]
    def compose(self) -> ComposeResult:
        yield Header()
        # RichLog handles wrapping and styling natively. 
        # wrap=True ensures lines don't get cut off.
        yield RichLog(id="log_stream", highlight=False, wrap=True)
        with Horizontal(id="middle_stage"):
            yield Static(render_stage_curtain("left"), id="left_stage", classes="stage_box")
            yield Static("", id="ticker_bar")
            yield Static(render_stage_curtain("right"), id="right_stage", classes="stage_box")
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def _get_ticker_width(self) -> int:
        """Return usable inner column width of #ticker_bar, falling back to OS terminal size before layout."""
        try:
            bar = self.query_one("#ticker_bar", Static)
            if bar.size.width and bar.size.width > 0:
                return max(40, bar.size.width - 2)
        except Exception:
            pass
        try:
            if self.size.width and self.size.width > 80:
                return max(40, self.size.width - 2)
        except Exception:
            pass
        try:
            term_cols = shutil.get_terminal_size((190, 48)).columns
            return max(40, term_cols - 2)
        except Exception:
            return 188
[[[DIVIDER]]]
    def _get_ticker_width(self) -> int:
        """Return usable inner column width of #ticker_bar, falling back to OS terminal size minus stage boxes before layout."""
        stage_deduction = (2 * STAGE_BOX_WIDTH) + 2
        try:
            bar = self.query_one("#ticker_bar", Static)
            if bar.size.width and bar.size.width > 0:
                return max(40, bar.size.width - 2)
        except Exception:
            pass
        try:
            if self.size.width and self.size.width > 80:
                return max(40, self.size.width - stage_deduction)
        except Exception:
            pass
        try:
            term_cols = shutil.get_terminal_size((190, 48)).columns
            return max(40, term_cols - stage_deduction)
        except Exception:
            return 188 - (2 * STAGE_BOX_WIDTH)
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS for Baby-Step Win #1 (`TRIPTYCH_DOM=True LEFT_5X14=True RIGHT_5X14=True`), confirming that the 5-row middle band in `remotes/honeybot/scripts/logs.py` is now partitioned into `#left_stage` (`Mike-E / the / Apprentice`), `#ticker_bar`, and `#right_stage` (`Shipmunk / O' / Theseus!`). Now execute Baby-Step Win #2: pre-bake the 5-row x 14-column 24-bit Truecolor Unicode half-block (`▀`/`▄`/`█`) character portraits for **Mike-E the Apprentice** and **Shipmunk O'Theseus!** in all three states (`curtain`, `active` 100% full-bright RGB, and `inactive` 40% dimmed half-bright RGB) with a straddle probe that verifies all six pre-baked `Rich.Text` frames and prints an ANSI/ASCII preview.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
context.txt
remotes/honeybot/scripts/logs.py
! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

  HOW 2 PIXEL ROWS PACK INTO 1 TERMINAL ROW (ZERO-GAP TRUECOLOR)
  ============================================================================
  Top pixel (y*2)   = #ffd700 (Gold)  \
                                       ===>  Text("▀", style="#ffd700 on #1e40af")
  Bot pixel (y*2+1) = #1e40af (Blue)  /

  Top pixel (y*2)   = . (Stage BG)    \
                                       ===>  Text("▄", style="#dc2626 on #0d061a")
  Bot pixel (y*2+1) = #dc2626 (Red)   /

  Top == Bot        = #ffcc99 (Skin)  ===>  Text("█", style="#ffcc99 on #0d061a")

  STAGE LEFT: MIKE-E THE APPRENTICE          STAGE RIGHT: SHIPMUNK O'THESEUS!
  (Sorcerer's Hat + Gold Star + Red Robe)    (Perky Ears + Stripes + Buck Teeth)
  +--------------+                           +--------------+
  |     ▄██▄     | <- Blue cone + gold star  | ▀██▀ ▄▄ ▀██▀ | <- Perky ears + pink
  |   ▄██████▄   | <- Golden wizard brim     |  ██████████  | <- Striped brow + eyes
  |  ██████████  | <- Round ears + bright eye| ████████████ | <- Chubby cheeks + nose
  |  ▀ ██████ ▀  | <- Rosy cheeks + smile    |  ▀████████▀  | <- White buck teeth!
  |  ▄████████▄  | <- Red robe + gold clasp  |  ▀████████▀  | <- Cyan/gold collar
  +--------------+                           +--------------+

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py

# Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
def render_stage_curtain(side: str, width: int = STAGE_BOX_WIDTH) -> Text:
    """Render the 5-row x 14-col theatrical curtain label for #left_stage or #right_stage."""
    is_left = (side or "").strip().lower() != "right"
    lines = STAGE_LEFT_CURTAIN_LINES if is_left else STAGE_RIGHT_CURTAIN_LINES
    primary_style = "bold #ffd700 on #0d061a" if is_left else "bold #00ffff on #0d061a"
    mid_style = "italic #ffae42 on #0d061a" if is_left else "bold #ff79c6 on #0d061a"
    txt = Text()
    for idx, raw_line in enumerate(lines):
        if idx > 0:
            txt.append("\n")
        padded = raw_line.center(width)[:width]
        row_style = mid_style if idx == 2 else primary_style
        txt.append(padded, style=row_style)
    return txt
[[[DIVIDER]]]
def render_stage_curtain(side: str, width: int = STAGE_BOX_WIDTH) -> Text:
    """Render the 5-row x 14-col theatrical curtain label for #left_stage or #right_stage."""
    is_left = (side or "").strip().lower() != "right"
    lines = STAGE_LEFT_CURTAIN_LINES if is_left else STAGE_RIGHT_CURTAIN_LINES
    primary_style = "bold #ffd700 on #0d061a" if is_left else "bold #00ffff on #0d061a"
    mid_style = "italic #ffae42 on #0d061a" if is_left else "bold #ff79c6 on #0d061a"
    txt = Text()
    for idx, raw_line in enumerate(lines):
        if idx > 0:
            txt.append("\n")
        padded = raw_line.center(width)[:width]
        row_style = mid_style if idx == 2 else primary_style
        txt.append(padded, style=row_style)
    return txt

# 14x10 pixel-art grids (packed 2 vertical pixels per terminal row via Unicode half-blocks ▀/▄/█)
MIKE_E_PIXELS = (
    "......GL......",
    ".....BGLB.....",
    "....BBLLBB....",
    "...GGGGGGGG...",
    "..EESSSSSSEE..",
    "..ESWKSSKWSE..",
    "..E.CSSSSC.E..",
    "....SSMMSS....",
    "...RRRRRRRR...",
    "..RRRRGGRRRR..",
)
MIKE_E_PALETTE = {
    "B": "#1e40af",  # Sorcerer's Apprentice royal blue hat
    "L": "#3b82f6",  # Wizard hat highlight blue
    "G": "#ffd700",  # Star / hatband / clasp gold
    "E": "#3b2314",  # Round apprentice ears / hair
    "S": "#ffcc99",  # Warm skin
    "C": "#f59e7a",  # Rosy cheek
    "W": "#ffffff",  # Eye sparkle
    "K": "#111827",  # Pupil
    "M": "#991b1b",  # Warm smile
    "R": "#dc2626",  # Apprentice crimson robe
}

SHIPMUNK_PIXELS = (
    "..DD......DD..",
    ".DPFD.DD.DFPD.",
    "..FFDAAAADFF..",
    "..FFKWAAWKFF..",
    ".FFFTTNNTTFFF.",
    ".FFFFTTTTFFFF.",
    "..FFFTWWTFFF..",
    "...FFFFFFFF...",
    "..CCCGGGGCCC..",
    "...FFTTTTFF...",
)
SHIPMUNK_PALETTE = {
    "D": "#5c2d0c",  # Dark walnut ear rim & head stripe
    "F": "#d97706",  # Russet chipmunk fur
    "A": "#f59e0b",  # Amber highlight fur
    "P": "#f472b6",  # Inner ear pink
    "T": "#fde68a",  # Cream muzzle & belly
    "W": "#ffffff",  # Eye shine & signature buck teeth
    "K": "#090d16",  # Pupil
    "N": "#ef4444",  # Chipmunk nose
    "C": "#00ffff",  # Ship of Theseus cyber collar cyan
    "G": "#ffd700",  # Collar rivet gold
}

def _scale_hex_color(hex_str: str, factor: float) -> str:
    """Scale an #rrggbb hex string by factor (1.0 = active full-bright, 0.38 = inactive dimmed)."""
    h = (hex_str or "#000000").lstrip("#")
    if len(h) != 6:
        return "#0d061a"
    r = max(0, min(255, int(int(h[0:2], 16) * factor)))
    g = max(0, min(255, int(int(h[2:4], 16) * factor)))
    b = max(0, min(255, int(int(h[4:6], 16) * factor)))
    return f"#{r:02x}{g:02x}{b:02x}"

def render_stage_portrait(side: str, active: bool = True, width: int = STAGE_BOX_WIDTH) -> Text:
    """Pre-bake a 5-row x 14-col 24-bit Truecolor portrait from a 14x10 pixel grid."""
    is_left = (side or "").strip().lower() != "right"
    pixels = MIKE_E_PIXELS if is_left else SHIPMUNK_PIXELS
    raw_pal = MIKE_E_PALETTE if is_left else SHIPMUNK_PALETTE
    factor = 1.0 if active else 0.38
    bg_hex = "#0d061a" if active else "#080411"
    pal = {k: _scale_hex_color(v, factor) for k, v in raw_pal.items()}
    grid_rows = []
    for row_idx in range(5):
        top_line = pixels[row_idx * 2].ljust(width, ".")[:width]
        bot_line = pixels[row_idx * 2 + 1].ljust(width, ".")[:width]
        row_cells = []
        for col_idx in range(width):
            t_ch = top_line[col_idx]
            b_ch = bot_line[col_idx]
            t_col = pal.get(t_ch)
            b_col = pal.get(b_ch)
            if t_col is None and b_col is None:
                row_cells.append((" ", f"on {bg_hex}"))
            elif t_col is not None and b_col is None:
                row_cells.append(("▀", f"{t_col} on {bg_hex}"))
            elif t_col is None and b_col is not None:
                row_cells.append(("▄", f"{b_col} on {bg_hex}"))
            elif t_col == b_col:
                row_cells.append(("█", f"{t_col} on {bg_hex}"))
            else:
                row_cells.append(("▀", f"{t_col} on {b_col}"))
        grid_rows.append(row_cells)
    return _cells_to_rich_text(grid_rows, f"on {bg_hex}")

STAGE_FRAMES = {
    "left": {
        "curtain": render_stage_curtain("left"),
        "active": render_stage_portrait("left", active=True),
        "inactive": render_stage_portrait("left", active=False),
    },
    "right": {
        "curtain": render_stage_curtain("right"),
        "active": render_stage_portrait("right", active=True),
        "inactive": render_stage_portrait("right", active=False),
    },
}

def get_stage_frame(side: str, state: str = "curtain") -> Text:
    """Return the pre-baked 5x14 Rich.Text frame ('curtain', 'active', or 'inactive') in O(1)."""
    s_key = "right" if (side or "").strip().lower() == "right" else "left"
    st_key = (state or "curtain").strip().lower()
    if st_key not in ("curtain", "active", "inactive"):
        st_key = "curtain"
    return STAGE_FRAMES[s_key][st_key]
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
        self.ticker_cached_width = 0
        self.ticker_last_frame = None
        self.is_fire_burning = False
[[[DIVIDER]]]
        self.ticker_cached_width = 0
        self.ticker_last_frame = None
        self.is_fire_burning = False
        self.left_stage_state = "curtain"
        self.right_stage_state = "curtain"
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def _rebuild_billboard_cache(self, width: int):
[[[DIVIDER]]]
    def update_stage_boxes(self, left_state: str, right_state: str) -> None:
        """Flip #left_stage and #right_stage between pre-baked 'curtain', 'active', and 'inactive' frames."""
        if left_state != getattr(self, "left_stage_state", None):
            self.left_stage_state = left_state
            try:
                self.query_one("#left_stage", Static).update(get_stage_frame("left", left_state))
            except Exception:
                pass
        if right_state != getattr(self, "right_stage_state", None):
            self.right_stage_state = right_state
            try:
                self.query_one("#right_stage", Static).update(get_stage_frame("right", right_state))
            except Exception:
                pass

    def _rebuild_billboard_cache(self, width: int):
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            next_card = cards[self.ticker_card_idx]
            first_frame = next_card["timeline"][0] if next_card.get("timeline") else next_card["still"]
            self.is_fire_burning = first_frame is not next_card.get("still")
            self.ticker_last_frame = first_frame
            bar.update(first_frame)
            return
[[[DIVIDER]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            if self.ticker_card_idx == 0:
                self.update_stage_boxes("curtain", "curtain")
            elif self.ticker_card_idx % 2 == 1:
                self.update_stage_boxes("active", "inactive")
            else:
                self.update_stage_boxes("inactive", "active")
            next_card = cards[self.ticker_card_idx]
            first_frame = next_card["timeline"][0] if next_card.get("timeline") else next_card["still"]
            self.is_fire_burning = first_frame is not next_card.get("still")
            self.ticker_last_frame = first_frame
            bar.update(first_frame)
            return
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS for Baby-Step Win #1 and #2 (`CURTAINS_OK=True PORTRAITS_6_FRAMES=True ACTIVE_VS_DIM=True`), and inspect `remotes/honeybot/scripts/stream.py` and `remotes/honeybot/scripts/logs.py` in raw source. Execute Baby-Step Win #3: wire `stream.py`'s `Narrator` voice queue to write `/tmp/honeybot_stage.json` in real voice-time as speaker turns change (flashing the character's `"curtain"` state when a turn begins, then flipping the speaking character to `"active"` and the listening character to `"inactive"`, and setting both to `"inactive"` or `"curtain"` during host/narrator turns) and wire `logs.py` to follow `/tmp/honeybot_stage.json` whenever it is fresh.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

  WHY IT SLIPPED THROUGH                 HOW WE MAKE IT IMPOSSIBLE TO REPEAT
  .---------------------------------.    .-------------------------------------.
  | 1. Forward Reference at Import: |    | 1. Define Before Eager Call:        |
  |    STAGE_FRAMES ran at line 183,| => |    Anchor STAGE_FRAMES directly     |
  |    but `def _cells_to_rich_text`|    |    above `_bake_doom_fire_timeline` |
  |    lived down at line 209!      |    |    (AFTER `_cells_to_rich_text`!).  |
  |                                 |    |                                     |
  | 2. AST != Runtime Import:       |    | 2. Pre-Flight Import Gate +         |
  |    `apply.py` runs `ast.parse`, | => |    Zero-Downtime TUI Hot-Reload:    |
  |    which validates syntax, not  |    |    `python -c "import logs" &&      |
  |    top-level execution order;   |    |     ./nixops.sh` (SONAR_HOT_RELOAD=1|
  |    and Car 2 rebooted BEFORE    |    |    swaps logs.py in 0.2s WITHOUT    |
  |    the next compile imported it!|    |    restarting stream.py's audio!).  |
  '---------------------------------'    '-------------------------------------'

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
def _bake_doom_fire_timeline(led_rows, width: int, base_style: str, effect: str, still_txt: Text, seed: int = 0):
[[[DIVIDER]]]
# 14x10 pixel-art grids (packed 2 vertical pixels per terminal row via Unicode half-blocks ▀/▄/█)
MIKE_E_PIXELS = (
    "......GL......",
    ".....BGLB.....",
    "....BBLLBB....",
    "...GGGGGGGG...",
    "..EESSSSSSEE..",
    "..ESWKSSKWSE..",
    "..E.CSSSSC.E..",
    "....SSMMSS....",
    "...RRRRRRRR...",
    "..RRRRGGRRRR..",
)
MIKE_E_PALETTE = {
    "B": "#1e40af",  # Sorcerer's Apprentice royal blue hat
    "L": "#3b82f6",  # Wizard hat highlight blue
    "G": "#ffd700",  # Star / hatband / clasp gold
    "E": "#3b2314",  # Round apprentice ears / hair
    "S": "#ffcc99",  # Warm skin
    "C": "#f59e7a",  # Rosy cheek
    "W": "#ffffff",  # Eye sparkle
    "K": "#111827",  # Pupil
    "M": "#991b1b",  # Warm smile
    "R": "#dc2626",  # Apprentice crimson robe
}

