feat(controller): Chief-of-Staff overnight keep-alive — SQLite queue drained by the launchd tick

Operator: "make sure this does not die after one or two iterations… use
sqlite… go all night, space out 10 revisions… check the tools first."
A session cron dies with the session; the durable substrate already runs
24/7 — local SQLite + the launchd controller tick. So the keep-alive is
STAGE 8 of the tick, draining a cos_overnight queue.

Per tick: MATCH ($0 SQL due-check, exit before any model touch) → FIRE
(one due vector, claude -p pinned to sonnet, wrapped in a perl-alarm wall
clock, default 540s) → MARK (iter++ in SQL; requeue +45m or done at
max_iter). Two-sided bound: won't die (SQLite + always-on tick survive
session death) and won't run away (model-pinned + time-bounded +
max_iter + single-flight + $0-idle — the $665 and receipt-backfill
lessons both encoded). Standing boundary in every fired prompt: refine +
grade >=95 + commit + STAGE; the only autonomous send allowed is a
self-channel milestone to elias@; NO cold external sends overnight.

Six vectors seeded (milestone-strategy, intro-scan, dinner-cigna-edit,
stage-cigna-sends, dinner-roster-venue, jai-followup@+8h), staggered.

Readiness verified: controller loaded+ticking, rust runner present, queue
created, dry-run matched the due vector, alarm wrapper killed a 30s hang
in 2s. Guards (same commit): cos-overnight-keepalive.test.js (durable +
bounded invariants) + cos-overnight.sh added to the model-pin allowlist.
14/14 green. Spec: docs/specs/cos-overnight-keepalive-2026-07-05.md.

Originating-Terminal: 🎩 Kitty Operator
Relevant-Rooms: 🧭 navigator, 📐 architect, 🎩 operator
Story: The overnight CoS run needed a heart that keeps beating after the session closes; the ratified local-first substrate (SQLite + launchd) is that heart, and this stage is the pacemaker — check the tools first, build on what already runs, guard both failure directions.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VRmFxPtHxYBiXRjiqRT54T

commit e0aecdb3a — 144×144 lattice triptych (on-chip)

◎ the ingest chose B1,B3 — Tactics·Speed × Tactics·Signal — acting on A3,B3 — Strategy·Fund × Tactics·Signal (grip 0.781) · the walk starts at the ACTOR's row · 10 file(s)

⏱ ingest 597.9ms (corpus→lattice) · definer walk 1182ms (row→transpose→row, ballistic XOR per hop) · total 1963ms
ingest here = commit-scoped SENSING only (this commit's message + changed files → lit anchors) — deep seed authoring lives in the reef-self-loop, off the commit path

rows = ACTOR (lens) · cols = PATIENT (object) · diagonal = self-reference; ◎ at actor-row × patient-col — every panel, same orientation. Heat above the diagonal = actors ranking ShortLex-earlier than their patients (the walk follows definers uphill in rank).
The 12 canonical lanes (rows top→bottom, cols left→right): A·Strategy · B·Tactics · C·Operations · A1·Law · A2·Goal · A3·Fund · B1·Speed · B2·Deal · B3·Signal · C1·Grid · C2·Loop · C3·Flow

PAIR LATTICE (the map we normally see)

PRE-WALK SENSE · INTENT grid — the snapshot, no point of view

the no-point-of-view snapshot — what the sensor lit before any walk; the leaf walk below is what bridges the two grids

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK SENSE · REALITY grid — the snapshot, no point of view

the no-point-of-view snapshot, reality side — same sensor law

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK Δ · straight comparison of the two ingested grids (green both · cyan intent-only · amber reality-only) · 4% overlap · ◎ = where the competence pixel sits in the raw overlap

good = green overlap without being a copy — near-100% is self-similarity, not alignment

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

◎ the pixel this commit belongs to

B1,B3 — Tactics·Speed × Tactics·Signal (actor) acting on A3,B3 — Strategy·Fund × Tactics·Signal (patient) · grip 0.781

actor seed B1,B3 — Measured velocity beats determine how fast the message signal can reach the intended broadcast channel bandwidth. Signal-to-noise is the inverse cube of bandwidth — narrow the channel, the message arrives; broaden it, recipients tune out. The right channel is the one whose floor matches the message's importance. A broadcast that addresses everyone addresses no one. Precision over reach until the address is calibrated; only then does scale convert prospects into recipients rather than into unsubscribes. Latency is the price of a missed beat. Every hundred milliseconds of round-trip turns a parallel race into a stalemate, because the first arrival sets the rules the second arrival has to honor. Throughput compounds: doubling the rate doubles the surface area where serendipity can hit. A pipeline at twice the velocity is not twice as productive — it is exponentially more lucky.

