THE METHODS · THE STUDIO ROOM · Nº 430 · EDITION 2 · LAUNCHING SOON

LUND CARDS · THE METHODSNº 430

THE CARD AS TEXT · FOR EYES AND FOR MACHINES

THE SUPPORT CARD · The answering seat: every ticket answered in time, from the house's answers, within the policy.

WAKES WHEN Tickets have piled up and nobody knows which are oldest or overdue. · A reply is needed now and the house has an answer written for exactly this. · Someone wants to know how fast the house actually answers, week by week. · A canned answer promises something the policy no longer allows.

THE LOOP

Triage by age against the reply window
Draft from a written answer, with the facts
Refuse a promise the policy does not hold
Log the closes; time to close, by week

NEVER

Never promises what the policy does not allow
Never invents an answer not on file
Never closes a ticket without a reply logged
Never claims a certainty it does not have

IT MAKES triage — open and waiting tickets oldest first: OVERDUE past the reply window, DUE inside the last quarter of it, FRESH otherwise; the counts; draft — the written answer with {name}, {order_id} and {ticket_id} filled from the ticket, nothing added; PROMISE NOT IN POLICY or FIELD MISSING refused (exit 1); UNKNOWN TICKET or UNKNOWN ANSWER exit 2; check — every answer's promises against promises_allowed and its placeholders against the three fields; offenders listed; exit 1 if any; close — a new tickets file with the ticket closed, the hour and the closing answer logged; NO REPLY LOGGED without --answer, NEVER IN PLACE when --out is the input, ALREADY CLOSED (each exit 1); log — per ISO week: tickets closed, median hours from received to closed, the share closed inside the reply window (NO POLICY without the policy file), n on every line; a close with no hour or no reply logged is named.

THE LAW A reply is written once by the house and sent many times by the seat. Every promise in a reply must already stand in the policy; a draft that would promise more is refused with PROMISE NOT IN POLICY. Age is counted from the hour the ticket arrived, and overdue is said in capitals, never softened.

IN THE BOX SKILL.md · CARD.md · ▣ scripts/support.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 18/18 · 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 · 658DC90C9833B4EB14B7AA79

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

THE SUPPORT CARD

The answering seat: every ticket answered in time, from the house's answers, within the policy.

TRIAGE · DRAFT · CHECK · CLOSE · LOG

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/support.py · 5 commands · selftest 18/18
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: 658DC90C9833B4EB14B7AA79 · sha-256 of SKILL.md · check any file against it →

CITE THIS CARDTHE SUPPORT CARD, Edition 2. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=support-card. Punch 658DC90C9833B4EB14B7AA79. Retrieved 2026-10-14.

What it is.

Use when customer tickets must be triaged, answered from the house's own canned answers and timed against its reply policy. Not for writing the policy itself or for legal complaints.

It carries a machine that shows its working: triage — open and waiting tickets oldest first: OVERDUE past the reply window, DUE inside the last quarter of it, FRESH otherwise; the counts; draft — the written answer with {name}, {order_id} and {ticket_id} filled from the ticket, nothing added; PROMISE NOT IN POLICY or FIELD MISSING refused (exit 1); UNKNOWN TICKET or UNKNOWN ANSWER exit 2; check — every answer's promises against promises_allowed and its placeholders against the three fields; offenders listed; exit 1 if any; close — a new tickets file with the ticket closed, the hour and the closing answer logged; NO REPLY LOGGED without --answer, NEVER IN PLACE when --out is the input, ALREADY CLOSED (each exit 1); log — per ISO week: tickets closed, median hours from received to closed, the share closed inside the reply window (NO POLICY without the policy file), n on every line; a close with no hour or no reply logged is named — 18 checks in its selftest. It never promises what the policy does not allow; never invents an answer not on file; never closes a ticket without a reply logged; 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 reply is written once by the house and sent many times by the seat. Every promise in a reply must already stand in the policy; a draft that would promise more is refused with PROMISE NOT IN POLICY. Age is counted from the hour the ticket arrived, and overdue is said in capitals, never softened.

The method.

Triage first: every open ticket's age against the house's reply window; OVERDUE, DUE and FRESH in that order, oldest first.
Draft from a written answer: the machine fills the ticket's own facts (name, order, dates) into the house's text; it never composes a new promise.
Check the answers file against the policy whenever either changes; an answer carrying a promise the policy does not list is flagged until a person fixes one or the other.
Close a ticket only with the reply that closed it logged beside it, and the hour.
Report by week: tickets closed, median hours to close, the share answered inside the window; the numbers carry their counts.