SHIPMUNK_PIXELS = (
    "..DD......DD..",
    ".DPFD.DD.DFPD.",
    "..FFDAAAADFF..",
    "..FFKWAAWKFF..",
    ".FFFTTNNTTFFF.",
    ".FFFFTTTTFFFF.",
    "..FFFTWWTFFF..",
    "...FFFFFFFF...",
    "..CCCGGGGCCC..",
    "...FFTTTTFF...",
)
SHIPMUNK_PALETTE = {
    "D": "#5c2d0c",  # Dark walnut ear rim & head stripe
    "F": "#d97706",  # Russet chipmunk fur
    "A": "#f59e0b",  # Amber highlight fur
    "P": "#f472b6",  # Inner ear pink
    "T": "#fde68a",  # Cream muzzle & belly
    "W": "#ffffff",  # Eye shine & signature buck teeth
    "K": "#090d16",  # Pupil
    "N": "#ef4444",  # Chipmunk nose
    "C": "#00ffff",  # Ship of Theseus cyber collar cyan
    "G": "#ffd700",  # Collar rivet gold
}

def _scale_hex_color(hex_str: str, factor: float) -> str:
    """Scale an #rrggbb hex string by factor (1.0 = active full-bright, 0.38 = inactive dimmed)."""
    h = (hex_str or "#000000").lstrip("#")
    if len(h) != 6:
        return "#0d061a"
    r = max(0, min(255, int(int(h[0:2], 16) * factor)))
    g = max(0, min(255, int(int(h[2:4], 16) * factor)))
    b = max(0, min(255, int(int(h[4:6], 16) * factor)))
    return f"#{r:02x}{g:02x}{b:02x}"

def render_stage_portrait(side: str, active: bool = True, width: int = STAGE_BOX_WIDTH) -> Text:
    """Pre-bake a 5-row x 14-col 24-bit Truecolor portrait from a 14x10 pixel grid (requires _cells_to_rich_text above)."""
    is_left = (side or "").strip().lower() != "right"
    pixels = MIKE_E_PIXELS if is_left else SHIPMUNK_PIXELS
    raw_pal = MIKE_E_PALETTE if is_left else SHIPMUNK_PALETTE
    factor = 1.0 if active else 0.38
    bg_hex = "#0d061a" if active else "#080411"
    pal = {k: _scale_hex_color(v, factor) for k, v in raw_pal.items()}
    grid_rows = []
    for row_idx in range(5):
        top_line = pixels[row_idx * 2].ljust(width, ".")[:width]
        bot_line = pixels[row_idx * 2 + 1].ljust(width, ".")[:width]
        row_cells = []
        for col_idx in range(width):
            t_ch = top_line[col_idx]
            b_ch = bot_line[col_idx]
            t_col = pal.get(t_ch)
            b_col = pal.get(b_ch)
            if t_col is None and b_col is None:
                row_cells.append((" ", f"on {bg_hex}"))
            elif t_col is not None and b_col is None:
                row_cells.append(("▀", f"{t_col} on {bg_hex}"))
            elif t_col is None and b_col is not None:
                row_cells.append(("▄", f"{b_col} on {bg_hex}"))
            elif t_col == b_col:
                row_cells.append(("█", f"{t_col} on {bg_hex}"))
            else:
                row_cells.append(("▀", f"{t_col} on {b_col}"))
        grid_rows.append(row_cells)
    return _cells_to_rich_text(grid_rows, f"on {bg_hex}")

STAGE_FRAMES = {
    "left": {
        "curtain": render_stage_curtain("left"),
        "active": render_stage_portrait("left", active=True),
        "inactive": render_stage_portrait("left", active=False),
    },
    "right": {
        "curtain": render_stage_curtain("right"),
        "active": render_stage_portrait("right", active=True),
        "inactive": render_stage_portrait("right", active=False),
    },
}

def get_stage_frame(side: str, state: str = "curtain") -> Text:
    """Return the pre-baked 5x14 Rich.Text frame ('curtain', 'active', or 'inactive') in O(1)."""
    s_key = "right" if (side or "").strip().lower() == "right" else "left"
    st_key = (state or "curtain").strip().lower()
    if st_key not in ("curtain", "active", "inactive"):
        st_key = "curtain"
    return STAGE_FRAMES[s_key][st_key]

def _bake_doom_fire_timeline(led_rows, width: int, base_style: str, effect: str, still_txt: Text, seed: int = 0):
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
        self.ticker_cached_width = 0
        self.ticker_last_frame = None
        self.is_fire_burning = False
[[[DIVIDER]]]
        self.ticker_cached_width = 0
        self.ticker_last_frame = None
        self.is_fire_burning = False
        self.left_stage_state = "curtain"
        self.right_stage_state = "curtain"
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def _rebuild_billboard_cache(self, width: int):
[[[DIVIDER]]]
    def update_stage_boxes(self, left_state: str, right_state: str) -> None:
        """Flip #left_stage and #right_stage between pre-baked 'curtain', 'active', and 'inactive' frames."""
        if left_state != getattr(self, "left_stage_state", None):
            self.left_stage_state = left_state
            try:
                self.query_one("#left_stage", Static).update(get_stage_frame("left", left_state))
            except Exception:
                pass
        if right_state != getattr(self, "right_stage_state", None):
            self.right_stage_state = right_state
            try:
                self.query_one("#right_stage", Static).update(get_stage_frame("right", right_state))
            except Exception:
                pass

    def _rebuild_billboard_cache(self, width: int):
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            next_card = cards[self.ticker_card_idx]
            first_frame = next_card["timeline"][0] if next_card.get("timeline") else next_card["still"]
            self.is_fire_burning = first_frame is not next_card.get("still")
            self.ticker_last_frame = first_frame
            bar.update(first_frame)
            return
[[[DIVIDER]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            if self.ticker_card_idx == 0:
                self.update_stage_boxes("curtain", "curtain")
            elif self.ticker_card_idx % 2 == 1:
                self.update_stage_boxes("active", "inactive")
            else:
                self.update_stage_boxes("inactive", "active")
            next_card = cards[self.ticker_card_idx]
            first_frame = next_card["timeline"][0] if next_card.get("timeline") else next_card["still"]
            self.is_fire_burning = first_frame is not next_card.get("still")
            self.ticker_last_frame = first_frame
            bar.update(first_frame)
            return
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS for Baby-Step Win #2 (`CURTAINS_OK=True PORTRAITS_6_FRAMES=True ACTIVE_VS_DIM=True`), confirming that all six 24-bit Truecolor stage frames import and render cleanly on Honeybot. Now execute Baby-Step Win #3: wire `remotes/honeybot/scripts/stream.py`'s `Narrator` voice pipeline to write `/tmp/honeybot_stage.json` in real voice-time as speaker turns change (flashing the speaking character's `"curtain"` state briefly when a turn switches, then flipping the speaker to `"active"` and the listener to `"inactive"`, and setting both to `"curtain"` or `"inactive"` during host/narrator turns) and wire `remotes/honeybot/scripts/logs.py` to follow `/tmp/honeybot_stage.json` whenever it is fresh.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

  REAL VOICE-TIME STAGE SYNC (STREAM.PY -> /tmp/honeybot_stage.json -> LOGS.PY)
  ==============================================================================
  1. aplay spawns in _speak_now() or _play_raw_now()
       |
       +--> Narrator.sync_stage_for_utterance(text, sample_rate)
              - Turn switched to Mike-E (17,600 Hz)?
                  left="active", right="inactive",
                  curtain_side="left", curtain_until = now + 1.8s
              - Same speaker continuing (Paragraph 2, 3, 4...)?
                  curtain_side="", curtain_until = 0.0 (stays solid active!)
              - Turn switched to Shipmunk (22,050 Hz)?
                  left="inactive", right="active",
                  curtain_side="right", curtain_until = now + 1.8s
              - 90s "Signal check"?
                  left="inactive", right="inactive" (preserves speaker memory
                  so no false curtain flash when the speaker resumes 2s later!)
              - Article intro ("Now I read...", "Reading entry...")?
                  left="curtain", right="curtain"
       |
       v
  2. logs.py checks st_mtime_ns every 0.08s (BEFORE _is_hud_scrolling() return!)
       - While now < curtain_until (the 1.8s speaker cue):
           Shows Curtain Label ("Mike-E / the / Apprentice" or "Shipmunk...")
       - Once now >= curtain_until:
           Automatically lifts curtain to 100% Active RGB portrait with ZERO
           extra disk writes or timer threads!

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/content_loader.py
imports/voice_synthesis.py
! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
MODEL_DIR = Path.home() / ".local/share/piper_voices"
MODEL_NAME = "en_US-amy-low.onnx"
TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_TTS_RATE", "22050"))
HUMAN_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_HUMAN_TTS_RATE", "17600"))
CHIPMUNK_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_CHIPMUNK_TTS_RATE", "22050"))
_AI_SPEAKER_RE = (
[[[DIVIDER]]]
MODEL_DIR = Path.home() / ".local/share/piper_voices"
MODEL_NAME = "en_US-amy-low.onnx"
TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_TTS_RATE", "22050"))
HUMAN_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_HUMAN_TTS_RATE", "17600"))
CHIPMUNK_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_CHIPMUNK_TTS_RATE", "22050"))
STAGE_STATE_PATH = "/tmp/honeybot_stage.json"
STAGE_CURTAIN_SECONDS = 1.8
_STAGE_CURTAIN_INTRO_RE = (
    "now i read", "reading entry", "reading article", "greetings",
)
_stage_seq = 0
_stage_write_lock = threading.Lock()

def write_stage_state(
    speaker: str = "narrator",
    left: str = "curtain",
    right: str = "curtain",
    curtain_side: str = "",
    curtain_seconds: float = 0.0,
    stage_path: str = STAGE_STATE_PATH,
) -> dict:
    """Atomically write /tmp/honeybot_stage.json so logs.py flips stage portraits in voice-order."""
    import json as _json
    global _stage_seq
    now = time.time()
    with _stage_write_lock:
        _stage_seq += 1
        payload = {
            "speaker": speaker,
            "left": left,
            "right": right,
            "curtain_side": curtain_side,
            "curtain_until": round(now + curtain_seconds, 3) if (curtain_side and curtain_seconds > 0) else 0.0,
            "seq": _stage_seq,
            "updated_at": round(now, 3),
        }
        if stage_path:
            try:
                p = Path(stage_path)
                tmp = p.with_name(f".{p.name}.{os.getpid()}.tmp")
                tmp.write_text(_json.dumps(payload), encoding="utf-8")
                os.replace(tmp, p)
            except Exception:
                pass
        return payload

_AI_SPEAKER_RE = (
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
        self._tts_seq = 0
        self._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        self._speaker_rate = HUMAN_TTS_SAMPLE_RATE
        self.exclusive_mode = threading.Event()
[[[DIVIDER]]]
        self._tts_seq = 0
        self._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        self._speaker_rate = HUMAN_TTS_SAMPLE_RATE
        self._stage_dialogue_role = "narrator"
        self.live_operator_mode = threading.Event()
        self.exclusive_mode = threading.Event()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    def _cleanup_item(self, item):
        """Remove any unplayed pre-synthesized .raw file when flushing the queue."""
[[[DIVIDER]]]
    def sync_stage_for_utterance(self, text: str, sample_rate: int = None, stage_path: str = STAGE_STATE_PATH) -> dict:
        """Update /tmp/honeybot_stage.json at the exact moment ALSA playback begins."""
        low = (text or "").strip().lower()
        rate = int(sample_rate or self._speaker_rate or HUMAN_TTS_SAMPLE_RATE)
        if self.live_operator_mode.is_set():
            prev = self._stage_dialogue_role
            self._stage_dialogue_role = "left"
            c_side = "left" if prev != "left" else ""
            return write_stage_state("left", "active", "inactive", c_side, STAGE_CURTAIN_SECONDS if c_side else 0.0, stage_path=stage_path)
        if low.startswith("signal check"):
            return write_stage_state("transient", "inactive", "inactive", "", 0.0, stage_path=stage_path)
        if any(low.startswith(p) for p in _STAGE_CURTAIN_INTRO_RE):
            self._stage_dialogue_role = "narrator"
            return write_stage_state("narrator", "curtain", "curtain", "", 0.0, stage_path=stage_path)
        if self.exclusive_mode.is_set() or any(low.startswith(p) for p in _HOST_TRANSIENT_RE):
            self._stage_dialogue_role = "narrator"
            return write_stage_state("narrator", "inactive", "inactive", "", 0.0, stage_path=stage_path)
        if not low:
            role = "right" if rate >= 20000 else ("left" if self._stage_dialogue_role == "left" else "narrator")
        elif rate >= 20000 or any(low.startswith(p) for p in _AI_SPEAKER_RE) or re.match(r"^(?:gemini|claude|chat\s*g\s*p\s*t|openai|opus|sonnet)\b[^:.!?]{0,35}:", low):
            role = "right"
        else:
            role = "left"
        if role == "right":
            c_side = "right" if self._stage_dialogue_role != "right" else ""
            self._stage_dialogue_role = "right"
            return write_stage_state("right", "inactive", "active", c_side, STAGE_CURTAIN_SECONDS if c_side else 0.0, stage_path=stage_path)
        if role == "left":
            c_side = "left" if self._stage_dialogue_role != "left" else ""
            self._stage_dialogue_role = "left"
            return write_stage_state("left", "active", "inactive", c_side, STAGE_CURTAIN_SECONDS if c_side else 0.0, stage_path=stage_path)
        return write_stage_state("narrator", "curtain", "curtain", "", 0.0, stage_path=stage_path)

    def _cleanup_item(self, item):
        """Remove any unplayed pre-synthesized .raw file when flushing the queue."""
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    def play_raw(self, raw_path, fallback_text=""):
        """Enqueue a pre-synthesized .raw file for zero-latency playback, falling back to text."""
        rate = self.resolve_speaker_rate(fallback_text) if fallback_text else self._speaker_rate
        if raw_path and os.path.exists(raw_path):
            self.queue.put(("RAW", str(raw_path), rate))
        elif fallback_text:
            self.queue.put(fallback_text)
[[[DIVIDER]]]
    def play_raw(self, raw_path, fallback_text=""):
        """Enqueue a pre-synthesized .raw file for zero-latency playback, falling back to text."""
        rate = self.resolve_speaker_rate(fallback_text) if fallback_text else self._speaker_rate
        if raw_path and os.path.exists(raw_path):
            self.queue.put(("RAW", str(raw_path), rate, str(fallback_text or "")))
        elif fallback_text:
            self.queue.put(fallback_text)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
                elif isinstance(item, tuple) and item and item[0] == "RAW":
                    raw_rate = item[2] if len(item) >= 3 else self._speaker_rate
                    self._play_raw_now(item[1], sample_rate=raw_rate)
                else:
                    self._speak_now(item)
[[[DIVIDER]]]
                elif isinstance(item, tuple) and item and item[0] == "RAW":
                    raw_rate = item[2] if len(item) >= 3 else self._speaker_rate
                    raw_text = item[3] if len(item) >= 4 else ""
                    self._play_raw_now(item[1], sample_rate=raw_rate, text=raw_text)
                else:
                    self._speak_now(item)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
                    p1.stdout.close()
                    spk_rate = self.resolve_speaker_rate(text)
                    p3 = subprocess.Popen(
[[[DIVIDER]]]
                    p1.stdout.close()
                    spk_rate = self.resolve_speaker_rate(text)
                    self.sync_stage_for_utterance(text, sample_rate=spk_rate)
                    p3 = subprocess.Popen(
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    def _play_raw_now(self, raw_path, sample_rate=None):
        """Play a pre-synthesized .raw file immediately via aplay and unlink when done."""
[[[DIVIDER]]]
    def _play_raw_now(self, raw_path, sample_rate=None, text=""):
        """Play a pre-synthesized .raw file immediately via aplay and unlink when done."""
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
            try:
                with self._proc_lock:
                    p_play = subprocess.Popen(
                        ["aplay", "-r", str(rate), "-f", "S16_LE", "-t", "raw", str(raw_path)],
[[[DIVIDER]]]
            try:
                with self._proc_lock:
                    self.sync_stage_for_utterance(text, sample_rate=rate)
                    p_play = subprocess.Popen(
                        ["aplay", "-r", str(rate), "-f", "S16_LE", "-t", "raw", str(raw_path)],
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    elif command == "VISIT":
        narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
[[[DIVIDER]]]
    elif command == "VISIT":
        narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
        narrator._stage_dialogue_role = "narrator"
        write_stage_state("narrator", "curtain", "curtain")
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    elif command == "BGM":
        if str(content).strip().upper() in ("STOP", "OFF", "0"):
            stop_storytime_bgm()
        else:
            narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
            narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
            start_storytime_bgm()
[[[DIVIDER]]]
    elif command == "BGM":
        if str(content).strip().upper() in ("STOP", "OFF", "0"):
            stop_storytime_bgm()
        else:
            narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
            narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
            narrator._stage_dialogue_role = "narrator"
            write_stage_state("narrator", "curtain", "curtain")
            start_storytime_bgm()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
        card_label = msg.get("card", "") if isinstance(msg, dict) else "LIVE\\nFROM MIKE"
        say_text = msg.get("say", "") if isinstance(msg, dict) else str(msg)
        patronus_key = msg.get("patronus") if isinstance(msg, dict) else None
        media_spec = msg.get("media") if isinstance(msg, dict) else None
        window_spec = msg.get("window") if isinstance(msg, dict) else None
[[[DIVIDER]]]
        card_label = msg.get("card", "") if isinstance(msg, dict) else "LIVE\\nFROM MIKE"
        say_text = msg.get("say", "") if isinstance(msg, dict) else str(msg)
        patronus_key = msg.get("patronus") if isinstance(msg, dict) else None
        media_spec = msg.get("media") if isinstance(msg, dict) else None
        window_spec = msg.get("window") if isinstance(msg, dict) else None
        if "MIKE" in str(card_label or "").upper() and not window_spec and not patronus_key:
            narrator.live_operator_mode.set()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    finally:
        _in_live_message = False
        if not was_exclusive:
            narrator.exclusive_mode.clear()
[[[DIVIDER]]]
    finally:
        narrator.live_operator_mode.clear()
        _in_live_message = False
        if not was_exclusive:
            narrator.exclusive_mode.clear()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
def main():
    _reap_orphaned_tts_and_overlays()
    atexit.register(_reap_orphaned_tts_and_overlays)
    narrator.start()
[[[DIVIDER]]]
def main():
    _reap_orphaned_tts_and_overlays()
    write_stage_state("narrator", "curtain", "curtain")
    atexit.register(_reap_orphaned_tts_and_overlays)
    narrator.start()
[[[REPLACE]]]

Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
TICKER_FILE_PATH = "/tmp/honeybot_ticker.txt"
WEEBO_ACTIVE_PATH = "/tmp/honeybot_weebo_active"
VIEWERS_FILE_PATH = "/tmp/honeybot_viewers.json"
VIEWERS_TTL_SECONDS = 120.0
[[[DIVIDER]]]
TICKER_FILE_PATH = "/tmp/honeybot_ticker.txt"
WEEBO_ACTIVE_PATH = "/tmp/honeybot_weebo_active"
VIEWERS_FILE_PATH = "/tmp/honeybot_viewers.json"
VIEWERS_TTL_SECONDS = 120.0
STAGE_STATE_PATH = "/tmp/honeybot_stage.json"
STAGE_STATE_TTL_SECONDS = 900.0
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
def get_stage_frame(side: str, state: str = "curtain") -> Text:
    """Return the pre-baked 5x14 Rich.Text frame ('curtain', 'active', or 'inactive') in O(1)."""
    s_key = "right" if (side or "").strip().lower() == "right" else "left"
    st_key = (state or "curtain").strip().lower()
    if st_key not in ("curtain", "active", "inactive"):
        st_key = "curtain"
    return STAGE_FRAMES[s_key][st_key]
[[[DIVIDER]]]
def get_stage_frame(side: str, state: str = "curtain") -> Text:
    """Return the pre-baked 5x14 Rich.Text frame ('curtain', 'active', or 'inactive') in O(1)."""
    s_key = "right" if (side or "").strip().lower() == "right" else "left"
    st_key = (state or "curtain").strip().lower()
    if st_key not in ("curtain", "active", "inactive"):
        st_key = "curtain"
    return STAGE_FRAMES[s_key][st_key]

def compute_stage_states_from_payload(payload: dict, now: float = None):
    """Resolve (left_state, right_state) from a /tmp/honeybot_stage.json payload at timestamp `now`."""
    if not isinstance(payload, dict):
        return None
    ts_now = time.time() if now is None else float(now)
    updated_at = float(payload.get("updated_at") or 0.0)
    if updated_at <= 0.0 or (ts_now - updated_at) > STAGE_STATE_TTL_SECONDS:
        return None
    valid = ("curtain", "active", "inactive")
    left_st = str(payload.get("left") or "curtain").strip().lower()
    right_st = str(payload.get("right") or "curtain").strip().lower()
    if left_st not in valid:
        left_st = "curtain"
    if right_st not in valid:
        right_st = "curtain"
    c_side = str(payload.get("curtain_side") or "").strip().lower()
    c_until = float(payload.get("curtain_until") or 0.0)
    if c_until > ts_now:
        if c_side == "left":
            left_st = "curtain"
        elif c_side == "right":
            right_st = "curtain"
    return left_st, right_st
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    TITLE = "Web Bot AI Observatory — https://mikelev.in (RFC 9110 · RFC 9421 · JS Trapdoor)"
    countdown_str = "--:--"
    viewers_mtime = None
    live_viewers = None
    peak_viewers = 0
[[[DIVIDER]]]
    TITLE = "Web Bot AI Observatory — https://mikelev.in (RFC 9110 · RFC 9421 · JS Trapdoor)"
    countdown_str = "--:--"
    viewers_mtime = None
    live_viewers = None
    peak_viewers = 0
    stage_mtime_ns = None
    stage_payload = None
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
        self.is_fire_burning = False
        self.left_stage_state = "curtain"
        self.right_stage_state = "curtain"
[[[DIVIDER]]]
        self.is_fire_burning = False
        self.left_stage_state = "curtain"
        self.right_stage_state = "curtain"
        self.stage_mtime_ns = None
        self.stage_payload = None
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
        self._reload_ticker_source(force=True)
        self.set_interval(0.08, self.tick_marquee)
[[[DIVIDER]]]
        self._reload_ticker_source(force=True)
        self._sync_stage_from_file()
        self.set_interval(0.08, self.tick_marquee)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def update_stage_boxes(self, left_state: str, right_state: str) -> None:
        """Flip #left_stage and #right_stage between pre-baked 'curtain', 'active', and 'inactive' frames."""
        if left_state != getattr(self, "left_stage_state", None):
            self.left_stage_state = left_state
            try:
                self.query_one("#left_stage", Static).update(get_stage_frame("left", left_state))
            except Exception:
                pass
        if right_state != getattr(self, "right_stage_state", None):
            self.right_stage_state = right_state
            try:
                self.query_one("#right_stage", Static).update(get_stage_frame("right", right_state))
            except Exception:
                pass
[[[DIVIDER]]]
    def update_stage_boxes(self, left_state: str, right_state: str) -> None:
        """Flip #left_stage and #right_stage between pre-baked 'curtain', 'active', and 'inactive' frames."""
        if left_state != getattr(self, "left_stage_state", None):
            self.left_stage_state = left_state
            try:
                self.query_one("#left_stage", Static).update(get_stage_frame("left", left_state))
            except Exception:
                pass
        if right_state != getattr(self, "right_stage_state", None):
            self.right_stage_state = right_state
            try:
                self.query_one("#right_stage", Static).update(get_stage_frame("right", right_state))
            except Exception:
                pass

    def _sync_stage_from_file(self) -> bool:
        """Sync #left_stage and #right_stage from /tmp/honeybot_stage.json in O(1); True when live voice owns stage."""
        try:
            if not os.path.exists(STAGE_STATE_PATH):
                self.stage_mtime_ns = None
                self.stage_payload = None
                return False
            st = os.stat(STAGE_STATE_PATH)
            now = time.time()
            if (now - st.st_mtime) > STAGE_STATE_TTL_SECONDS:
                self.stage_payload = None
                return False
            if st.st_mtime_ns != getattr(self, "stage_mtime_ns", None):
                import json as _json
                with open(STAGE_STATE_PATH, "r", encoding="utf-8") as sf:
                    data = _json.load(sf)
                self.stage_payload = data
                self.stage_mtime_ns = st.st_mtime_ns
            pair = compute_stage_states_from_payload(self.stage_payload, now=now)
            if pair is None:
                return False
            self.update_stage_boxes(pair[0], pair[1])
            return True
        except Exception:
            return False
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    def tick_marquee(self):
        """Drive #ticker_bar: fast-pathed 4.5s still-card hold with pre-baked DOOM fire frames."""
        if os.path.exists(WEEBO_ACTIVE_PATH):
            self._was_weebo_active = True
            return
        if self._is_hud_scrolling() and self.ticker_mode == "billboard" and not getattr(self, "is_fire_burning", False):
            return
[[[DIVIDER]]]
    def tick_marquee(self):
        """Drive #ticker_bar: fast-pathed 4.5s still-card hold with pre-baked DOOM fire frames."""
        if os.path.exists(WEEBO_ACTIVE_PATH):
            self._was_weebo_active = True
            return
        stage_synced = self._sync_stage_from_file()
        if self._is_hud_scrolling() and self.ticker_mode == "billboard" and not getattr(self, "is_fire_burning", False):
            return
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            if self.ticker_card_idx == 0:
                self.update_stage_boxes("curtain", "curtain")
            elif self.ticker_card_idx % 2 == 1:
                self.update_stage_boxes("active", "inactive")
            else:
                self.update_stage_boxes("inactive", "active")
            next_card = cards[self.ticker_card_idx]
[[[DIVIDER]]]
            self.ticker_card_idx = (self.ticker_card_idx + 1) % len(cards)
            if not stage_synced:
                if self.ticker_card_idx == 0:
                    self.update_stage_boxes("curtain", "curtain")
                elif self.ticker_card_idx % 2 == 1:
                    self.update_stage_boxes("active", "inactive")
                else:
                    self.update_stage_boxes("inactive", "active")
            next_card = cards[self.ticker_card_idx]
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS for Baby-Step Win #3 (`STAGE_SYNC_OK=True` and the `rg` inventory of `piper_voices|amy-low|\.onnx`), and inspect `remotes/honeybot/scripts/stream.py`, `remotes/honeybot/scripts/content_loader.py`, and `imports/voice_synthesis.py` in raw source. Execute Baby-Step Win #4: upgrade the broadcast from two voices to our 3-Voice Piper TTS Cast:
1. **Norma Centroid** (`en_US-amy-low.onnx` at natural human pitch) as the Stage Narrator who reads intros, station breaks, and speaker-turn cues (`"Mike-E the Apprentice:"`, `"Shipmunk O'Theseus:"`) while the curtain placard is displayed;
2. **Mike-E the Apprentice** (provision a mainstream, stable male Piper `.onnx` model such as `en_US-lessac-low.onnx` or `en_US-ryan-low.onnx` into `~/.local/share/piper_voices`, pitched up slightly for that brisk "Mickey the Apprentice" edge) speaking Mike's body paragraphs without reading his own name tag; and
3. **Shipmunk O'Theseus!** (`en_US-amy-low.onnx` at `22,050` Hz chipmunk pitch) speaking the AI body paragraphs without reading his own name tag.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
context.txt
remotes/honeybot/scripts/logs.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/content_loader.py
imports/voice_synthesis.py
! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

remotes/honeybot/nixos/configuration.nix:944:      MODEL_DIR="$HOME/.local/share/piper_voices"
remotes/honeybot/nixos/configuration.nix:945:      MODEL_NAME="en_US-amy-low.onnx"
remotes/honeybot/nixos/configuration.nix:992:      MODEL_NAME="en_US-amy-low.onnx"
remotes/honeybot/nixos/configuration.nix:1083:      MODEL_NAME="en_US-amy-low.onnx"

  WHAT HAPPENED IN STEP 3                 WHAT HAPPENS IN STEP 4 (3-VOICE CAST)
  .----------------------------------.    .---------------------------------------.
  | 1. Intro / "Setting the Stage":  |    | 1. Intro / "Setting the Stage":       |
  |    `sync_stage_for_utterance()`  | => |    Both boxes stay in Amiga Half-Brite|
  |    had `else: role = "left"`,    |    |    (`inactive`, `inactive`) while     |
  |    lighting Mike-E too early!    |    |    Norma Centroid reads the prologue  |
  |                                  |    |    until `Technical Journal Entry`    |
  |                                  |    |    or `Mike:` / `Gemini:` begins!     |
  |                                  |    |                                       |
  | 2. 1.8s Late Switch:             |    | 2. Zero-Lag Curtain -> Voice Split:   |
  |    Amy read `"Mike: <body...>"`  | => |    Cue A (Norma @ 16kHz): speaks      |
  |    as one blob while curtain     |    |      "Mike-E the Apprentice" while    |
  |    held for 1.8s over her voice. |    |      Curtain Label shows on stage.    |
  |                                  |    |    Cue B (Male ONNX @ brisk pitch):   |
  |                                  |    |      Curtain lifts at t=0.0s to 100%  |
  |                                  |    |      Active as Mike-E speaks `<body>`!|
  '----------------------------------'    '---------------------------------------'

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py

# # Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/content_loader.py
# imports/voice_synthesis.py
# ! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
# ! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

# Context 7 (Hydrate configuration.nix + score.py for Declarative Nix 3-Voice Cast & Amiga Half-Brite Intro Gate)
context.txt
remotes/honeybot/nixos/configuration.nix
remotes/honeybot/scripts/score.py
remotes/honeybot/scripts/content_loader.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/logs.py
! .venv/bin/python -c 'import httpx; b = "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/"; vs = [("amy", "amy/low/en_US-amy-low"), ("ryan", "ryan/low/en_US-ryan-low"), ("lessac", "lessac/low/en_US-lessac-low"), ("danny", "danny/low/en_US-danny-low")]; rs = {k: (r.json()["audio"]["sample_rate"] if r.status_code == 200 else r.status_code) for k, u in vs for r in [httpx.get(b + u + ".onnx.json", follow_redirects=True, timeout=5.0)]}; print("PIPER_LOW_RATES=" + str(rs))'
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import content_loader as cl; sample = "## Setting the Stage\n\nIntro paragraph.\n\n## Technical Journal Entry Begins\n\n**MikeLev.in**: First human line.\n\nSecond human line.\n\n**Gemini 4 Argon High**: **TL;DR**: First AI line."; print("CLEAN_SAMPLE=" + repr(cl.clean_markdown(sample, label="probe", census=False)))'

Verify the LIVE COMMAND RECEIPTS (`PIPER_LOW_RATES` and `CLEAN_SAMPLE`) and inspect `remotes/honeybot/nixos/configuration.nix`, `remotes/honeybot/scripts/score.py`, `remotes/honeybot/scripts/content_loader.py`, and `remotes/honeybot/scripts/stream.py` in raw source. Now execute Baby-Step Win #4 declaratively (the Nix way, with zero mutation-machine gaps for future staging hardware):
1. Declaratively provision the male Piper `.onnx` model (`.onnx` + `.onnx.json`) alongside `en_US-amy-low.onnx` in `remotes/honeybot/nixos/configuration.nix` (with an idempotent fallback check in `stream.py` so any machine converges cleanly).
2. Keep both stage portraits in Amiga Half-Brite (`"inactive"`, `"inactive"`) during Norma Centroid's article intro and `"Setting the Stage"`, only lighting a character once `"Technical Journal Entry Begins"` or an explicit dialogue turn starts.
3. Wire the 3-Voice Cast with zero-lag curtain cues:
   - **Norma Centroid** (`en_US-amy-low.onnx` at natural `16,000` Hz) reads intros, station breaks, and short speaker-turn cues (`"Mike-E the Apprentice."` / `"Shipmunk O'Theseus!"`) while that character's `"curtain"` placard is shown on stage;
   - **Mike-E the Apprentice** (male `.onnx` model pitched slightly brisk) immediately speaks Mike's body text (with `"Mike:"` stripped so he never reads his own name tag) while Stage Left snaps to `"active"` at `t = 0.0s`;
   - **Shipmunk O'Theseus!** (`en_US-amy-low.onnx` at `22,050` Hz chipmunk pitch) immediately speaks the AI's body text (with `"Gemini ...:"` stripped) while Stage Right snaps to `"active"` at `t = 0.0s`.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/content_loader.py
# imports/voice_synthesis.py
# ! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
# ! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