patient seed A3,B3 — Financial budget runway funds the signal bandwidth and message reach of the intended broadcast channel. Every dollar builds either the floor (substrate that lifts future bets) or the ceiling (cap on the next round). Allocation is which one this dollar chooses, and the choice is irreversible by the time the wire clears. Burn rate divides into substrate-building and exhaust. Substrate compounds; exhaust evaporates by quarter-end. The investor who pays for substrate is buying a different asset than the investor who pays for exhaust. Capital allocation is the irreversible vote — the bet that closes options to keep the others open. Each dollar deployed buys both the asset and the foreclosure of every alternative the dollar could have funded. Signal-to-noise is the inverse cube of bandwidth — narrow the channel, the message arrives; broaden it, recipients tune out. The right channel is the one whose floor matches the message's importance. A broadcast that addresses everyone addresses no one. Precision over reach until the address is calibrated; only then does scale convert prospects into recipients rather than into unsubscribes. Repetition is the cost of being heard. Three sends to one recipient beats one send to three; the first send announces, the second send is heard, the third send is remembered.

INTENT · leaf-walk clouds · BLUE→VIOLET by ply (◎ = the competence pixel: actor∩patient)

good = the clouds concentrate where the commit SAYS it works

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

REALITY · leaf-walk clouds · ORANGE→RED by ply (◎ = the same pixel — the shared perspective)

good = the same clouds as intent — shipped where declared

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

DELTA · graded cloud divergence (green agree · magenta declared-not-done · amber done-not-declared) · 2% agree

good = mostly green; magenta = said-not-done (chase it in the code), amber = done-not-said (chase it in the docs)

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

TOLERANCE · cloud topology (dim green in-shape · hot amber bleed · red orthogonal concentration)

red only matters above the flip — a few amber is normal bleed

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

SHORTLEX-3 PROJECTION (ABC · A1..C3 · the 132 children — candidate, pre-ratchet)

PROJECTED INTENT · the commit’s claims, head→axis_i tail→axis_j (cyan)

the three-length view — good = claims land inside their own zone

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PROJECTED REALITY · same projection law (amber)

same projection law — good = reality occupies the zones intent does

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

STRAIGHT COMPARISON · green both · cyan intent-only · amber reality-only · 4% overlap

good = the zones agree; cross-zone scatter is the drift to read

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

shortlex-3 zone occupancy (lit cells per diagonal square)
INTENT: 2 in the 3×3 ABC corner · 1 in the 9×9 A1..C3 square · 919 in the 132×132 children square · 188 cross-zone
REALITY: 3 in the 3×3 ABC corner · 19 in the 9×9 A1..C3 square · 993 in the 132×132 children square · 317 cross-zone
axes: identical 144 ShortLex names both sides (symmetric); weights asymmetric
zone 3 = candidate children, pre-ratchet (repo-derived dumps not yet past the perturbation-probe gate) · intent 85 claims @ θ 0.6406 · reality 32 claims @ θ 0.625 · projection 875ms

SHAPE-MATCH σ = 1.61 — the real intent↔reality definer walk vs random-reality (actual 0.1216 vs random 0.061). >0 = the walk distinguishes real alignment from noise; aggregate over independent walks → the divergent series.
σ_drift · weak · panel story is not yet evidence

Most out-of-lane row: C2 · Loop — that lane is drifting most (read the red band in that row).

Three-colour tolerance (out-of-lane = reality fires where intent is weak, in orthogonal blocks): green=in-lane agreement · amber=a few out-of-lane (tolerated) · red=armed — orthogonal out-of-lane 22% vs tolerance 25%.
614 green · 749 amber · 0 red · 203 XOR-friction cells.

Both axes = the 144 ShortLex anchors, cell=row⊕col, on the chip. INTENT=docs+rules · REALITY=code · DELTA=friction.

tile dump — what the SimHash matched on the diagonal (inspect the sensor; tap to expand)

A,A The lattice we build today is what next year's underwriter inherits.

intent 78% **2. AI strategy specific — the best-matched role at this table** - **Katya Andresen**, EVP, Ch

reality 72% # there is real work, bounded to max_iter per vector — the "space out 10 revisions, don't die a

B,B Tactics is the choice of which leverage to pull at which beat.

intent 72% # The Chief-of-Staff Overnight Keep-Alive

reality 67% # scripts/controller/cos-overnight.sh — STAGE: the Chief-of-Staff overnight refinement loop.

