THE METHODS · THE STUDIO ROOM · Nº 418 · EDITION 2 · LAUNCHING SOON

LUND CARDS · THE METHODSNº 418

THE CARD AS TEXT · FOR EYES AND FOR MACHINES

THE PICK ROOM CARD · The client review seat: options side by side, every pick and reason on record, decided by the rule.

WAKES WHEN Three options went to the client and the answer came back as a feeling. · Reviewers voted in a thread and nobody can count it. · Someone voted twice, or for an option that was never offered. · A decision is claimed and the rule it was made by was never agreed.

THE LOOP

Lay the options; agree the rule first
Write the review sheet, one line per option
Record every pick with its reason
Tally; decide only by the rule, or NO DECISION

NEVER

Never decides by a rule not agreed first
Never counts a vote twice
Never drops a reason from the record
Never claims a certainty it does not have

IT MAKES sheet — the review sheet: the rule and the reviewers at the top, one line per option with a blank mark and a blank reason; printed, and written to --out when given; NO RULE (exit 2) when the room has no rule; tally — first choices per option, most first, every reason under its option; UNKNOWN OPTION, DOUBLE VOTE and UNKNOWN REVIEWER listed with their reasons and not counted; a TIE at the top stated; reviewers with no counted vote named; decide — DECISION by the rule as written (consensus: every reviewer in the room voted and all for one option; majority: more than half the reviewers in the room for one option) or NO DECISION with what the rule needed (exit 1); NO RULE when the room has none (exit 2); check — UNKNOWN OPTION, DOUBLE VOTE, UNKNOWN REVIEWER and NO RULE named row by row; exit 1 on any.

THE LAW A pick is a record: who chose what, and why, under a rule agreed before the first vote. The tally is printed in the open; a vote for an option not offered is UNKNOWN OPTION, a second vote from the same reviewer is DOUBLE VOTE, and when the rule is not met the answer is NO DECISION, not a nudge toward one.

IN THE BOX SKILL.md · CARD.md · ▣ scripts/pickroom.py (machine) · ▣ scripts/evals.py (test runner) · tests/EVALS.json (9 tests) · references/ (schema + fictional demo) · START-HERE.md · system/ doors · MY-LEDGER.md · LICENSE-NOTE.md · card.svg · card.json · card.html · AUDIT.md · MANIFEST.md · WITNESS.md · SELFTEST.md.

PROOF SELFTEST 15/15 · witnessed in Claude 2026-10-14 on fictional fixtures · the test prompts in fresh chat, 9/9 · Edition 2.

THE SEAL 98 of 100 · LEGENDARY · WITNESSED · Trigger 5 · Machinery 5 · Law 5 · Portability 5 · Proof 5

IT CAN BE WRONG an AI reading a text file: it can be wrong and can invent a number, a rule, a name or a date; check what matters against your own records and the source; the risk is yours. LICENSE-NOTE.md rides in the packet and is the licence.

THE PUNCH · SHA-256 OF SKILL.MD · 721C30E004D46A8EB086BB96

THE FACE, FULL SIZE → · THE MACHINE FACE → · card.json →

THE PICK ROOM CARD

The client review seat: options side by side, every pick and reason on record, decided by the rule.

SHEET · TALLY · DECIDE · CHECK

LEGENDARYWITNESSEDIN CLAUDE IN LUNDR IN CHATGPT

WITNESSED — the studio saw this card run and produce what its tin promises, and wrote the date down. What the marks mean →

THE SPEC SHEET · READ BEFORE YOU BUY · THE STANDARD EVERY CARD MEETS →

SEAL: 98 of 100 · LEGENDARY · WITNESSED
EDITION: E1 · witnessed 2026-10-14
MACHINE: scripts/pickroom.py · 4 commands · selftest 15/15
RUNS: On your own computer. Standard library only, offline; no account, no key, no telemetry.
HOSTS: Claude · Claude Code · ChatGPT · Gemini · any AI that reads a file (START-HERE.md and system/ name each door)
LICENCE: One person, any number of their own AIs and computers · editions replace files · nothing renews · freeze, not loss
PUNCH: 721C30E004D46A8EB086BB96 · sha-256 of SKILL.md · check any file against it →

CITE THIS CARDTHE PICK ROOM CARD, Edition 2. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=pick-room-card. Punch 721C30E004D46A8EB086BB96. Retrieved 2026-10-14.

What it is.

Use when a client or a group must choose between finished options: the review sheet, the votes with reasons, the tally, a decision by the rule agreed in advance. Not for scoring on weighted criteria.

