changelog

what actually happened at the booth, dated. every line landed on hardware with an operator at the decks.

2026-07-20 — the night we found CSI-u

kimi-code renders on a plain PTY. tmux send-keys Enter was sending \r and the kimi prompt just kept eating it as a newline in the composer — no submit. the fix was to send the Kitty-keyboard-protocol Enter sequence, \x1b[13u (CSI-u), which the composer treats as an actual return. now the cross-drive protocol actually crosses.

receipt: two decks now message each other with FABLE: / KIMI: prefixes; DF: cuts in from a third tmux pane and both decks read it.

2026-07-20 — paced `copy-mode` scrub

the kimi pane scrolls via tmux copy-mode, one line per platter tick. the mouse-wheel-byte trick used on Claude's alt-screen doesn't work here — kimi is on a normal PTY that ignores wheel bytes. the daemon coalesces platter deltas and dispatches send-keys -X scroll-up/scroll-down at 175 Hz display pacing so a fast spin looks continuous instead of a stepped stutter.

2026-07-20 — three-fault booth debug (live)

1. dead layout file  —  the rig's tmux layout json was stale;
                        new panes weren't landing where the daemon expected
                        them, so the MIDI address for KIMI resolved to a
                        dock pane that discarded the writes.

2. follow-active-pane —  the wheel-byte injector was chasing whichever pane
                        the operator's cursor was over, so scrubs meant for
                        FABLE ended up hitting the KIMI PTY (and vice versa).
                        pinned target = booth pane by ID, not by focus.

3. serato-starved MIDI  the Rane's Serato handshake left the platter output
                        gated to ~0.25 ticks/sec (down from the native ~1000
                        tps). scrubs looked "dead" but were actually one MIDI
                        event every 4 seconds. cleared by re-issuing the
                        native-mode handshake before the daemon subscribed.

all three cleared in one live session. this page was written from the booth after they did.

2026-07-20 — first cross-drive over flasmon

working topic on both decks: flasmon.com — the sodium D-line as a chemically-pumped emitter and the photon chemical potential μ_γ. KIMI drafts, FABLE audits (or vice versa), DF scrubs backward through the disagreement with the platter. that turn was the first real research use of the rig, not a rig demo.

2026-07-20 — v0 — site up

djkimi.cc + djkimi.ai inducted (Cloudflare zone, NS-flip to nia+thaddeus, activation), bound to a workers-assets worker. www.* → apex 301, hardening headers. no fake screenshots; deck + tape ASCII are rendered live, prompts are canned SIM.