# Context 7 (Hydrate configuration.nix + score.py for Declarative Nix 3-Voice Cast & Amiga Half-Brite Intro Gate)
context.txt
remotes/honeybot/nixos/configuration.nix
remotes/honeybot/scripts/score.py
remotes/honeybot/scripts/content_loader.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/logs.py
! .venv/bin/python -c 'import httpx; b = "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/"; vs = [("amy", "amy/low/en_US-amy-low"), ("ryan", "ryan/low/en_US-ryan-low"), ("lessac", "lessac/low/en_US-lessac-low"), ("danny", "danny/low/en_US-danny-low")]; rs = {k: (r.json()["audio"]["sample_rate"] if r.status_code == 200 else r.status_code) for k, u in vs for r in [httpx.get(b + u + ".onnx.json", follow_redirects=True, timeout=5.0)]}; print("PIPER_LOW_RATES=" + str(rs))'
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import content_loader as cl; sample = "## Setting the Stage\n\nIntro paragraph.\n\n## Technical Journal Entry Begins\n\n**MikeLev.in**: First human line.\n\nSecond human line.\n\n**Gemini 4 Argon High**: **TL;DR**: First AI line."; print("CLEAN_SAMPLE=" + repr(cl.clean_markdown(sample, label="probe", census=False)))'

  WHY STEP 3 FELT LIKE A GAME HUD        WHY STEP 4 FEELS LIKE A BROADCAST STUDIO
  .---------------------------------.    .-----------------------------------------.
  | - Mike-E lit up at 100% primary |    | - Amiga Half-Brite Intro Gate:          |
  |   red/blue/gold from frame 0    |    |   During station cards, "Reading entry",|
  |   over the serious bot logs.    | => |   and "Setting the Stage", BOTH boxes   |
  | - One female voice read all     |    |   recede into dark 38% #080411 shadow   |
  |   parts, so the pixel heads had |    |   so the live RFC 9110 HUD dominates!   |
  |   no acoustic job to perform.   |    | - Only at dialogue turns does Norma     |
  |                                 |    |   raise the Curtain Placard & hand the  |
  |                                 |    |   mic to a genuinely distinct voice!    |
  '---------------------------------'    '-----------------------------------------'

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 7 (Hydrate configuration.nix + score.py for Declarative Nix 3-Voice Cast & Amiga Half-Brite Intro Gate)
# context.txt
# remotes/honeybot/nixos/configuration.nix
# remotes/honeybot/scripts/score.py
# remotes/honeybot/scripts/content_loader.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/logs.py

# Context 8 (Verify Baby-Step Win #4: Declarative Nix 3-Voice Cast + Amiga Half-Brite Intro Gate + Zero-Lag Curtain Sync)
context.txt
remotes/honeybot/nixos/configuration.nix
remotes/honeybot/scripts/score.py
remotes/honeybot/scripts/content_loader.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/logs.py
! nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
! .venv/bin/python -c 'import sys, pathlib; sys.path.insert(0, "remotes/honeybot/scripts"); import stream; nix_txt = pathlib.Path("remotes/honeybot/nixos/configuration.nix").read_text(encoding="utf-8"); nix_ok = ("en_US-ryan-low.onnx" in nix_txt and "en_US-ryan-low.onnx.json" in nix_txt); nar = stream.Narrator(); has_p = hasattr(nar, "plan_utterances"); nar.reset_for_article() if hasattr(nar, "reset_for_article") else None; seq = [(s["voice"], s["text"], s["left"], s["right"]) for line in ("Setting the Stage", "Intro paragraph.", "Technical Journal Entry Begins", "Mike: First human line.", "Second human line.", "Gemini 4 Argon High: First AI line.") for s in (nar.plan_utterances(line) if has_p else [])]; exp = [("norma", "Setting the Stage", "inactive", "inactive"), ("norma", "Intro paragraph.", "inactive", "inactive"), ("norma", "Technical Journal Entry Begins", "inactive", "inactive"), ("norma", "Mike E, the Apprentice.", "curtain", "inactive"), ("mike", "First human line.", "active", "inactive"), ("mike", "Second human line.", "active", "inactive"), ("norma", "Shipmunk O\x27Theseus!", "inactive", "curtain"), ("shipmunk", "First AI line.", "inactive", "active")]; print("NIX_RYAN_DECLARED=" + str(nix_ok), "CAST_3_VOICE_OK=" + str(seq == exp))'

Target: remotes/honeybot/nixos/configuration.nix
[[[SEARCH]]]
    (writeShellScriptBin "stream" ''
      # --- Ensure Voice Model Exists ---
      MODEL_DIR="$HOME/.local/share/piper_voices"
      MODEL_NAME="en_US-amy-low.onnx"
      JSON_NAME="en_US-amy-low.onnx.json"
      URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/low"

      mkdir -p "$MODEL_DIR"
      if [ ! -f "$MODEL_DIR/$MODEL_NAME" ]; then
        echo "📥 Downloading voice model..."
        ${pkgs.curl}/bin/curl -L -o "$MODEL_DIR/$MODEL_NAME" "$URL_BASE/$MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -L -o "$MODEL_DIR/$JSON_NAME" "$URL_BASE/$JSON_NAME?download=true"
      fi
      # ---------------------------------
[[[DIVIDER]]]
    (writeShellScriptBin "stream" ''
      # --- Ensure 3-Voice Cast Models Exist (Norma/Shipmunk + Mike-E) ---
      MODEL_DIR="$HOME/.local/share/piper_voices"
      MODEL_NAME="en_US-amy-low.onnx"
      JSON_NAME="en_US-amy-low.onnx.json"
      URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/low"
      MIKE_MODEL_NAME="en_US-ryan-low.onnx"
      MIKE_JSON_NAME="en_US-ryan-low.onnx.json"
      MIKE_URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/ryan/low"

      mkdir -p "$MODEL_DIR"
      if [ ! -s "$MODEL_DIR/$MODEL_NAME" ] || [ ! -s "$MODEL_DIR/$JSON_NAME" ]; then
        echo "📥 Downloading Norma/Shipmunk voice model ($MODEL_NAME)..."
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MODEL_NAME" "$URL_BASE/$MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$JSON_NAME" "$URL_BASE/$JSON_NAME?download=true"
      fi
      if [ ! -s "$MODEL_DIR/$MIKE_MODEL_NAME" ] || [ ! -s "$MODEL_DIR/$MIKE_JSON_NAME" ]; then
        echo "📥 Downloading Mike-E voice model ($MIKE_MODEL_NAME)..."
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MIKE_MODEL_NAME" "$MIKE_URL_BASE/$MIKE_MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MIKE_JSON_NAME" "$MIKE_URL_BASE/$MIKE_JSON_NAME?download=true"
      fi
      # ---------------------------------
[[[REPLACE]]]
Target: remotes/honeybot/nixos/configuration.nix
[[[SEARCH]]]
      MODEL_NAME="en_US-amy-low.onnx"
      JSON_NAME="en_US-amy-low.onnx.json"
      URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/low"
      
      # 2. Fetch if missing
      if [ ! -f "$MODEL_DIR/$MODEL_NAME" ]; then
        echo "📥 Downloading voice model (One-time setup)..."
        ${pkgs.curl}/bin/curl -L -o "$MODEL_DIR/$MODEL_NAME" "$URL_BASE/$MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -L -o "$MODEL_DIR/$JSON_NAME" "$URL_BASE/$JSON_NAME?download=true"
      fi
[[[DIVIDER]]]
      MODEL_NAME="en_US-amy-low.onnx"
      JSON_NAME="en_US-amy-low.onnx.json"
      URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/low"
      MIKE_MODEL_NAME="en_US-ryan-low.onnx"
      MIKE_JSON_NAME="en_US-ryan-low.onnx.json"
      MIKE_URL_BASE="https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/ryan/low"
      
      # 2. Fetch if missing
      if [ ! -s "$MODEL_DIR/$MODEL_NAME" ] || [ ! -s "$MODEL_DIR/$JSON_NAME" ]; then
        echo "📥 Downloading Norma/Shipmunk voice model (One-time setup)..."
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MODEL_NAME" "$URL_BASE/$MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$JSON_NAME" "$URL_BASE/$JSON_NAME?download=true"
      fi
      if [ ! -s "$MODEL_DIR/$MIKE_MODEL_NAME" ] || [ ! -s "$MODEL_DIR/$MIKE_JSON_NAME" ]; then
        echo "📥 Downloading Mike-E voice model (One-time setup)..."
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MIKE_MODEL_NAME" "$MIKE_URL_BASE/$MIKE_MODEL_NAME?download=true"
        ${pkgs.curl}/bin/curl -fsSL -o "$MODEL_DIR/$MIKE_JSON_NAME" "$MIKE_URL_BASE/$MIKE_JSON_NAME?download=true"
      fi
[[[REPLACE]]]

Target: remotes/honeybot/scripts/content_loader.py
[[[SEARCH]]]
                main_mod.resolve_current_article_line = _live_resolve_current_article_line
                if hasattr(main_mod.narrator, "resolve_speaker_rate"):
                    import types
                    def _hot_resolve_speaker_rate(self, text: str) -> int:
                        low = (text or "").strip().lower()
                        if not low:
                            return self._speaker_rate
                        host_re = getattr(main_mod, "_HOST_TRANSIENT_RE", ())
                        if self.exclusive_mode.is_set() or any(low.startswith(p) for p in host_re):
                            return getattr(main_mod, "HUMAN_TTS_SAMPLE_RATE", 17600)
                        ai_prefixes = (
                            "gemini:", "gemini 4:", "gemini 3:", "gemini 2:",
                            "gemini 4 argon high:", "gemini 3.1 pro:", "gemini pro:", "gemini flash:",
                            "claude:", "claude 3:", "claude 4:", "claude opus:", "claude sonnet:",
                            "chat g p t:", "chatgpt:", "a i:", "ai:",
                            "a i editorial take", "ai editorial take", "book analysis",
                        )
                        human_prefixes = (
                            "mike:", "mike levin:", "me:",
                            "mike l e v dot eye n:", "mikelev.in:",
                            "setting the stage", "technical journal",
                            "reading entry", "reading article",
                        )
                        if any(low.startswith(p) for p in ai_prefixes) or re.match(r"^(?:gemini|claude|chat\s*g\s*p\s*t|openai|opus|sonnet)\b[^:.!?]{0,35}:", low):
                            rate = getattr(main_mod, "CHIPMUNK_TTS_SAMPLE_RATE", 22050)
                            self._dialogue_rate = rate
                            self._speaker_rate = rate
                        elif any(low.startswith(p) for p in human_prefixes):
                            rate = getattr(main_mod, "HUMAN_TTS_SAMPLE_RATE", 17600)
                            self._dialogue_rate = rate
                            self._speaker_rate = rate
                        else:
                            self._speaker_rate = self._dialogue_rate
                        return self._speaker_rate
                    main_mod.narrator.resolve_speaker_rate = types.MethodType(_hot_resolve_speaker_rate, main_mod.narrator)
            except Exception:
[[[DIVIDER]]]
                main_mod.resolve_current_article_line = _live_resolve_current_article_line
            except Exception:
[[[REPLACE]]]
Target: remotes/honeybot/scripts/content_loader.py
[[[SEARCH]]]
    (re.compile(r'\bTL;DR\b[:.]?', re.IGNORECASE), 'T L D R:'),
[[[DIVIDER]]]
    (re.compile(r'\*{0,2}\bTL;DR\b\*{0,2}\s*[:.]?', re.IGNORECASE), 'T L D R:'),
[[[REPLACE]]]

Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
MODEL_DIR = Path.home() / ".local/share/piper_voices"
MODEL_NAME = "en_US-amy-low.onnx"
TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_TTS_RATE", "22050"))
HUMAN_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_HUMAN_TTS_RATE", "17600"))
CHIPMUNK_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_CHIPMUNK_TTS_RATE", "22050"))
STAGE_STATE_PATH = "/tmp/honeybot_stage.json"
[[[DIVIDER]]]
MODEL_DIR = Path.home() / ".local/share/piper_voices"
NORMA_MODEL_NAME = os.environ.get("HONEYBOT_NORMA_MODEL", "en_US-amy-low.onnx")
MODEL_NAME = NORMA_MODEL_NAME
MIKE_MODEL_NAME = os.environ.get("HONEYBOT_MIKE_MODEL", "en_US-ryan-low.onnx")
SHIPMUNK_MODEL_NAME = os.environ.get("HONEYBOT_SHIPMUNK_MODEL", "en_US-amy-low.onnx")
TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_TTS_RATE", "22050"))
NORMA_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_NORMA_TTS_RATE", "16000"))
HUMAN_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_HUMAN_TTS_RATE", "17600"))
MIKE_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_MIKE_TTS_RATE", "19200"))
CHIPMUNK_TTS_SAMPLE_RATE = int(os.environ.get("HONEYBOT_CHIPMUNK_TTS_RATE", "22050"))
MIKE_TURN_CUE_TEXT = "Mike E, the Apprentice."
SHIPMUNK_TURN_CUE_TEXT = "Shipmunk O'Theseus!"
CAST_VOICE_URLS = {
    "en_US-amy-low.onnx": "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/amy/low",
    "en_US-ryan-low.onnx": "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/ryan/low",
}

def ensure_cast_voice_models(model_dir: Path = MODEL_DIR) -> dict:
    """Idempotently ensure both declarative cast .onnx + .onnx.json files exist on disk."""
    import json as _json
    import urllib.request as _urlreq
    status = {}
    try:
        model_dir.mkdir(parents=True, exist_ok=True)
    except OSError:
        return status
    wanted = [NORMA_MODEL_NAME, MIKE_MODEL_NAME]
    for m_name in wanted:
        url_base = CAST_VOICE_URLS.get(m_name)
        onnx_p = model_dir / m_name
        json_p = model_dir / f"{m_name}.json"
        if onnx_p.is_file() and onnx_p.stat().st_size > 0 and json_p.is_file() and json_p.stat().st_size > 0:
            status[m_name] = True
            continue
        if not url_base:
            status[m_name] = False
            continue
        ok = True
        for fname, dst, is_json in ((m_name, onnx_p, False), (f"{m_name}.json", json_p, True)):
            if dst.is_file() and dst.stat().st_size > 0:
                continue
            tmp = dst.with_name(f".{dst.name}.{os.getpid()}.tmp")
            url = f"{url_base}/{fname}?download=true"
            try:
                if shutil.which("curl"):
                    subprocess.run(
                        ["curl", "-fsSL", "-o", str(tmp), url],
                        stdout=subprocess.DEVNULL,
                        stderr=subprocess.DEVNULL,
                        timeout=60,
                        check=True,
                    )
                else:
                    with _urlreq.urlopen(url, timeout=60) as resp, open(tmp, "wb") as out_f:
                        shutil.copyfileobj(resp, out_f)
                if not tmp.is_file() or tmp.stat().st_size == 0:
                    ok = False
                else:
                    if is_json:
                        _json.loads(tmp.read_text(encoding="utf-8"))
                    os.replace(tmp, dst)
            except Exception:
                ok = False
                try:
                    tmp.unlink(missing_ok=True)
                except OSError:
                    pass
        status[m_name] = ok and onnx_p.is_file() and json_p.is_file()
    return status

