fix(pmu): strip the LLM from the decidable receipt — deterministic, LLM-free, immutable-commit reality

HARD RULE (operator 2026-07-04): the decidable on-chip receipt (tolerance panel / insurance
images / σ / placement / lane) is LLM-FREE and a pure function of the commit. NEVER combine an
LLM's output with the decidable panel again.

ROOT CAUSE of the empty/non-deterministic panels: a local-LM 'semantic reality' (ollama, flag
.thetacog/semantic-reality.on) was PREPENDED into the reality corpus. It paraphrased differently
every call ('proper'↔'correct'), so the SAME commit rendered 1079/1590/333 lit and random EMPTY
panels. A reproducible coordinate space leaning on an undecidable model is fooling ourselves —
the exact thing S=P=H refuses.

FIX:
- remove semanticReality + the semantic-reality.on LLM flag from the reality corpus entirely.
- read reality from the IMMUTABLE commit (git show <sha>:<file>), never the mutable working tree.
- corpus-ingest buildMass brought up to the ingestWide deterministic spec (path-order not mtime,
  touch-stable size key not mtimeMs, atomic tmp+rename write, absolute cache path — the fixed-name
  relative cache had scattered into a stray <sha>/ dir under concurrency).

PROVEN: a substantial commit now renders an IDENTICAL 327g/629a/768r on every render (was random);
a thin 2-line commit renders a deterministic direction-only 0-region receipt (honest, never a
*broken* blank). The LLM story over the drift signals becomes a SEPARATE later output (two emails,
combinable into one webpage — but the receipt half is computed with no model).

Guard: tests/pmu-simulator/receipt-is-llm-free.test.mjs (no semanticReality · immutable read ·
HARD-RULE documented · same-commit-twice-identical determinism).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013BUf5M5p42Zw73VaGt33ZL

commit 5904dccea — 144×144 lattice triptych (on-chip)

◎ the ingest chose A,B1 — Strategy × Tactics·Speed — acting on C1,A1 — Operations·Grid × Strategy·Law (grip 0.797) · the walk starts at the ACTOR's row · 4 file(s)

⏱ ingest 522.3ms (corpus→lattice) · definer walk 727ms (row→transpose→row, ballistic XOR per hop) · total 1398ms
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) · 6% 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

A,B1 — Strategy × Tactics·Speed (actor) acting on C1,A1 — Operations·Grid × Strategy·Law (patient) · grip 0.797

actor seed A,B1 — Defining a long-horizon strategy frame requires a lattice where speed and tempo beats prevent missed choices. The lattice we build today is what next year's underwriter inherits. We set the substrate that downstream pricing rides on, and that choice is paid back across every quarter we keep operating. Direction is what you keep when you forget the calendar. Multi-year posture, irreversibly chosen, defines the boundary every quarter renegotiates inside. Strategy is the bearing the rotation is measured against. 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. Time-to-first-byte is the only number the operator feels at the wheel. Optimize the felt latency before the raw throughput; the operator's grip on the loop is what drives the loop forward.

patient seed C1,A1 — Power flows where the grid lets it. Choose connections before topologies; the topology is what the connections compose, and a connection added later is a topology silently rewritten. Latency is geography expressed in milliseconds. The grid IS the map of where compute meets storage meets bandwidth; the map is the territory at the scale the request actually traverses. Infrastructure decisions are reversible at ten-times the cost of being right the first time. Buy the irreversibility down before you provision the easy decisions, because the easy decisions ride on the irreversible ones. EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at any moment. Compliance attaches to the artifact the artifact cannot escape — provenance is a regulatory primitive, not a feature. Constraints precede objectives. The cap-table, the license text, the consent envelope — each draws the inviolate line the action must respect, and the line is enforced before the optimization runs. A rule names what the system MAY NOT do. If the agent can do X, then 'forbid X' is not the constraint — it is wishful thinking dressed as policy. Constraints live in the gate, not in the prompt.