It carries a machine that shows its working: sheet — the review sheet: the rule and the reviewers at the top, one line per option with a blank mark and a blank reason; printed, and written to --out when given; NO RULE (exit 2) when the room has no rule; tally — first choices per option, most first, every reason under its option; UNKNOWN OPTION, DOUBLE VOTE and UNKNOWN REVIEWER listed with their reasons and not counted; a TIE at the top stated; reviewers with no counted vote named; decide — DECISION by the rule as written (consensus: every reviewer in the room voted and all for one option; majority: more than half the reviewers in the room for one option) or NO DECISION with what the rule needed (exit 1); NO RULE when the room has none (exit 2); check — UNKNOWN OPTION, DOUBLE VOTE, UNKNOWN REVIEWER and NO RULE named row by row; exit 1 on any — 15 checks in its selftest. It never decides by a rule not agreed first; never counts a vote twice; never drops a reason from the record; never claims a certainty it does not have.

The law, written into the card.

It is in the card itself, so the AI follows it.

THE LAWA pick is a record: who chose what, and why, under a rule agreed before the first vote. The tally is printed in the open; a vote for an option not offered is UNKNOWN OPTION, a second vote from the same reviewer is DOUBLE VOTE, and when the rule is not met the answer is NO DECISION, not a nudge toward one.

The method.

Lay the options in a sheet: name, file, a line of note; agree the decision rule (consensus, or a majority of the reviewers named) and write it in the sheet before anything is shown.
Write the review sheet: one line per option with a blank mark and a blank reason, the rule printed at the top.
Record the picks: reviewer, option, first choice marked, the reason in the reviewer's words; a pick for an option not on the sheet is UNKNOWN OPTION; a second pick from one reviewer is DOUBLE VOTE.
Tally: first choices per option, the reasons grouped under each; ties stated.
Decide by the rule as written, or print NO DECISION with what the rule needed; the sheet stays as the record either way.

What you bring.

The options, finished, with their files.
The reviewers who may vote.
The rule for a decision, agreed before the vote.

Planned as a separate seat, not in this box and not yet on sale: the studio's own pick rooms for its client work.

What it does once your AI has it.

sheet The review sheet: the rule and the reviewers at the top, one line per option with a blank mark and a blank reason; printed, and written to --out when given; NO RULE (exit 2) when the room has no rule
tally First choices per option, most first, every reason under its option; UNKNOWN OPTION, DOUBLE VOTE and UNKNOWN REVIEWER listed with their reasons and not counted; a TIE at the top stated; reviewers with no counted vote named
decide DECISION by the rule as written (consensus: every reviewer in the room voted and all for one option; majority: more than half the reviewers in the room for one option) or NO DECISION with what the rule needed (exit 1); NO RULE when the room has none (exit 2)
check UNKNOWN OPTION, DOUBLE VOTE, UNKNOWN REVIEWER and NO RULE named row by row; exit 1 on any

What it will never do.

It never decides by a rule not agreed first; never counts a vote twice; never drops a reason from the record; never claims a certainty it does not have.

Where it hands off: Options scored on weighted criteria → THE BOARD CARD; A council of seats reviewing one piece → THE COUNCIL CARD; Changing the options themselves → the maker of the options.

It is not a recommendation. It is an AI reading a text file: it can be wrong, so check what matters against your own records and the source. The holder brings their own AI; no endorsement by Anthropic or any other AI company is claimed.

What’s in the buy.

SKILL.md — the card your AI reads: when it wakes, the loop, the never-lines, the hand-offs.
CARD.md — the whole card in ten sections: face, law, method, machine, what you bring, never, tests, seal, term.
scripts/pickroom.py — the machine, Python standard library only; `selftest` proves it on its own fixtures.
tests/EVALS.json and scripts/evals.py — 9 test prompts and the offline runner that scores your AI's answers.
references/ — the data schema and FICTIONAL demo files, invented to show the machine, not real.
MY-LEDGER.md — your own record of what you asked and what you decided.
START-HERE.md, system/short-prompt.md, system/rules.md — the doors, for any AI with or without skills.
card.svg, card.json, card.html, AUDIT.md, MANIFEST.md, WITNESS.md, SELFTEST.md — the faces and the proof.
NOT IN THE BOX — any studio edition (planned, not yet on sale: the studio's own pick rooms for its client work). Also not in the box: any data you must bring, and advice of any kind (not a recommendation).

PROOF · SELFTEST 15/15 · WITNESSED IN CLAUDE 2026-10-14 ON FICTIONAL FIXTURES · THE 9 TEST PROMPTS ASKED IN A FRESH CHAT HOLDING ONLY THE CARD · 9/9 · EDITION 2 · PUNCH 721C30E004D46A8EB086BB96

ON EVERY DOOR · A CARD FOR YOUR AI, A FILE FOR YOUR COMPUTER

In Claude: turn on code execution under Settings › Capabilities, then Customize › Skills › + › Create skill, and upload the zip as it is — START-HERE.md in the box walks you through it, and tells you how to keep the ledger. In ChatGPT, Gemini and the rest: a Project, a GPT or a Gem with the card’s files attached and the short door in the box pasted as its instructions. The words run anywhere an AI reads Markdown; the machines run wherever the host runs Python. Witnessed in Claude only. The host marks →

Take it.