C,C Operations is the loop that runs whether anyone watches.

intent 78% 1. **MATCH ($0):** `SELECT … FROM cos_overnight WHERE state='pending' AND next_at <= now ORDER

reality 69% (no distinctive match)

A1,A1 EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at

intent 81% In order: (1) run one more verification pass specifically for the AI Center of Enablement / Off

reality 70% (no distinctive match)

A2,A2 The target is the cell coordinate the system MUST occupy by end-of-quarter.

intent 78% scheduled job never inherits the saved interactive default; time-bounded so a hang can never

reality 70% # TIME-BOUNDED + MODEL-PINNED worker. perl alarm survives exec; SIGALRM kills a hang.

A3,A3 Every dollar builds either the floor (substrate that lifts future bets) or the ceiling (cap on the next round)

intent 73% | id | max_iter | what it advances | |---|---|---| | `milestone-strategy` | 3 | self-milestone

reality 67% echo "$(date -u +%FT%TZ) tick: stage cos-overnight rc=$?" >> "$LOG"

B1,B1 Latency is the price of a missed beat.

intent 75% (no distinctive match)

reality 78% # THE KEEP-ALIVE (operator 2026-07-05: "make sure this does not die after one or two iterations

B2,B2 The deal sets the exchange rate between two value substrates.

intent 77% (no distinctive match)

reality 67% (no distinctive match)

B3,B3 Signal-to-noise is the inverse cube of bandwidth — narrow the channel, the message arrives; broaden it, recipi

intent 73% What is solid: Tom Reagan's practice-head role and Tom Srail's analytics/infrastructure role ar

reality 70% WHERE state='pending' AND next_at <= datetime('now')

C1,C1 Power flows where the grid lets it.

intent 80% (no distinctive match)

reality 69% sqlite3 "$DB" "UPDATE cos_overnight SET next_at=datetime('now','+${INTERVAL_MIN} minutes'), upd

C2,C2 Each iteration is a hypothesis tested against the ground truth of the world.

intent 75% (no distinctive match)

reality 67% (no distinctive match)

C3,C3 Flow is the rate at which committed work crosses the finish line.

intent 77% 2. **FIRE (bounded, pinned):** one due vector → one `claude -p --model claude-sonnet-5` worker,

reality 70% # ═══════════════════════════════════════════════════════════════════════════════════════════

commit context — 10 file(s)

feat(controller): Chief-of-Staff overnight keep-alive — SQLite queue drained by the launchd tick

Operator: "make sure this does not die after one or two iterations… use
sqlite… go all night, space out 10 revisions… check the tools first."
A session cron dies with the session; the durable substrate already runs
24/7 — local SQLite + the launchd controller tick. So the keep-alive is
STAGE 8 of the tick, draining a cos_overnight queue.

Per tick: MATCH ($0 SQL due-check, exit before any model touch) → FIRE
(one due vector, claude -p pinned to sonnet, wrapped in a perl-alarm wall
clock, default 540s) → MARK (iter++ in SQL; requeue +45m or done at
max_iter). Two-sided bound: won't die (SQLite + always-on tick survive
session death) and won't run away (model-pinned + time-bounded +
max_iter + single-flight + $0-idle — the $665 and receipt-backfill
lessons both encoded). Standing boundary in every fired prompt: refine +
grade >=95 + commit + STAGE; the only autonomous send allowed is a
self-channel milestone to elias@; NO cold external sends overnight.

Six vectors seeded (milestone-strategy, intro-scan, dinner-cigna-edit,
stage-cigna-sends, dinner-roster-venue, jai-followup@+8h), sta

docs/comms/self/approved/2026-07-06-cigna-head-to-head.grade.json
docs/comms/self/approved/2026-07-06-cigna-head-to-head.md
docs/comms/self/approved/2026-07-06-marsh-broker-angle.grade.json
docs/comms/self/approved/2026-07-06-marsh-broker-angle.md
docs/comms/self/sent/2026-07-06-cigna-head-to-head.grade.json
docs/comms/self/sent/2026-07-06-cigna-head-to-head.md
docs/comms/self/sent/2026-07-06-marsh-broker-angle.grade.json
docs/comms/self/sent/2026-07-06-marsh-broker-angle.md
docs/specs/cos-overnight-keepalive-2026-07-05.md
scripts/controller/cos-overnight.sh

Produced on the chip by the running pipeline (scripts/pmu/pipeline.mjs → runPipeline → walk + xor), commit-scoped · on-chip per-walk 688.99ns (this commit). Recompute: node scripts/pmu/commit-triptych.mjs --commit e0aecdb3a

Commit artifacts (open on a computer — Gmail mobile won't follow file:// links; the inline images above are the phone-viewable form): reef grid · this triptych · the HDL — all three are also attached to this email.