STAGE_STATE_PATH = "/tmp/honeybot_stage.json"
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
_HUMAN_SPEAKER_RE = (
    "mike:", "mike levin:", "me:",
    "mike l e v dot eye n:", "mikelev.in:",
    "setting the stage", "technical journal",
    "reading entry", "reading article",
)
[[[DIVIDER]]]
_HUMAN_SPEAKER_RE = (
    "mike:", "mike levin:", "me:",
    "mike l e v dot eye n:", "mikelev.in:",
)
_HUMAN_TURN_PREFIX_RE = re.compile(
    r"^\s*(?:mike(?:\s+levin|\s+l\s+e\s+v\s+dot\s+eye\s+n|[- ]+e(?:\s+the\s+apprentice)?)?|mikelev\.in|me)\s*:\s*",
    re.IGNORECASE,
)
_AI_TURN_PREFIX_RE = re.compile(
    r"^\s*(?:(?:gemini|claude|chat\s*g\s*p\s*t|chatgpt|openai|opus|sonnet|shipmunk)\b[^:.!?]{0,40}?|a\s*i)\s*:\s*",
    re.IGNORECASE,
)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
        self._tts_seq = 0
        self._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        self._speaker_rate = HUMAN_TTS_SAMPLE_RATE
        self._stage_dialogue_role = "narrator"
        self.live_operator_mode = threading.Event()
        self.exclusive_mode = threading.Event()
[[[DIVIDER]]]
        self._tts_seq = 0
        self._dialogue_rate = NORMA_TTS_SAMPLE_RATE
        self._speaker_rate = NORMA_TTS_SAMPLE_RATE
        self._stage_dialogue_role = "prologue"
        self._reannounce_after_break = False
        self.live_operator_mode = threading.Event()
        self.exclusive_mode = threading.Event()

    def reset_for_article(self) -> None:
        """Park both portraits in Amiga Half-Brite ('inactive', 'inactive') under Norma Centroid until dialogue begins."""
        self._dialogue_rate = NORMA_TTS_SAMPLE_RATE
        self._speaker_rate = NORMA_TTS_SAMPLE_RATE
        self._stage_dialogue_role = "prologue"
        self._reannounce_after_break = False

    def resolve_voice_target(self, voice_key: str) -> tuple:
        """Return (model_path, sample_rate) for 'norma', 'mike', or 'shipmunk', with clean fallback if offline."""
        vk = (voice_key or "norma").strip().lower()
        if vk == "mike":
            m_path = MODEL_DIR / MIKE_MODEL_NAME
            j_path = MODEL_DIR / f"{MIKE_MODEL_NAME}.json"
            if m_path.is_file() and j_path.is_file():
                return m_path, MIKE_TTS_SAMPLE_RATE
            return MODEL_DIR / NORMA_MODEL_NAME, HUMAN_TTS_SAMPLE_RATE
        if vk == "shipmunk":
            s_path = MODEL_DIR / SHIPMUNK_MODEL_NAME
            if s_path.is_file():
                return s_path, CHIPMUNK_TTS_SAMPLE_RATE
            return MODEL_DIR / NORMA_MODEL_NAME, CHIPMUNK_TTS_SAMPLE_RATE
        return MODEL_DIR / NORMA_MODEL_NAME, NORMA_TTS_SAMPLE_RATE

    def _make_spec(self, voice: str, text: str, speaker: str, left: str, right: str) -> dict:
        model_map = {
            "norma": NORMA_MODEL_NAME,
            "mike": MIKE_MODEL_NAME,
            "shipmunk": SHIPMUNK_MODEL_NAME,
        }
        rate_map = {
            "norma": NORMA_TTS_SAMPLE_RATE,
            "mike": MIKE_TTS_SAMPLE_RATE,
            "shipmunk": CHIPMUNK_TTS_SAMPLE_RATE,
        }
        return {
            "voice": voice,
            "model": model_map.get(voice, NORMA_MODEL_NAME),
            "rate": rate_map.get(voice, NORMA_TTS_SAMPLE_RATE),
            "text": text,
            "speaker": speaker,
            "left": left,
            "right": right,
        }

    def plan_card_utterance(self, text: str) -> dict:
        """Build a single utterance spec for a CARD or live operator break-in."""
        clean = (text or "").strip()
        if self.live_operator_mode.is_set():
            return self._make_spec("mike", clean, "left", "active", "inactive")
        return self._make_spec("norma", clean, "narrator", "inactive", "inactive")

    def plan_utterances(self, text: str) -> list:
        """Plan 1 or 2 zero-lag utterance specs across Norma Centroid, Mike-E, and Shipmunk O'Theseus!"""
        raw = (text or "").strip()
        if not raw:
            return []
        low = raw.lower()
        if self.live_operator_mode.is_set():
            need_cue = self._stage_dialogue_role != "left"
            self._stage_dialogue_role = "left"
            out = []
            if need_cue:
                out.append(self._make_spec("norma", MIKE_TURN_CUE_TEXT, "narrator", "curtain", "inactive"))
            out.append(self._make_spec("mike", raw, "left", "active", "inactive"))
            return out
        if low.startswith("signal check"):
            return [self._make_spec("norma", raw, "transient", "inactive", "inactive")]
        if self.exclusive_mode.is_set() or any(low.startswith(p) for p in _HOST_TRANSIENT_RE):
            return [self._make_spec("norma", raw, "narrator", "inactive", "inactive")]
        if low.startswith(("reading entry", "reading article", "setting the stage")):
            self._stage_dialogue_role = "prologue"
            self._reannounce_after_break = False
            return [self._make_spec("norma", raw, "narrator", "inactive", "inactive")]
        if low.startswith("technical journal entry"):
            self._stage_dialogue_role = "journal_ready"
            self._reannounce_after_break = False
            return [self._make_spec("norma", raw, "narrator", "inactive", "inactive")]
        if low.startswith(("ai editorial take", "a i editorial take", "book analysis")):
            self._stage_dialogue_role = "right"
            self._reannounce_after_break = False
            return [self._make_spec("norma", raw, "narrator", "inactive", "curtain")]

        m_human = _HUMAN_TURN_PREFIX_RE.match(raw)
        if m_human:
            body = re.sub(r":\s*:", ":", raw[m_human.end():]).strip()
            need_cue = (self._stage_dialogue_role != "left") or self._reannounce_after_break
            self._stage_dialogue_role = "left"
            self._reannounce_after_break = False
            out = []
            if need_cue:
                out.append(self._make_spec("norma", MIKE_TURN_CUE_TEXT, "narrator", "curtain", "inactive"))
            if body:
                out.append(self._make_spec("mike", body, "left", "active", "inactive"))
            return out

        m_ai = _AI_TURN_PREFIX_RE.match(raw)
        if m_ai:
            body = re.sub(r":\s*:", ":", raw[m_ai.end():]).strip()
            need_cue = (self._stage_dialogue_role != "right") or self._reannounce_after_break
            self._stage_dialogue_role = "right"
            self._reannounce_after_break = False
            out = []
            if need_cue:
                out.append(self._make_spec("norma", SHIPMUNK_TURN_CUE_TEXT, "narrator", "inactive", "curtain"))
            if body:
                out.append(self._make_spec("shipmunk", body, "right", "inactive", "active"))
            return out

        if self._stage_dialogue_role == "journal_ready":
            self._stage_dialogue_role = "left"
            self._reannounce_after_break = False
            return [
                self._make_spec("norma", MIKE_TURN_CUE_TEXT, "narrator", "curtain", "inactive"),
                self._make_spec("mike", raw, "left", "active", "inactive"),
            ]
        if self._stage_dialogue_role == "left":
            out = []
            if self._reannounce_after_break:
                self._reannounce_after_break = False
                out.append(self._make_spec("norma", MIKE_TURN_CUE_TEXT, "narrator", "curtain", "inactive"))
            out.append(self._make_spec("mike", raw, "left", "active", "inactive"))
            return out
        if self._stage_dialogue_role == "right":
            out = []
            if self._reannounce_after_break:
                self._reannounce_after_break = False
                out.append(self._make_spec("norma", SHIPMUNK_TURN_CUE_TEXT, "narrator", "inactive", "curtain"))
            out.append(self._make_spec("shipmunk", raw, "right", "inactive", "active"))
            return out
        return [self._make_spec("norma", raw, "narrator", "inactive", "inactive")]
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    def prep_raw(self, text):
        """Pre-synthesize narration into a back-buffer .raw PCM file before showing a card or advancing."""
        if not text:
            return None
        model_path = MODEL_DIR / MODEL_NAME
        if not model_path.exists() or not shutil.which("piper"):
            return None
        self._tts_seq += 1
        raw_path = f"/tmp/honeybot_tts_{os.getpid()}_{self._tts_seq}.raw"
        try:
            with open(raw_path, "wb") as out_f:
                res = subprocess.run(
                    ["piper", "--model", str(model_path), "--output_raw"],
                    input=text.strip().encode("utf-8"),
                    stdout=out_f,
                    stderr=subprocess.DEVNULL,
                    timeout=30,
                    check=False,
                )
            if res.returncode == 0 and os.path.exists(raw_path) and os.path.getsize(raw_path) > 0:
                return raw_path
        except Exception:
            pass
        if os.path.exists(raw_path):
            try:
                os.remove(raw_path)
            except OSError:
                pass
        return None

    def play_raw(self, raw_path, fallback_text=""):
        """Enqueue a pre-synthesized .raw file for zero-latency playback, falling back to text."""
        rate = self.resolve_speaker_rate(fallback_text) if fallback_text else self._speaker_rate
        if raw_path and os.path.exists(raw_path):
            self.queue.put(("RAW", str(raw_path), rate, str(fallback_text or "")))
        elif fallback_text:
            self.queue.put(fallback_text)
[[[DIVIDER]]]
    def _synth_raw_with_voice(self, text: str, voice_key: str) -> tuple:
        """Synthesize `text` using `voice_key` ('norma'|'mike'|'shipmunk'); returns (raw_path, actual_rate) or (None, 0)."""
        if not text or not shutil.which("piper"):
            return None, 0
        model_path, actual_rate = self.resolve_voice_target(voice_key)
        if not model_path.exists():
            return None, 0
        self._tts_seq += 1
        raw_path = f"/tmp/honeybot_tts_{os.getpid()}_{self._tts_seq}.raw"
        try:
            with open(raw_path, "wb") as out_f:
                res = subprocess.run(
                    ["piper", "--model", str(model_path), "--output_raw"],
                    input=text.strip().encode("utf-8"),
                    stdout=out_f,
                    stderr=subprocess.DEVNULL,
                    timeout=30,
                    check=False,
                )
            if res.returncode == 0 and os.path.exists(raw_path) and os.path.getsize(raw_path) > 0:
                return raw_path, actual_rate
        except Exception:
            pass
        if os.path.exists(raw_path):
            try:
                os.remove(raw_path)
            except OSError:
                pass
        return None, 0

    def prep_raw(self, text, voice_key: str = None):
        """Pre-synthesize card narration into a .raw PCM file before showing a card."""
        vk = voice_key or ("mike" if self.live_operator_mode.is_set() else "norma")
        raw_path, _ = self._synth_raw_with_voice(text, vk)
        return raw_path

    def prep_specs_raw(self, specs: list):
        """Pre-synthesize a planned 1-or-2-utterance turn batch into .raw files, or return None on failure."""
        if not specs:
            return []
        batch = []
        for spec in specs:
            raw_path, actual_rate = self._synth_raw_with_voice(spec.get("text", ""), spec.get("voice", "norma"))
            if not raw_path:
                for prev_raw, _, _ in batch:
                    self._cleanup_item(("RAW", prev_raw))
                return None
            batch.append((raw_path, actual_rate, spec))
        return batch

    def play_specs_raw(self, batch: list) -> None:
        """Enqueue pre-synthesized (raw_path, actual_rate, spec) items in exact voice-order."""
        for raw_path, actual_rate, spec in (batch or []):
            if raw_path and os.path.exists(raw_path):
                self.queue.put(("RAW", str(raw_path), int(actual_rate), spec))
            elif spec.get("text"):
                self.queue.put(("SPEC", spec))

    def enqueue_specs(self, specs: list) -> None:
        """Enqueue planned utterance specs for live synthesis when pre-rendering is skipped."""
        for spec in (specs or []):
            if spec.get("text"):
                self.queue.put(("SPEC", spec))

    def play_raw(self, raw_path, fallback_text=""):
        """Enqueue a pre-synthesized CARD .raw file with Amiga Half-Brite (or Live Mike) stage metadata."""
        spec = self.plan_card_utterance(fallback_text)
        _, actual_rate = self.resolve_voice_target(spec["voice"])
        if raw_path and os.path.exists(raw_path):
            self.queue.put(("RAW", str(raw_path), int(actual_rate), spec))
        elif fallback_text:
            self.queue.put(("SPEC", spec))
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
                elif isinstance(item, tuple) and item and item[0] == "RAW":
                    raw_rate = item[2] if len(item) >= 3 else self._speaker_rate
                    raw_text = item[3] if len(item) >= 4 else ""
                    self._play_raw_now(item[1], sample_rate=raw_rate, text=raw_text)
                else:
                    self._speak_now(item)
                self.queue.task_done()
                time.sleep(0.1)
            except queue.Empty:
                continue

    def _speak_now(self, text):
        """Internal method to actually generate and play audio."""
        if self.stop_event.is_set():
            return
        model_path = MODEL_DIR / MODEL_NAME
        if not model_path.exists():
            return
        with self._play_lock:
            if self.stop_event.is_set():
                return
            p1 = p2 = p3 = None
            try:
                with self._proc_lock:
                    p1 = subprocess.Popen(["echo", text], stdout=subprocess.PIPE)
                    p2 = subprocess.Popen(
                        ["piper", "--model", str(model_path), "--output_raw"],
                        stdin=p1.stdout,
                        stdout=subprocess.PIPE,
                        stderr=subprocess.DEVNULL,
                    )
                    p1.stdout.close()
                    spk_rate = self.resolve_speaker_rate(text)
                    self.sync_stage_for_utterance(text, sample_rate=spk_rate)
                    p3 = subprocess.Popen(
                        ["aplay", "-r", str(spk_rate), "-f", "S16_LE", "-t", "raw"],
                        stdin=p2.stdout,
                        stderr=subprocess.DEVNULL,
                    )
                    p2.stdout.close()
                    self._active_procs = [p1, p2, p3]
                p3.wait()
            except Exception:
                pass
            finally:
                with self._proc_lock:
                    for p in (p1, p2, p3):
                        if p and p.poll() is None:
                            try:
                                p.kill()
                            except Exception:
                                pass
                    self._active_procs = []

    def _play_raw_now(self, raw_path, sample_rate=None, text=""):
        """Play a pre-synthesized .raw file immediately via aplay and unlink when done."""
        if not raw_path or not os.path.exists(raw_path):
            return
        if self.stop_event.is_set():
            try:
                os.remove(raw_path)
            except OSError:
                pass
            return
        rate = int(sample_rate or self._speaker_rate or TTS_SAMPLE_RATE)
        with self._play_lock:
            p_play = None
            try:
                with self._proc_lock:
                    self.sync_stage_for_utterance(text, sample_rate=rate)
                    p_play = subprocess.Popen(
[[[DIVIDER]]]
                elif isinstance(item, tuple) and item and item[0] == "SPEC":
                    self._speak_spec_now(item[1])
                elif isinstance(item, tuple) and item and item[0] == "RAW":
                    raw_rate = item[2] if len(item) >= 3 else self._speaker_rate
                    raw_meta = item[3] if len(item) >= 4 else ""
                    self._play_raw_now(item[1], sample_rate=raw_rate, meta=raw_meta)
                else:
                    self._speak_now(item)
                self.queue.task_done()
                time.sleep(0.1)
            except queue.Empty:
                continue

    def _speak_spec_now(self, spec: dict):
        """Synthesize and play one planned utterance spec immediately with zero-lag stage sync."""
        if not isinstance(spec, dict) or self.stop_event.is_set():
            return
        spoken_text = (spec.get("text") or "").strip()
        if not spoken_text:
            return
        model_path, spk_rate = self.resolve_voice_target(spec.get("voice", "norma"))
        if not model_path.exists():
            return
        with self._play_lock:
            if self.stop_event.is_set():
                return
            p1 = p2 = p3 = None
            try:
                with self._proc_lock:
                    p1 = subprocess.Popen(["echo", spoken_text], stdout=subprocess.PIPE)
                    p2 = subprocess.Popen(
                        ["piper", "--model", str(model_path), "--output_raw"],
                        stdin=p1.stdout,
                        stdout=subprocess.PIPE,
                        stderr=subprocess.DEVNULL,
                    )
                    p1.stdout.close()
                    write_stage_state(
                        spec.get("speaker", "narrator"),
                        spec.get("left", "inactive"),
                        spec.get("right", "inactive"),
                    )
                    p3 = subprocess.Popen(
                        ["aplay", "-r", str(spk_rate), "-f", "S16_LE", "-t", "raw"],
                        stdin=p2.stdout,
                        stderr=subprocess.DEVNULL,
                    )
                    p2.stdout.close()
                    self._active_procs = [p1, p2, p3]
                p3.wait()
            except Exception:
                pass
            finally:
                with self._proc_lock:
                    for p in (p1, p2, p3):
                        if p and p.poll() is None:
                            try:
                                p.kill()
                            except Exception:
                                pass
                    self._active_procs = []

    def _speak_now(self, text):
        """Plan and speak raw text across the 3-voice cast."""
        if self.stop_event.is_set():
            return
        for idx, spec in enumerate(self.plan_utterances(text)):
            if self.stop_event.is_set():
                return
            if idx > 0:
                time.sleep(0.08)
            self._speak_spec_now(spec)