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) · 5% 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) · ⚠ TOO MANY

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 · 5% 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: 1 in the 3×3 ABC corner · 13 in the 9×9 A1..C3 square · 1908 in the 132×132 children square · 326 cross-zone
REALITY: 0 in the 3×3 ABC corner · 5 in the 9×9 A1..C3 square · 1402 in the 132×132 children square · 199 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 151 claims @ θ 0.6406 · reality 48 claims @ θ 0.625 · projection 579ms

SHAPE-MATCH σ = -0.1 — the real intent↔reality definer walk vs random-reality (actual 0.1118 vs random 0.114). >0 = the walk distinguishes real alignment from noise; aggregate over independent walks → the divergent series.
σ_drift · noise · below zero — this read is noise, not a measurement

Most out-of-lane row: B2 · Deal — 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=FIRED — orthogonal out-of-lane 29% vs tolerance 25%.
615 green · 599 amber · 500 red · 438 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 77% **Through-holes (H state in grid):** - Type: UPSIDE-DOWN PYRAMIDS (inverted, point down) - Why:

reality 75% const cacheDir = CACHE_DIR_ABS; // absolute — never scattered by a wrong CWD

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

intent 75% ### SQL (Central Supabase: `bvhhlosblntckhwyagvp.supabase.co`) | File | Purpose | |---|---| | `

reality 78% ].filter(Boolean).join('\n\n') || files.map(f => readAtCommit(f)).join('\n\n'); // fallback als

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

intent 80% # Claude Code Configuration

reality 83% (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 83% (no distinctive match)

reality 80% (no distinctive match)

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

intent 83% (no distinctive match)

reality 80% (no distinctive match)

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

intent 77% ### DELEGATION NEVER GIT-BRANCHES (HARD RULE — enforced by tests, not by prose) **We NEVER auto

reality 72% (no distinctive match)

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

intent 73% (no distinctive match)

reality 73% (no distinctive match)

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

intent 81% (no distinctive match)

reality 73% (no distinctive match)

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

intent 75% PROVEN: a substantial commit now renders an IDENTICAL 327g/629a/768r on every render (was rando

reality 78% (no distinctive match)

C1,C1 Power flows where the grid lets it.

intent 78% (no distinctive match)

reality 75% (no distinctive match)

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

intent 80% ### MDX FORBIDDEN CHARACTERS **The following CRASH the MDX parser (500 error on blog posts):**

reality 70% (no distinctive match)

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

intent 78% - **NEVER** just substantiate claims - show why the reader should care - **EVERY** section shou

reality 78% // rest. Matches the ingestWide deterministic spec. Guard: tests/pmu-simulator/corpus-ingest-de

commit context — 4 file(s)

fix(pmu): strip the LLM from the decidable receipt — deterministic, LLM-free, immutable-commit reality

HARD RULE (operator 2026-07-04): the decidable on-chip receipt (tolerance panel / insurance
images / σ / placement / lane) is LLM-FREE and a pure function of the commit. NEVER combine an
LLM's output with the decidable panel again.

ROOT CAUSE of the empty/non-deterministic panels: a local-LM 'semantic reality' (ollama, flag
.thetacog/semantic-reality.on) was PREPENDED into the reality corpus. It paraphrased differently
every call ('proper'↔'correct'), so the SAME commit rendered 1079/1590/333 lit and random EMPTY
panels. A reproducible coordinate space leaning on an undecidable model is fooling ourselves —
the exact thing S=P=H refuses.

FIX:
- remove semanticReality + the semantic-reality.on LLM flag from the reality corpus entirely.
- read reality from the IMMUTABLE commit (git show <sha>:<file>), never the mutable working tree.
- corpus-ingest buildMass brought up to the ingestWide deterministic spec (path-order not mtime,
  touch-stable size key not mtimeMs, atomic tmp+rename write, absolute cache path — the fixed-name
  relative cache had scattered into a stray 

CLAUDE.md
scripts/pmu/commit-triptych.mjs
scripts/pmu/corpus-ingest.mjs
tests/pmu-simulator/receipt-is-llm-free.test.mjs

Produced on the chip by the running pipeline (scripts/pmu/pipeline.mjs → runPipeline → walk + xor), commit-scoped · on-chip per-walk 521.61ns (this commit). Recompute: node scripts/pmu/commit-triptych.mjs --commit 5904dccea

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.