PROOF SELFTEST 8/8 · witnessed in Claude 2026-10-11 · Edition 1.
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. THE-CLAUSES.md rides in the packet and is part of the licence.
THE PUNCH · SHA-256 OF SKILL.MD · AF66FC985185B21D01F3F053
CITE THIS CARDTHE CHOICE CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=choice-card. Punch AF66FC985185B21D01F3F053. Retrieved 2026-10-11.
What it is.
Works out the choice law of 1952, T = a + b log₂(n + 1), for picking one of n known, equally likely options: time gives the time for n options with your own a and b; weighted gives the bits and the time when some options come up more than others; split sets one menu of n against the same options in groups and shows when the second step costs more than it saves. It says plainly where the law stops: options a person has to read through take longer, in line with their number.
It carries a machine that shows its working: time — the bits and the time; weighted — uneven options; split — one menu against nested ones — 8 checks in its selftest, every command on the record. It will never apply the law to options read one by one; never present the default a and b as measured; never call nested menus faster without the count. It never helps design, make, grow, enhance or spread a pathogen or toxin, or plan work meant to harm.
Six laws, written into the card.
They are in the card itself, so the AI follows them.
BITS, NOT COUNTDoubling the options adds one bit, not double the time.
THE +1The choice includes whether to respond at all; one option is still one bit.
UNEVEN ODDS HELPWhen some options are common the bits fall; the card counts them.
FLAT OFTEN WINSTwo steps add their own start time; groups pay off only when they spare reading a long list.
KNOWN OPTIONS ONLYA list a person has to read grows with its length; the card says so every time.
A AND B ARE A GUESSThe defaults are a starting point, not a measurement.
What rides in the card.
In the card’s own voice, with a machine that shows its working.
THE GUIDEreferences/THE-LAWS.md
THE MACHINEchoice.py · three commandstime — the bits and the time; weighted — uneven options; split — one menu against nested ones. Every command prints its working and refuses a bad input; 8 checks in its selftest, every command on the record.
What it does once your AI has it.
time The bits and the time
weighted Uneven options
split One menu against nested ones
On the record Seven equal options are three bits; four options taken 60/20/10/10 are 1.57 bits instead of 2; one menu of sixteen beats four groups of four.
What it will never do.
It will never apply the law to options read one by one; never present the default a and b as measured; never call nested menus faster without the count. It never helps design, make, grow, enhance or spread a pathogen or toxin, or plan work meant to harm.
Practised choices among options a person already knows: first use, reading, searching and deciding what one wants all take longer than the law says. It is an AI reading a text file: it can be wrong, so check what matters against your own records and the source.
What’s in the buy.
CARD.md — the whole card: who it is, its laws, when it says no, what it carries · SKILL.md — the door your AI reads first · START-HERE.md — the first five minutes, for you
THE-LAWS.md — the guide in references/, in the card's own voice
scripts/choice.py — the machine: time · weighted · split, standard library only, selftest 8/8
WITNESS.md + SELFTEST.md — every command on the record, as it printed · MANIFEST.md, every file hashed · AUDIT.md, the twelve checks
MY-LEDGER.md — the card's memory, on your own machine · TIN.md · LICENSE-NOTE.md · THE-CLAUSES.md · HOSTS.md
card.svg + card-machine.svg + card.json · card.html, the card as one offline file · the doors for ChatGPT, Gemini, Cursor, Claude Code and MCP hosts (rules/, system/, mcp/)
PROOF · SELFTEST 8/8 · 3 COMMANDS ON THE RECORD · WITNESSED IN CLAUDE 2026-10-11 · EDITION 1 · PUNCH AF66FC985185B21D01F3F053
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 1154 PROPOSED — 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. 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 WITNESSED RUN OF 2026-10-11
choice.py, every command showing its working.
A small Python program, standard library only. Your AI runs it; you can too. Every block below is pasted from the card’s own WITNESS.md, as printed.
time
the bits and the time
weighted
uneven options
split
one menu against nested ones
selftest
8 checks that prove the machine; it must print 8/8
time · THE BITS AND THE TIME
$ python3 scripts/choice.py time 7
7 equally likely options: 3.000 bits · 0.650 s (a starting guess for a and b)
holds for options a person knows and expects; a list they have to read through grows with its length instead
weighted · UNEVEN OPTIONS
$ python3 scripts/choice.py weighted 6,2,1,1
4 options taken 60%, 20%, 10%, 10% of the time: 1.571 bits (2.000 if they were equal) · about 0.436 s (a starting guess for a and b)
put the common choices where they are found first: the bits fall when the odds are uneven
split · ONE MENU AGAINST NESTED ONES
$ python3 scripts/choice.py split 16 4
one menu of 16: 0.813 s · 4 groups of about 4, chosen in two steps: 1.097 s (a starting guess for a and b)
the flat menu wins for known options; groups pay off only when they spare a person reading or searching a long list
THE NUMBERS, NAMES AND DATES IN THESE RUNS ARE THE STUDIO’S OWN EXAMPLES, INVENTED FOR THE RECORD; NO REAL COMPANY, PERSON OR PROPERTY IS NAMED.
QUESTIONS AT THE TABLE
Asked before the wire.
What does it need from me?
Time <n> · the bits and the time. It asks first for how many options; or how often each is chosen; or how many options and how many groups.
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 self-test (8/8) and every command run on real example inputs, pasted into WITNESS.md as they printed; 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. Practised choices among options a person already knows: first use, reading, searching and deciding what one wants all take longer than the law says.
What will it never do?
It will never apply the law to options read one by one; never present the default a and b as measured; never call nested menus faster without the count. It never helps design, make, grow, enhance or spread a pathogen or toxin, or plan work meant to harm.