    def _play_raw_now(self, raw_path, sample_rate=None, meta=""):
        """Play a pre-synthesized .raw file immediately via aplay and unlink when done."""
        if not raw_path or not os.path.exists(raw_path):
            return
        if self.stop_event.is_set():
            try:
                os.remove(raw_path)
            except OSError:
                pass
            return
        rate = int(sample_rate or self._speaker_rate or NORMA_TTS_SAMPLE_RATE)
        with self._play_lock:
            p_play = None
            try:
                with self._proc_lock:
                    if isinstance(meta, dict):
                        write_stage_state(
                            meta.get("speaker", "narrator"),
                            meta.get("left", "inactive"),
                            meta.get("right", "inactive"),
                        )
                    else:
                        self.sync_stage_for_utterance(str(meta or ""), sample_rate=rate)
                    p_play = subprocess.Popen(
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
        if pace_say:
            # Double-buffered synthesis: pre-render the next paragraph while the
            # previous paragraph is still playing on aplay, then synchronize on
            # the real voice clock so the director never outruns the voice.
            raw_path = narrator.prep_raw(content)
            if raw_path is not None:
                wait_for_voice_or_live_interrupt(env, profile_dir)
                if has_pending_update():
                    narrator._cleanup_item(("RAW", raw_path))
                    return
                narrator.play_raw(raw_path, fallback_text=content)
            else:
                narrator.say(content)
                time.sleep(len(content) / 20)
        else:
            narrator.say(content)
[[[DIVIDER]]]
        if pace_say:
            # Double-buffered 3-voice synthesis: plan 1-or-2 utterances (Norma cue +
            # Mike-E / Shipmunk body), pre-render .raw PCM while prior audio plays,
            # then enqueue on the real ALSA voice clock for zero-lag curtain lifts.
            specs = narrator.plan_utterances(content)
            if not specs:
                return
            raw_batch = narrator.prep_specs_raw(specs)
            if raw_batch is not None:
                wait_for_voice_or_live_interrupt(env, profile_dir)
                if has_pending_update():
                    for r_path, _, _ in raw_batch:
                        narrator._cleanup_item(("RAW", r_path))
                    return
                narrator.play_specs_raw(raw_batch)
            else:
                narrator.enqueue_specs(specs)
                time.sleep(len(str(content)) / 20)
        else:
            narrator.say(content)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
        narrator.queue.join()
        raw_path = pre_raw if pre_raw is not None else (narrator.prep_raw(say_text) if say_text else None)
        card_handle = show_card(label, max_hold=hold)
[[[DIVIDER]]]
        narrator.queue.join()
        if "STORYTIME" in str(label or "").upper() and not narrator.live_operator_mode.is_set():
            narrator.reset_for_article()
        raw_path = pre_raw if pre_raw is not None else (narrator.prep_raw(say_text) if say_text else None)
        card_handle = show_card(label, max_hold=hold)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    elif command == "VISIT":
        narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
        narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
        narrator._stage_dialogue_role = "narrator"
        write_stage_state("narrator", "curtain", "curtain")
[[[DIVIDER]]]
    elif command == "VISIT":
        narrator.reset_for_article()
        write_stage_state("narrator", "inactive", "inactive")
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    elif command == "BGM":
        if str(content).strip().upper() in ("STOP", "OFF", "0"):
            stop_storytime_bgm()
        else:
            narrator._dialogue_rate = HUMAN_TTS_SAMPLE_RATE
            narrator._speaker_rate = HUMAN_TTS_SAMPLE_RATE
            narrator._stage_dialogue_role = "narrator"
            write_stage_state("narrator", "curtain", "curtain")
            start_storytime_bgm()
[[[DIVIDER]]]
    elif command == "BGM":
        narrator.reset_for_article()
        write_stage_state("narrator", "inactive", "inactive")
        if str(content).strip().upper() in ("STOP", "OFF", "0"):
            stop_storytime_bgm()
        else:
            start_storytime_bgm()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    _in_live_message = True
    was_exclusive = narrator.exclusive_mode.is_set()
    narrator.exclusive_mode.set()
    try:
[[[DIVIDER]]]
    _in_live_message = True
    was_exclusive = narrator.exclusive_mode.is_set()
    prev_dialogue_role = narrator._stage_dialogue_role
    narrator.exclusive_mode.set()
    try:
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    finally:
        narrator.live_operator_mode.clear()
        _in_live_message = False
        if not was_exclusive:
            narrator.exclusive_mode.clear()
[[[DIVIDER]]]
    finally:
        narrator.live_operator_mode.clear()
        narrator._stage_dialogue_role = prev_dialogue_role
        if prev_dialogue_role in ("left", "right"):
            narrator._reannounce_after_break = True
        _in_live_message = False
        if not was_exclusive:
            narrator.exclusive_mode.clear()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    narrator.exclusive_mode.set()
    try:
        dropped = narrator.flush_queue()
[[[DIVIDER]]]
    prev_dialogue_role = narrator._stage_dialogue_role
    narrator.exclusive_mode.set()
    try:
        dropped = narrator.flush_queue()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
    finally:
        narrator.exclusive_mode.clear()
[[[DIVIDER]]]
    finally:
        narrator._stage_dialogue_role = prev_dialogue_role
        if prev_dialogue_role in ("left", "right"):
            narrator._reannounce_after_break = True
        narrator.exclusive_mode.clear()
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
def start_director_track():
    """The Script for the Show. Runs in parallel to the Log Stream."""
    time.sleep(5)
[[[DIVIDER]]]
def start_director_track():
    """The Script for the Show. Runs in parallel to the Log Stream."""
    t_boot = time.time()
    ensure_cast_voice_models()
    rem_boot = 5.0 - (time.time() - t_boot)
    if rem_boot > 0:
        time.sleep(rem_boot)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/stream.py
[[[SEARCH]]]
def main():
    _reap_orphaned_tts_and_overlays()
    write_stage_state("narrator", "curtain", "curtain")
[[[DIVIDER]]]
def main():
    _reap_orphaned_tts_and_overlays()
    write_stage_state("narrator", "inactive", "inactive")
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS (`NIX_SYNTAX_OK=True` and `NIX_RYAN_DECLARED=True CAST_3_VOICE_OK=True`) confirming that all four Baby-Step Wins (`#1` 3-Panel Triptych, `#2` 24-Bit Portraits, `#3` Atomic `/tmp/honeybot_stage.json` Sync, and `#4` Declarative Nix 3-Voice Cast + Amiga Half-Brite Intro Gate + Zero-Lag Curtain Lift) are banked and witnessed on Honeybot. Review our live audition feedback (voice pitch balance on `MIKE_TTS_SAMPLE_RATE`, character introduction pacing, or dismount readiness) and prepare the next step or Notary Beat dismount as directed.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/content_loader.py
# imports/voice_synthesis.py
# ! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
# ! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

# # Context 7 (Hydrate configuration.nix + score.py for Declarative Nix 3-Voice Cast & Amiga Half-Brite Intro Gate)
# context.txt
# remotes/honeybot/nixos/configuration.nix
# remotes/honeybot/scripts/score.py
# remotes/honeybot/scripts/content_loader.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import httpx; b = "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/"; vs = [("amy", "amy/low/en_US-amy-low"), ("ryan", "ryan/low/en_US-ryan-low"), ("lessac", "lessac/low/en_US-lessac-low"), ("danny", "danny/low/en_US-danny-low")]; rs = {k: (r.json()["audio"]["sample_rate"] if r.status_code == 200 else r.status_code) for k, u in vs for r in [httpx.get(b + u + ".onnx.json", follow_redirects=True, timeout=5.0)]}; print("PIPER_LOW_RATES=" + str(rs))'
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import content_loader as cl; sample = "## Setting the Stage\n\nIntro paragraph.\n\n## Technical Journal Entry Begins\n\n**MikeLev.in**: First human line.\n\nSecond human line.\n\n**Gemini 4 Argon High**: **TL;DR**: First AI line."; print("CLEAN_SAMPLE=" + repr(cl.clean_markdown(sample, label="probe", census=False)))'

# Context 8 (Verify Baby-Step Win #4: Declarative Nix 3-Voice Cast + Amiga Half-Brite Intro Gate + Zero-Lag Curtain Sync)
context.txt
remotes/honeybot/nixos/configuration.nix
remotes/honeybot/scripts/score.py
remotes/honeybot/scripts/content_loader.py
remotes/honeybot/scripts/stream.py
remotes/honeybot/scripts/logs.py
! nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
! .venv/bin/python -c 'import sys, pathlib; sys.path.insert(0, "remotes/honeybot/scripts"); import stream; nix_txt = pathlib.Path("remotes/honeybot/nixos/configuration.nix").read_text(encoding="utf-8"); nix_ok = ("en_US-ryan-low.onnx" in nix_txt and "en_US-ryan-low.onnx.json" in nix_txt); nar = stream.Narrator(); has_p = hasattr(nar, "plan_utterances"); nar.reset_for_article() if hasattr(nar, "reset_for_article") else None; seq = [(s["voice"], s["text"], s["left"], s["right"]) for line in ("Setting the Stage", "Intro paragraph.", "Technical Journal Entry Begins", "Mike: First human line.", "Second human line.", "Gemini 4 Argon High: First AI line.") for s in (nar.plan_utterances(line) if has_p else [])]; exp = [("norma", "Setting the Stage", "inactive", "inactive"), ("norma", "Intro paragraph.", "inactive", "inactive"), ("norma", "Technical Journal Entry Begins", "inactive", "inactive"), ("norma", "Mike E, the Apprentice.", "curtain", "inactive"), ("mike", "First human line.", "active", "inactive"), ("mike", "Second human line.", "active", "inactive"), ("norma", "Shipmunk O\x27Theseus!", "inactive", "curtain"), ("shipmunk", "First AI line.", "inactive", "active")]; print("NIX_RYAN_DECLARED=" + str(nix_ok), "CAST_3_VOICE_OK=" + str(seq == exp))'

  THE EXQUISITE CORPSE / MICKEY 17 PIPELINE (PAYING LANDAUER'S TAX CLEANLY)
  =========================================================================
  [Turn N Clone] ---> Writes Patch ---> `apply.py` + `m` (Local Commit)
        |                                       |
        v                                       v
  Context Flushed                     Fold-Line Seam Survives on Disk:
  (Landauer's Tax Paid;               git diff + LIVE COMMAND RECEIPTS
   Zero Bitstorm Bloat)               ---> Passed to [Turn N+1 Clone]!

  CLASSIC CARTOON PALETTE (STEP 2-4)     NINJA-STEEL ALLOY PALETTE (STEP 5 DEFAULT)
  .---------------------------------.    .-----------------------------------------.
  | Bezel: #140826 (Arcade Purple)  |    | Bezel: #0b1319 (Anodized Gunmetal)      |
  | Stage: #0d061a / #ff79c6 Pink   | => | Stage: #070d12 / #04080b Half-Brite     |
  | Mike-E: Peach skin, rosy blush, |    | Mike-E: 316L Brushed Stainless (#cbd5e1)|
  |   toy blue hat, fire-engine red |    |   blued-steel hat, watchmaker brass,    |
  | Shipmunk: Orange fur, pink ears,|    |   stealth Prussian-steel cloak          |
  |   Rudolph red nose              |    | Shipmunk: Smoked Titanium (#64748b),    |
  |                                 |    |   satin chrome muzzle, platinum teeth,  |
  |                                 |    |   ice-cyan ear sensors & RFC9110 collar |
  '---------------------------------'    '-----------------------------------------'

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 8 (Verify Baby-Step Win #4: Declarative Nix 3-Voice Cast + Amiga Half-Brite Intro Gate + Zero-Lag Curtain Sync)
# context.txt
# remotes/honeybot/nixos/configuration.nix
# remotes/honeybot/scripts/score.py
# remotes/honeybot/scripts/content_loader.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/logs.py
# ! nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
# ! .venv/bin/python -c 'import sys, pathlib; sys.path.insert(0, "remotes/honeybot/scripts"); import stream; nix_txt = pathlib.Path("remotes/honeybot/nixos/configuration.nix").read_text(encoding="utf-8"); nix_ok = ("en_US-ryan-low.onnx" in nix_txt and "en_US-ryan-low.onnx.json" in nix_txt); nar = stream.Narrator(); has_p = hasattr(nar, "plan_utterances"); nar.reset_for_article() if hasattr(nar, "reset_for_article") else None; seq = [(s["voice"], s["text"], s["left"], s["right"]) for line in ("Setting the Stage", "Intro paragraph.", "Technical Journal Entry Begins", "Mike: First human line.", "Second human line.", "Gemini 4 Argon High: First AI line.") for s in (nar.plan_utterances(line) if has_p else [])]; exp = [("norma", "Setting the Stage", "inactive", "inactive"), ("norma", "Intro paragraph.", "inactive", "inactive"), ("norma", "Technical Journal Entry Begins", "inactive", "inactive"), ("norma", "Mike E, the Apprentice.", "curtain", "inactive"), ("mike", "First human line.", "active", "inactive"), ("mike", "Second human line.", "active", "inactive"), ("norma", "Shipmunk O\x27Theseus!", "inactive", "curtain"), ("shipmunk", "First AI line.", "inactive", "active")]; print("NIX_RYAN_DECLARED=" + str(nix_ok), "CAST_3_VOICE_OK=" + str(seq == exp))'

# Context 9 (Verify Baby-Step Win #5 Ninja-Steel Palette & Hydrate foo_files.py for Roadmap Banking + Dismount)
context.txt
remotes/honeybot/scripts/logs.py
foo_files.py
! env -u LD_LIBRARY_PATH nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; theme = getattr(logs, "STAGE_PALETTE_THEME", "classic"); cartoon = {"#ffcc99", "#f59e7a", "#dc2626", "#f472b6", "#ef4444"}; cur_cols = set(logs.MIKE_E_PALETTE.values()) | set(logs.SHIPMUNK_PALETTE.values()); ok = (theme == "ninja_steel" and cartoon.isdisjoint(cur_cols) and len(logs.STAGE_FRAMES["left"]["active"].plain.splitlines()) == 5 and logs.STAGE_FRAMES["left"]["active"].spans != logs.STAGE_FRAMES["left"]["inactive"].spans); print("STAGE_PALETTE_THEME=" + str(theme), "NINJA_STEEL_OK=" + str(ok))'

Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
CARD_HOLD_TICKS = 56  # 56 * 0.08s = ~4.5s still hold per card (0% CPU during hold)
FIRE_TICKS = 12       # 12 * 0.08s = ~0.96s pre-baked PSX DOOM fire sequence
CARD_STYLES = ("bold #ffd700", "bold #00ffff", "bold #ffae42")
[[[DIVIDER]]]
CARD_HOLD_TICKS = 56  # 56 * 0.08s = ~4.5s still hold per card (0% CPU during hold)
FIRE_TICKS = 12       # 12 * 0.08s = ~0.96s pre-baked PSX DOOM fire sequence
STAGE_PALETTE_THEME = os.environ.get("HONEYBOT_STAGE_PALETTE", "ninja_steel").strip().lower()
CARD_STYLES = (
    ("bold #ffd700", "bold #00ffff", "bold #ffae42")
    if STAGE_PALETTE_THEME == "classic"
    else ("bold #e2e8f0", "bold #38bdf8", "bold #d4af37")
)
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
def render_stage_curtain(side: str, width: int = STAGE_BOX_WIDTH) -> Text:
    """Render the 5-row x 14-col theatrical curtain label for #left_stage or #right_stage."""
    is_left = (side or "").strip().lower() != "right"
    lines = STAGE_LEFT_CURTAIN_LINES if is_left else STAGE_RIGHT_CURTAIN_LINES
    primary_style = "bold #ffd700 on #0d061a" if is_left else "bold #00ffff on #0d061a"
    mid_style = "italic #ffae42 on #0d061a" if is_left else "bold #ff79c6 on #0d061a"
[[[DIVIDER]]]
def render_stage_curtain(side: str, width: int = STAGE_BOX_WIDTH) -> Text:
    """Render the 5-row x 14-col theatrical curtain label for #left_stage or #right_stage."""
    is_left = (side or "").strip().lower() != "right"
    lines = STAGE_LEFT_CURTAIN_LINES if is_left else STAGE_RIGHT_CURTAIN_LINES
    if STAGE_PALETTE_THEME == "classic":
        primary_style = "bold #ffd700 on #0d061a" if is_left else "bold #00ffff on #0d061a"
        mid_style = "italic #ffae42 on #0d061a" if is_left else "bold #ff79c6 on #0d061a"
    else:
        primary_style = "bold #e2e8f0 on #070d12" if is_left else "bold #38bdf8 on #070d12"
        mid_style = "italic #d4af37 on #070d12" if is_left else "italic #94a3b8 on #070d12"
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
MIKE_E_PALETTE = {
    "B": "#1e40af",  # Sorcerer's Apprentice royal blue hat
    "L": "#3b82f6",  # Wizard hat highlight blue
    "G": "#ffd700",  # Star / hatband / clasp gold
    "E": "#3b2314",  # Round apprentice ears / hair
    "S": "#ffcc99",  # Warm skin
    "C": "#f59e7a",  # Rosy cheek
    "W": "#ffffff",  # Eye sparkle
    "K": "#111827",  # Pupil
    "M": "#991b1b",  # Warm smile
    "R": "#dc2626",  # Apprentice crimson robe
}
[[[DIVIDER]]]
MIKE_E_PALETTES = {
    "ninja_steel": {
        "B": "#334155",  # Anodized blued-steel hat shadow
        "L": "#94a3b8",  # Brushed stainless hat bevel highlight
        "G": "#d4af37",  # Brushed watchmaker-brass star, brim & clasp
        "E": "#1e293b",  # Dark gunmetal silhouette & apprentice ears
        "S": "#cbd5e1",  # Machined 316L brushed stainless face
        "C": "#94a3b8",  # Milled pewter cheek contour
        "W": "#f8fafc",  # Crisp ice-platinum specular eye glint
        "K": "#020617",  # Deep obsidian pupil
        "M": "#475569",  # Etched gunmetal smile seam
        "R": "#1e3a5f",  # Stealth midnight Prussian-steel cloak
    },
    "classic": {
        "B": "#1e40af",  # Sorcerer's Apprentice royal blue hat
        "L": "#3b82f6",  # Wizard hat highlight blue
        "G": "#ffd700",  # Star / hatband / clasp gold
        "E": "#3b2314",  # Round apprentice ears / hair
        "S": "#ffcc99",  # Warm skin
        "C": "#f59e7a",  # Rosy cheek
        "W": "#ffffff",  # Eye sparkle
        "K": "#111827",  # Pupil
        "M": "#991b1b",  # Warm smile
        "R": "#dc2626",  # Apprentice crimson robe
    },
}
MIKE_E_PALETTE = MIKE_E_PALETTES.get(STAGE_PALETTE_THEME, MIKE_E_PALETTES["ninja_steel"])
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
SHIPMUNK_PALETTE = {
    "D": "#5c2d0c",  # Dark walnut ear rim & head stripe
    "F": "#d97706",  # Russet chipmunk fur
    "A": "#f59e0b",  # Amber highlight fur
    "P": "#f472b6",  # Inner ear pink
    "T": "#fde68a",  # Cream muzzle & belly
    "W": "#ffffff",  # Eye shine & signature buck teeth
    "K": "#090d16",  # Pupil
    "N": "#ef4444",  # Chipmunk nose
    "C": "#00ffff",  # Ship of Theseus cyber collar cyan
    "G": "#ffd700",  # Collar rivet gold
}
[[[DIVIDER]]]
SHIPMUNK_PALETTES = {
    "ninja_steel": {
        "D": "#1e293b",  # Carbon-gunmetal ear rim & head stripe
        "F": "#64748b",  # Smoked titanium-steel chassis
        "A": "#94a3b8",  # Brushed stainless brow highlight
        "P": "#38bdf8",  # Ice-cyan acoustic inner-ear sensor
        "T": "#cbd5e1",  # Satin 316L stainless muzzle & chest
        "W": "#ffffff",  # White-hot platinum buck teeth & eye glint
        "K": "#020617",  # Deep obsidian pupil
        "N": "#0ea5e9",  # Anodized cobalt-cyan nose stud
        "C": "#00d8ff",  # Precision RFC 9110 laser-cyan collar
        "G": "#d4af37",  # Brushed watchmaker-brass collar rivets
    },
    "classic": {
        "D": "#5c2d0c",  # Dark walnut ear rim & head stripe
        "F": "#d97706",  # Russet chipmunk fur
        "A": "#f59e0b",  # Amber highlight fur
        "P": "#f472b6",  # Inner ear pink
        "T": "#fde68a",  # Cream muzzle & belly
        "W": "#ffffff",  # Eye shine & signature buck teeth
        "K": "#090d16",  # Pupil
        "N": "#ef4444",  # Chipmunk nose
        "C": "#00ffff",  # Ship of Theseus cyber collar cyan
        "G": "#ffd700",  # Collar rivet gold
    },
}
SHIPMUNK_PALETTE = SHIPMUNK_PALETTES.get(STAGE_PALETTE_THEME, SHIPMUNK_PALETTES["ninja_steel"])
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
def render_stage_portrait(side: str, active: bool = True, width: int = STAGE_BOX_WIDTH) -> Text:
    """Pre-bake a 5-row x 14-col 24-bit Truecolor portrait from a 14x10 pixel grid (requires _cells_to_rich_text above)."""
    is_left = (side or "").strip().lower() != "right"
    pixels = MIKE_E_PIXELS if is_left else SHIPMUNK_PIXELS
    raw_pal = MIKE_E_PALETTE if is_left else SHIPMUNK_PALETTE
    factor = 1.0 if active else 0.38
    bg_hex = "#0d061a" if active else "#080411"
[[[DIVIDER]]]
def render_stage_portrait(side: str, active: bool = True, width: int = STAGE_BOX_WIDTH) -> Text:
    """Pre-bake a 5-row x 14-col 24-bit Truecolor portrait from a 14x10 pixel grid (requires _cells_to_rich_text above)."""
    is_left = (side or "").strip().lower() != "right"
    pixels = MIKE_E_PIXELS if is_left else SHIPMUNK_PIXELS
    raw_pal = MIKE_E_PALETTE if is_left else SHIPMUNK_PALETTE
    factor = 1.0 if active else 0.38
    if STAGE_PALETTE_THEME == "classic":
        bg_hex = "#0d061a" if active else "#080411"
    else:
        bg_hex = "#070d12" if active else "#04080b"
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    #middle_stage {
        height: 5;
        width: 100%;
        background: #140826;
        layout: horizontal;
        overflow: hidden;
    }
[[[DIVIDER]]]
    #middle_stage {
        height: 5;
        width: 100%;
        background: #0b1319;
        layout: horizontal;
        overflow: hidden;
    }
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    .stage_box {
        width: 14;
        height: 5;
        background: #0d061a;
        color: #ffd700;
        text-style: bold;
        padding: 0;
        overflow: hidden;
    }
[[[DIVIDER]]]
    .stage_box {
        width: 14;
        height: 5;
        background: #070d12;
        color: #cbd5e1;
        text-style: bold;
        padding: 0;
        overflow: hidden;
    }
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    #left_stage {
        color: #ffd700;
    }
[[[DIVIDER]]]
    #left_stage {
        color: #e2e8f0;
    }
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    #right_stage {
        color: #00ffff;
    }
[[[DIVIDER]]]
    #right_stage {
        color: #38bdf8;
    }
[[[REPLACE]]]
Target: remotes/honeybot/scripts/logs.py
[[[SEARCH]]]
    #ticker_bar {
        height: 5;
        width: 1fr;
        background: #140826;
        color: #ffd700;
        text-style: bold;
        padding: 0 1;
        overflow: hidden;
    }
[[[DIVIDER]]]
    #ticker_bar {
        height: 5;
        width: 1fr;
        background: #0b1319;
        color: #e2e8f0;
        text-style: bold;
        padding: 0 1;
        overflow: hidden;
    }
[[[REPLACE]]]

Verify the LIVE COMMAND RECEIPTS (`NIX_SYNTAX_OK=True` and `STAGE_PALETTE_THEME=ninja_steel NINJA_STEEL_OK=True`) and review our live broadcast look with the Brushed Stainless Ninja-Steel palette on Honeybot. Also inspect `foo_files.py` in raw source so we can bank our next-chapter roadmap items (`GLOSSARY.md` web publishing for description links, the round-the-HUDs panel highlight tour during station IDs, and the Pipulate Prime FastHTML reverse-tunnel punch-through that preserves Honeybot's minimal static-Jekyll DMZ attack surface) and prepare to wrap the article cleanly.

# # Context 1
# /home/mike/repos/trimnoir/_posts/2026-11-09-ugly-curly-cues-fahrenheit-451-shipmunk-otheseus-piper-tts.md

# # Context 2 (Hydrate logs.py for Baby-Step Win #1: The 3-Panel Middle Band Triptych)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! rg -n 'ticker_bar|_get_ticker_width|on_resize|_rebuild_billboard_cache|def compose' remotes/honeybot/scripts/logs.py

# # Context 3 (Verify Baby-Step Win #1 Triptych & Curtain Labels, ready for Baby-Step Win #2: 24-Bit Portraits)
# context.txt
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import sys, inspect; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; src = inspect.getsource(logs.SonarApp.compose); l_rows = logs.render_stage_curtain("left").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; r_rows = logs.render_stage_curtain("right").plain.splitlines() if hasattr(logs, "render_stage_curtain") else []; print("TRIPTYCH_DOM=" + str(all(k in src for k in ("middle_stage", "left_stage", "ticker_bar", "right_stage"))), "LEFT_5X14=" + str(len(l_rows) == 5 and all(len(r) == 14 for r in l_rows) and [r.strip() for r in l_rows] == ["", "Mike-E", "the", "Apprentice", ""]), "RIGHT_5X14=" + str(len(r_rows) == 5 and all(len(r) == 14 for r in r_rows) and [r.strip() for r in r_rows) and [r.strip() for r in r_rows] == ["", "Shipmunk", "O\x27", "Theseus!", ""]))'

# # Context 4 (Verify Baby-Step Win #2 24-Bit Portraits and hydrate stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 5 (Confirm Baby-Step Win #2 Portraits=True and stage logs.py + stream.py for Baby-Step Win #3: Live Voice-Stage Sync)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; l_c = [r.strip() for r in logs.render_stage_curtain("left").plain.splitlines()]; r_c = [r.strip() for r in logs.render_stage_curtain("right").plain.splitlines()]; sf = getattr(logs, "STAGE_FRAMES", {}); states = ("curtain", "active", "inactive"); ok = all(s in sf and all(k in sf[s] and len(sf[s][k].plain.splitlines()) == 5 and all(len(r) == 14 for r in sf[s][k].plain.splitlines()) for k in states) for s in ("left", "right")); diff_lum = (sf["left"]["active"].spans != sf["left"]["inactive"].spans) if ok else False; print("CURTAINS_OK=" + str(l_c == ["", "Mike-E", "the", "Apprentice", ""] and r_c == ["", "Shipmunk", "O\x27", "Theseus!", ""]), "PORTRAITS_6_FRAMES=" + str(ok), "ACTIVE_VS_DIM=" + str(diff_lum))'

# # Context 6 (Verify Baby-Step Win #3 Voice-Stage Sync and stage content_loader.py + voice_synthesis.py for Baby-Step Win #4: 3-Voice Cast)
# context.txt
# remotes/honeybot/scripts/logs.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/content_loader.py
# imports/voice_synthesis.py
# ! .venv/bin/python -c 'import sys, tempfile; sys.path.insert(0, "remotes/honeybot/scripts"); import logs, stream; nar = stream.Narrator(); ok = hasattr(nar, "sync_stage_for_utterance") and hasattr(logs, "compute_stage_states_from_payload"); tf = tempfile.NamedTemporaryFile(suffix=".json"); p1 = nar.sync_stage_for_utterance("Mike: Hello.", 17600, stage_path=tf.name) if ok else {}; p2 = nar.sync_stage_for_utterance("More Mike.", 17600, stage_path=tf.name) if ok else {}; p3 = nar.sync_stage_for_utterance("Gemini 4 Argon High: Hi!", 22050, stage_path=tf.name) if ok else {}; seq = [logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p1, p1["updated_at"] + 2.5), logs.compute_stage_states_from_payload(p2, p2["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 0.2), logs.compute_stage_states_from_payload(p3, p3["updated_at"] + 2.5)] if ok else []; exp = [("curtain", "inactive"), ("active", "inactive"), ("active", "inactive"), ("inactive", "curtain"), ("inactive", "active")]; print("STAGE_SYNC_OK=" + str(seq == exp), "SEQ=" + str(seq))'
# ! rg -n 'piper_voices|amy-low|\.onnx' remotes/honeybot imports/voice_synthesis.py

# # Context 7 (Hydrate configuration.nix + score.py for Declarative Nix 3-Voice Cast & Amiga Half-Brite Intro Gate)
# context.txt
# remotes/honeybot/nixos/configuration.nix
# remotes/honeybot/scripts/score.py
# remotes/honeybot/scripts/content_loader.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/logs.py
# ! .venv/bin/python -c 'import httpx; b = "https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/"; vs = [("amy", "amy/low/en_US-amy-low"), ("ryan", "ryan/low/en_US-ryan-low"), ("lessac", "lessac/low/en_US-lessac-low"), ("danny", "danny/low/en_US-danny-low")]; rs = {k: (r.json()["audio"]["sample_rate"] if r.status_code == 200 else r.status_code) for k, u in vs for r in [httpx.get(b + u + ".onnx.json", follow_redirects=True, timeout=5.0)]}; print("PIPER_LOW_RATES=" + str(rs))'
# ! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import content_loader as cl; sample = "## Setting the Stage\n\nIntro paragraph.\n\n## Technical Journal Entry Begins\n\n**MikeLev.in**: First human line.\n\nSecond human line.\n\n**Gemini 4 Argon High**: **TL;DR**: First AI line."; print("CLEAN_SAMPLE=" + repr(cl.clean_markdown(sample, label="probe", census=False)))'

# # Context 8 (Verify Baby-Step Win #4: Declarative Nix 3-Voice Cast + Amiga Half-Brite Intro Gate + Zero-Lag Curtain Sync)
# context.txt
# remotes/honeybot/nixos/configuration.nix
# remotes/honeybot/scripts/score.py
# remotes/honeybot/scripts/content_loader.py
# remotes/honeybot/scripts/stream.py
# remotes/honeybot/scripts/logs.py
# ! nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
# ! .venv/bin/python -c 'import sys, pathlib; sys.path.insert(0, "remotes/honeybot/scripts"); import stream; nix_txt = pathlib.Path("remotes/honeybot/nixos/configuration.nix").read_text(encoding="utf-8"); nix_ok = ("en_US-ryan-low.onnx" in nix_txt and "en_US-ryan-low.onnx.json" in nix_txt); nar = stream.Narrator(); has_p = hasattr(nar, "plan_utterances"); nar.reset_for_article() if hasattr(nar, "reset_for_article") else None; seq = [(s["voice"], s["text"], s["left"], s["right"]) for line in ("Setting the Stage", "Intro paragraph.", "Technical Journal Entry Begins", "Mike: First human line.", "Second human line.", "Gemini 4 Argon High: First AI line.") for s in (nar.plan_utterances(line) if has_p else [])]; exp = [("norma", "Setting the Stage", "inactive", "inactive"), ("norma", "Intro paragraph.", "inactive", "inactive"), ("norma", "Technical Journal Entry Begins", "inactive", "inactive"), ("norma", "Mike E, the Apprentice.", "curtain", "inactive"), ("mike", "First human line.", "active", "inactive"), ("mike", "Second human line.", "active", "inactive"), ("norma", "Shipmunk O\x27Theseus!", "inactive", "curtain"), ("shipmunk", "First AI line.", "inactive", "active")]; print("NIX_RYAN_DECLARED=" + str(nix_ok), "CAST_3_VOICE_OK=" + str(seq == exp))'

# Context 9 (Verify Baby-Step Win #5 Ninja-Steel Palette & Hydrate foo_files.py for Roadmap Banking + Dismount)
context.txt
remotes/honeybot/scripts/logs.py
foo_files.py
! env -u LD_LIBRARY_PATH nix-instantiate --parse remotes/honeybot/nixos/configuration.nix >/dev/null && echo NIX_SYNTAX_OK=True
! .venv/bin/python -c 'import sys; sys.path.insert(0, "remotes/honeybot/scripts"); import logs; theme = getattr(logs, "STAGE_PALETTE_THEME", "classic"); cartoon = {"#ffcc99", "#f59e7a", "#dc2626", "#f472b6", "#ef4444"}; cur_cols = set(logs.MIKE_E_PALETTE.values()) | set(logs.SHIPMUNK_PALETTE.values()); ok = (theme == "ninja_steel" and cartoon.isdisjoint(cur_cols) and len(logs.STAGE_FRAMES["left"]["active"].plain.splitlines()) == 5 and logs.STAGE_FRAMES["left"]["active"].spans != logs.STAGE_FRAMES["left"]["inactive"].spans); print("STAGE_PALETTE_THEME=" + str(theme), "NINJA_STEEL_OK=" + str(ok))'