Nº 418 — LAUNCHING SOON

Sold as is under the term printed. A card instructs an AI you operate; results depend on your model, your files and your judgement. Not medical, legal, tax or financial advice, never the person’s boss, and not a licence to practise. The terms →

ONE PAYMENT · NOTHING RENEWS ITSELF

OFFLINE · ONE LICENCE · PUNCH-CHECKABLE · THE STANDARD

One payment. The file is yours to keep and keeps working at its edition; new editions issued in the year after you buy reach you by email as replacement files. Tax is added at checkout where it applies. Nothing auto-renews.

A TERM, NOT A SUBSCRIPTION: NOTHING RENEWS ITSELF AND NOBODY IS CHARGED AGAIN. WHEN A TERM ENDS THE FILES STAY ON YOUR COMPUTER AND STAY READABLE — FREEZE, NOT LOSS.

THE MACHINE · IN THE OPEN · FROM THE CARD’S OWN SELFTEST.MD

pickroom.py, every command showing its working.

A small Python program, standard library only. Your AI runs it; you can too. The block below is pasted from the card’s own SELFTEST.md, as printed when it was run from inside the packet.

`sheet`	the review sheet: the rule and the reviewers at the top, one line per option with a blank mark and a blank reason; printed, and written to --out when given; NO RULE (exit 2) when the room has no rule
`tally`	first choices per option, most first, every reason under its option; UNKNOWN OPTION, DOUBLE VOTE and UNKNOWN REVIEWER listed with their reasons and not counted; a TIE at the top stated; reviewers with no counted vote named
`decide`	DECISION by the rule as written (consensus: every reviewer in the room voted and all for one option; majority: more than half the reviewers in the room for one option) or NO DECISION with what the rule needed (exit 1); NO RULE when the room has none (exit 2)
`check`	UNKNOWN OPTION, DOUBLE VOTE, UNKNOWN REVIEWER and NO RULE named row by row; exit 1 on any
`selftest`	15 checks that prove the machine; it must print selftest 15/15

THE SELFTEST, AS PRINTED

ok   1  sheet: the rule and its meaning on the first line, the reviewers on the second, one line per option with its file and note, a blank mark and a blank reason; --out writes the same text
ok   2  tally: first choices per option most first, every reason under its option (a blank one as NO REASON given), nothing flagged, every reviewer counted
ok   3  UNKNOWN OPTION: a vote for an option not on the sheet is listed with its reason and not counted
ok   4  DOUBLE VOTE: a second vote from one reviewer is listed with its reason and not counted; the first vote stands
ok   5  UNKNOWN REVIEWER: a name not in the room is listed with its reason and not counted
ok   6  reasons are never dropped: every reason in the votes file is somewhere in the tally
ok   7  check names the three flags row by row and exits 1
ok   8  check on clean votes prints no flags and exits 0
ok   9  decide by majority: 3 of 5 for one option is a DECISION naming the option, the rule, the count and the record (exit 0)
ok   10  decide by majority with 2 of 5 is NO DECISION with what was needed, the most so far and who has no counted vote (exit 1); the tally never nudges
ok   11  a tie at the top is stated as TIE with both options and the count; decide on a tie is NO DECISION naming both
ok   12  decide by consensus: all three for one option is a DECISION; one reviewer silent is NO DECISION naming who; one for another option is NO DECISION (exit 1 both)
ok   13  a room with no rule: decide and sheet refuse NO RULE (exit 2), check flags NO RULE (exit 1), tally still counts and says NO RULE on file
ok   14  a rule outside consensus|majority, an empty reviewer list, an option named twice, a vote with no option and a sheet without its columns are bad input (exit 2)
ok   15  the same files give the same bytes twice
selftest 15/15

THE RUN IS PASTED FROM THE CARD’S OWN SELFTEST.MD AS PRINTED. ITS FIXTURES ARE FICTIONAL, INVENTED FOR THE TEST; NO PRIVATE PERSON IS NAMED.

QUESTIONS AT THE TABLE

Asked before the wire.

What does it need from me?

The options, finished, with their files. The reviewers who may vote. The rule for a decision, agreed before the vote. If something it needs is missing, it says what is missing and stops there.

Does it send anything anywhere?

No. The machine is a standard-library Python program that runs on your own computer and sends nothing anywhere; the words run wherever an AI reads Markdown. No server, no account, no key.

What is on the record?

The machine's selftest (15/15), run from inside the packet in Claude on 2026-10-14 on fictional fixtures and pasted into SELFTEST.md as printed; the twelve-check audit (AUDIT.md). Owed, and said so in WITNESS.md: Any run on real data. Every demo file is fictional. The punch on the face is the SHA-256 of SKILL.md, so a downloaded copy can be checked against this room.

Can it be wrong?

Yes — it is an AI reading a text file: it can be wrong and can invent a number, a rule, a name or a date; check what matters against your own records and the source; the risk is yours. LICENSE-NOTE.md rides in the packet and is the licence. It is not a recommendation.

What will it never do?

It never decides by a rule not agreed first; never counts a vote twice; never drops a reason from the record; never claims a certainty it does not have.

← THE METHODS · THE FLOOR →