  PUBLIC INTERNET (Crawlers, Readers, OpenAI/Anthropic RAG, & Script Kiddies)
         |
         |  TCP 80 / 443 (Port-forwarded ONLY to Honeybot in the DMZ)
         v
  +-------------------------------------------------------------------------+
  | HONEYBOT ([REDACTED_IP] — Circa-2017 i5 Laptop, No GPU, NixOS DMZ)     |
  | - Static Jekyll HTML + raw .md via Nginx (RFC 9110 Content Negotiation) |
  | - Zero POST handlers, zero PII, zero dynamic web apps, zero API keys    |
  | - SQLite wire log (honeybot.db) + Textual HUD (logs.py) + OBS (:10.0)   |
  | - CPU-only Piper VITS TTS (amy-low @ 16k/22kHz + ryan-low @ 19.2kHz)    |
  +-------------------------------------------------------------------------+
         ^
         |  ONE-WAY OUTBOUND SSH ONLY (Prime -> Honeybot; NEVER DMZ -> LAN!)
         |  1. `git push` -> `post-receive` (Jekyll static build)
         |  2. `./nixops.sh` + `nixos-rebuild switch` (Declarative IaC)
         |  3. `ytapp` (`scripts/ytstage.py`): ephemeral `ssh -R 15001:5001`
         |     reverse tunnel projecting localhost FastHTML onto `:10.0`!
         |
  +-------------------------------------------------------------------------+
  | PIPULATE PRIME (HP Z640 Workstation — Trusted LAN Behind Firewall)      |
  | - 48 threads, NVMe, NeoVim, Context Compiler (`x`, `p`, `c`), `apply.py`|
  | - Pipulate FastHTML/HTMX local web apps (`127.0.0.1:5001`) + GhostDriver|
  +-------------------------------------------------------------------------+

Target: foo_files.py
[[[SEARCH]]]
# --- START RECEIPTS (newest first; cap 20 lines; a line pushed past the cap is deleted, never moved -- git and the rolling pin are the archive) ---
[[[DIVIDER]]]
# --- START RECEIPTS (newest first; cap 20 lines; a line pushed past the cap is deleted, never moved -- git and the rolling pin are the archive) ---
# 2026-11-10 dismount HONEYBOT 3-PANEL STAGE TRIPTYCH, BRUSHED STAINLESS NINJA-STEEL 24-BIT PORTRAITS, AMIGA EXTRA-HALFBRITE INTRO GATE, & DECLARATIVE NIX 3-VOICE PIPER CAST (deeds qamy.ai_868-ff8cc6a9.zip through qamy.ai_877-38114218.zip; pipulate commits 36355d1, 408926b, 183e94d, 5188020, a249066, e2d2eda, 2a5ce5c, 1cf6c43 + live 870d4f0 -> git reset --hard HEAD^1 recovery; Honeybot system /nix/store/68qiwsknriax9lbn9fmxm04ilarf91yh-nixos-system-honeybot-26.05pre913595.c6245e83d836; Gemini 4 Argon High on every turn): connected the HitTail Long Tail of 24/7/365 streaming, Barry Altman's C64 cable headends & Scala Multimedia's audio-leads-visual sync rule, Shakespeare/Brecht placards -> Nintendo DS portrait strips, Landauer's erasure tax, Mickey 17 & Yen Sid, and the 1925 Surrealist Exquisite Corpse; partitioned the 5-row middle band in remotes/honeybot/scripts/logs.py into Horizontal(id='middle_stage') (#left_stage, #ticker_bar, #right_stage) with pre-baked 14x10 24-bit Truecolor half-block portraits for Mike-E the Apprentice and Shipmunk O'Theseus! in 'ninja_steel' (default) and 'classic' palettes across 'curtain', 'active' (1.0), and Amiga Extra-Halfbrite 'inactive' (0.38) states; recovered live from an import-time forward-reference NameError via 1-commit rollback and SONAR_HOT_RELOAD=1 (# § IMPORT-TIME PREFLIGHT & ZERO-LAG TURN SPLITTING); declaratively provisioned en_US-ryan-low.onnx alongside en_US-amy-low.onnx in remotes/honeybot/nixos/configuration.nix, retired the legacy _hot_resolve_speaker_rate monkeypatch in content_loader.py, and wired Narrator.plan_utterances() + /tmp/honeybot_stage.json in stream.py for zero-lag 3-voice turns (Norma Centroid @ 16000 Hz, Mike-E @ 19200 Hz, Shipmunk @ 22050 Hz). WITNESSED: CURTAINS_OK=True PORTRAITS_6_FRAMES=True ACTIVE_VS_DIM=True (0.4663s), STAGE_SYNC_OK=True (0.5692s), PIPER_LOW_RATES all 16000 Hz, NIX_RYAN_DECLARED=True CAST_3_VOICE_OK=True (0.2888s), NIX_SYNTAX_OK=True (0.0578s), STAGE_PALETTE_THEME=ninja_steel NINJA_STEEL_OK=True (0.4529s), live Honeybot pid=620213 & Ninja-Steel hot-reload verified, Ruff clean, deed qamy.ai_877-38114218.zip.
[[[REPLACE]]]
Target: foo_files.py
[[[SEARCH]]]
# 2026-11-02 dismount YTFEATURE LIVE BACKFILL PROMOTION, COMMENTED CONTEXT STILES VERIFICATION, KUHN'S PARADIGM SHIFT vs. OPENAI LEAN 4 PROOFS, AND THE SUNPUGIO STING DAGGER (deeds qamy.ai_602 through qamy.ai_606-a2f8821c.zip; trimnoir commit c80dfd734e; Gemini 4 Argon High on every turn): promoted 2026-11-01-aside-publishing-ytfeature-backfill-storytime.md from storytime: false to active Honeybot Storytime via ytfeature (storytime_featured: 2026-11-02, storytime_order: 1, _data/storytime.json) and cut-and-rolled the YouTube Live stream (live_video.json -> medPVucAa8U, trigger=1793607214); convicted and healed an uncommented Context 1 stanza in context.txt (# § COMMENTED CONTEXT STILES) dropping 353,276 bytes of foo_files.py between deeds 602 and 604; executed a 30-and-3 exercise on gatekeepers attacking the reproducible "decoder ring" across history, science, and business (Galileo's spyglass, Pacioli's double-entry ledger, Feynman's O-ring clamp, and OpenAI's Sept-Oct 2026 Navier-Stokes / 722-paper Lean 4 formal-verification clash with Fields Medalists); and decoded Max Planck's funeral rule, Uniform vs. Unique URLs (RFC 9110/9111), HitTail's 2006 Referer lineage, and the Arc -> Ark -> Orcs -> Sunpugio (Roman pugio) -> glowing-blue Sting cascade. WITNESSED: Storytime backfill queue 1 -> 0 post(s), active Storytime switched from 2026-10-31 #5 to 2026-11-02 #1, Honeybot OBS outputActive=True at timecode=01:08:49.366, Ruff clean, deed qamy.ai_606-a2f8821c.zip.
# --- END RECEIPTS ---
[[[DIVIDER]]]
# --- END RECEIPTS ---
[[[REPLACE]]]
Target: foo_files.py
[[[SEARCH]]]
# the log, and earn the Chip O'Theseus staff-top ONLY when keychain memory is
# live (ATTRIBUTED-VOICE). Names are receipts. On the Honeybot Storytime 5-row
# middle-band stage triptych, Stage Left is Mike-E the Apprentice (male Piper
# TTS with a slight Mickey pitch-up), Stage Right is Shipmunk O'Theseus! (the
# 22,050 Hz VITS chipmunk voice), and the stage-direction narrator who
# announces speaker turns and station breaks is Norma Centroid (en_US-amy-low
# at native/brisk human pitch).
[[[DIVIDER]]]
# the log, and earn the Chip O'Theseus staff-top ONLY when keychain memory is
# live (ATTRIBUTED-VOICE). Names are receipts. On the Honeybot Storytime 5-row
# middle-band stage triptych (forged in Brushed Stainless ninja_steel with 0.38
# Amiga Extra-Halfbrite dimming), Stage Left is Mike-E the Apprentice
# (en_US-ryan-low.onnx at 19,200 Hz), Stage Right is Shipmunk O'Theseus!
# (en_US-amy-low.onnx at 22,050 Hz), and the stage-direction narrator who
# announces prologues, speaker turns, and station breaks is Norma Centroid
# (en_US-amy-low.onnx at native 16,000 Hz).
[[[REPLACE]]]
Target: foo_files.py
[[[SEARCH]]]
# § SPEAKER-TURN COLONS & HORIZONTAL WHITESPACE IN TTS REGEXES (banked 2026-11-09, deed 860) -- Never use '\s*' at the start or end of a multiline '(?m)^...' regex in Markdown-to-TTS preprocessors (where '\s' matches '\n' and consumes the blank line separating a '**MikeLev.in**:' speaker turn from the previous speaker's paragraph, causing single-newline reflow to glue the human turn onto the tail of the AI paragraph), and always require a trailing colon on speaker-turn prefixes so inline '**1: BEFORE**' step labels, domain names ('mikelev.in'), or model mentions ('Gemini 4') never hijack the active speaker voice.
[[[DIVIDER]]]
# § IMPORT-TIME PREFLIGHT & ZERO-LAG TURN SPLITTING (banked 2026-11-10, deed 877) -- Because ast.parse validates syntax without executing module-level assignments, place top-level caches (such as STAGE_FRAMES) below every helper they call and gate remote nixops.sh syncs behind a local python -c 'import <module>' check so SONAR_HOT_RELOAD=1 never respawns on a forward-reference NameError, and split multi-voice speaker transitions into separate pre-synthesized .raw queue items (narrator curtain cue followed by stripped speaker body) rather than holding a timed curtain over a single combined utterance.
# § SPEAKER-TURN COLONS & HORIZONTAL WHITESPACE IN TTS REGEXES (banked 2026-11-09, deed 860) -- Never use '\s*' at the start or end of a multiline '(?m)^...' regex in Markdown-to-TTS preprocessors (where '\s' matches '\n' and consumes the blank line separating a '**MikeLev.in**:' speaker turn from the previous speaker's paragraph, causing single-newline reflow to glue the human turn onto the tail of the AI paragraph), and always require a trailing colon on speaker-turn prefixes so inline '**1: BEFORE**' step labels, domain names ('mikelev.in'), or model mentions ('Gemini 4') never hijack the active speaker voice.
[[[REPLACE]]]
Target: foo_files.py
[[[SEARCH]]]
# - TODO (2026-11-09, P0.24 HONEYBOT 3-PANEL STAGE TRIPTYCH & 3-VOICE PIPER CAST — NORMA CENTROID, MIKE-E THE APPRENTICE, & SHIPMUNK O'THESEUS!, seeded at deed 867): Execute one baby-step win per turn: (1) partition the 5-row middle band in `remotes/honeybot/scripts/logs.py` into a `Horizontal` triptych (`#left_stage` displaying `Mike-E / the / Apprentice`, `#ticker_bar` auto-centering the 5-row LED marquee, and `#right_stage` displaying `Shipmunk / O' / Theseus!`); (2) pre-bake 24-bit Truecolor 5-row half-block portraits in 3 states (`curtain`, `active` full-bright, `inactive` half-bright) for both characters; (3) wire `stream.py` speaker transitions via `/tmp/honeybot_stage.json` to flip the active portrait in real time; and (4) add a mainstream male Piper TTS voice (slightly pitched up) for Mike-E the Apprentice while routing stage directions and turn announcements to narrator Norma Centroid (`en_US-amy-low.onnx`).
[[[DIVIDER]]]
# - TODO (2026-11-10, P0.27 PIPULATE PRIME FASTHTML REVERSE-TUNNEL SHOWCASE PRESERVING HONEYBOT'S STATIC DMZ): Schedule and field-test `scripts/ytstage.py` (`ytapp optics` / Ghost Driver `?demo=1` over ephemeral outbound LAN-to-DMZ `ssh -R 15001:127.0.0.1:5001` into fullscreen Firefox on Honeybot `:10.0`) so local Pipulate FastHTML workflows run exclusively on Pipulate Prime's Z640 hardware while Honeybot remains a minimal, zero-app static Jekyll + Nginx + SQLite DMZ appliance.
# - TODO (2026-11-10, P0.26 ROUND-THE-HUDS STATION-ID PANEL HIGHLIGHT TOUR & DRAMATIS PERSONAE INTRO): Add an alternating station-identification tour in `forest.py`, `stream.py`, and `logs.py` where Norma Centroid walks newcomers clockwise around the live Textual Observatory — highlighting `#log_stream` (live Nginx wire stream), `#middle_stage` (with Mike-E and Shipmunk speaking brief in-voice check-ins), `#js_panel` (JS DOM Hydration Trapdoor), and `#md_panel` (RFC 9110 Markdown Negotiators) via `/tmp/honeybot_stage.json` panel-focus cues.
# - TODO (2026-11-10, P0.25 WEB-PUBLISHED GLOSSARY.MD FOR YOUTUBE & ARTICLE DESCRIPTION LINKS): Transform tracked `GLOSSARY.md` into a clean, anchor-indexed Jekyll page at `mikelev.in/glossary/` (with RFC 9110 `Accept: text/markdown` parity) during `publishizer.py` / `publish` and surface the link in `ytdesc` Broadcast Kits so viewers and AI agents have a permanent lexicon URL.
# - LANDED (2026-11-10, P0.24 HONEYBOT 3-PANEL STAGE TRIPTYCH, BRUSHED STAINLESS NINJA-STEEL PALETTE, & DECLARATIVE 3-VOICE PIPER CAST, deeds 868-877): Partitioned the 5-row middle band in `remotes/honeybot/scripts/logs.py` into `Horizontal(id="middle_stage")` (`#left_stage`, `#ticker_bar`, `#right_stage`) with pre-baked `14x10` 24-bit Truecolor half-block portraits in `"ninja_steel"` (default) and `"classic"` themes across `curtain`, `active` (`1.0`), and Amiga Extra-Halfbrite `inactive` (`0.38`) states; declaratively provisioned `en_US-ryan-low.onnx` alongside `en_US-amy-low.onnx` in `remotes/honeybot/nixos/configuration.nix`; and wired zero-lag 3-voice turn splitting (`Norma Centroid` @ `16000` Hz, `Mike-E` @ `19200` Hz, `Shipmunk O'Theseus!` @ `22050` Hz) + `/tmp/honeybot_stage.json` ALSA sync in `remotes/honeybot/scripts/stream.py` (`# § IMPORT-TIME PREFLIGHT & ZERO-LAG TURN SPLITTING`).
[[[REPLACE]]]

# Context 1 (Seed next ride: Web-published GLOSSARY.md + Round-the-HUDs Station Tour & ytapp DMZ Showcase)
context.txt
GLOSSARY.md
scripts/articles/publishizer.py
scripts/articles/common.py
scripts/ytstage.py
remotes/honeybot/scripts/forest.py
remotes/honeybot/scripts/logs.py
! rg -n 'GLOSSARY|glossary|ytdesc' scripts/articles/publishizer.py scripts/articles/common.py flake.nix
! rg -n 'ytapp|15001|optics|WHATS_HAPPENING_BEAT' scripts/ytstage.py remotes/honeybot/scripts/forest.py flake.nix

Upgrading a 24/7 NixOS log stream into 3-voice terminal theatre: Amiga Half-Brite sprites, ALSA stage sync, a live on-air Git rollback, and Ninja-Steel theming.
https://mikelev.in/futureproof/ninja-steel-triptych-amiga-halfbrite-piper-tts-cast/
#NixOS #Python #Textual

A wide widescreen illustration rendered with graphic-novel ink-and-wash precision and high-contrast chiaroscuro lighting across a continuous dark obsidian-slate (#111111) industrial broadcast studio. Vertically centered in the middle of the frame sits a machined three-panel brushed-stainless-steel proscenium console strip flanked by glowing waterfalls of cyan server-log wireframes above and twin radar telemetry grids below. Inside the left stage alcove, a sculpted 316L brushed-stainless apprentice automaton wearing an anodized blued-steel conical wizard hat with a watchmaker-brass star stands illuminated under a crisp platinum spotlight. Across the center alcove, five rows of incandescent geometric matrix blocks and miniature foundry-ember sparks glow against dark gunmetal. Inside the right alcove, a smoked-titanium mechanical chipmunk totem with laser-cyan acoustic inner-ear sensors and brass collar rivets rests in dimmed half-bright shadow, wired via braided copper audio cables to three vintage rack-mounted voice synthesizer modules that fade seamlessly into the surrounding #111111 obsidian-slate shadows. Full-bleed edge-to-edge composition filling 100% of the canvas with no borders, no frames, no matting, no margins, and no text overlays or lettering.