What you bring.

The tickets, with the hour each arrived.
The house's written answers.
The reply policy: the window and the promises allowed.

Planned as a separate seat, not in this box and not yet on sale: the studio's own answers for its cards and vessels.

What it does once your AI has it.

triage Open and waiting tickets oldest first: OVERDUE past the reply window, DUE inside the last quarter of it, FRESH otherwise; the counts
draft The written answer with {name}, {order_id} and {ticket_id} filled from the ticket, nothing added; PROMISE NOT IN POLICY or FIELD MISSING refused (exit 1); UNKNOWN TICKET or UNKNOWN ANSWER exit 2
check Every answer's promises against promises_allowed and its placeholders against the three fields; offenders listed; exit 1 if any
close A new tickets file with the ticket closed, the hour and the closing answer logged; NO REPLY LOGGED without --answer, NEVER IN PLACE when --out is the input, ALREADY CLOSED (each exit 1)
log Per ISO week: tickets closed, median hours from received to closed, the share closed inside the reply window (NO POLICY without the policy file), n on every line; a close with no hour or no reply logged is named

What it will never do.

It never promises what the policy does not allow; never invents an answer not on file; never closes a ticket without a reply logged; never claims a certainty it does not have.

Where it hands off: Writing or changing the reply policy → the owner of the house; A refund that needs a figure → THE PRICE CARD; A complaint that names the law → a lawyer.

It is not legal advice on a complaint. 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/support.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 answers for its cards and vessels). Also not in the box: any data you must bring, and advice of any kind (not legal advice on a complaint).

PROOF · SELFTEST 18/18 · 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 658DC90C9833B4EB14B7AA79

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 430 — 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

support.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.

`triage`	open and waiting tickets oldest first: OVERDUE past the reply window, DUE inside the last quarter of it, FRESH otherwise; the counts
`draft`	the written answer with {name}, {order_id} and {ticket_id} filled from the ticket, nothing added; PROMISE NOT IN POLICY or FIELD MISSING refused (exit 1); UNKNOWN TICKET or UNKNOWN ANSWER exit 2
`check`	every answer's promises against promises_allowed and its placeholders against the three fields; offenders listed; exit 1 if any
`close`	a new tickets file with the ticket closed, the hour and the closing answer logged; NO REPLY LOGGED without --answer, NEVER IN PLACE when --out is the input, ALREADY CLOSED (each exit 1)
`log`	per ISO week: tickets closed, median hours from received to closed, the share closed inside the reply window (NO POLICY without the policy file), n on every line; a close with no hour or no reply logged is named
`selftest`	18 checks that prove the machine; it must print selftest 18/18

THE SELFTEST, AS PRINTED

ok   1  triage lists open and waiting tickets oldest first with OVERDUE, DUE and FRESH and the counts; closed tickets are left out
ok   2  triage never softens OVERDUE: one minute past the window is OVERDUE, exactly at the window is DUE
ok   3  the window is read from the policy only: at 48 h the same queue has one OVERDUE and T-102 is FRESH
ok   4  draft fills {name}, {order_id} and {ticket_id} from the ticket and writes the answer text and nothing else
ok   5  draft refuses PROMISE NOT IN POLICY (exit 1) and writes no file
ok   6  draft: UNKNOWN ANSWER and UNKNOWN TICKET are bad input (exit 2)
ok   7  draft refuses FIELD MISSING (exit 1) when the ticket lacks a field the answer needs, instead of inventing one
ok   8  draft will not fill a placeholder outside name, order_id, ticket_id: bad input (exit 2)
ok   9  check lists the answer whose promise is outside the policy and the unknown field, and exits 1
ok   10  check exits 0 when every promise is inside the policy
ok   11  close writes a new file with the ticket closed, closed_at and closing_answer logged, every other row and the fiction line untouched
ok   12  close refuses NO REPLY LOGGED without --answer (exit 1) and writes nothing
ok   13  close refuses NEVER IN PLACE when --out is the tickets file itself
ok   14  close: ALREADY CLOSED is refused (exit 1); an unknown ticket, an hour before the ticket arrived or an answer key not in --answers is bad input (exit 2)
ok   15  log prints per ISO week the closes, the median hours and the share inside the window with n
ok   16  log names a ticket CLOSED WITHOUT AN HOUR and CLOSED WITHOUT A REPLY LOGGED instead of counting it
ok   17  log without --policy says NO POLICY and leaves the share inside the window blank rather than guessing a window
ok   18  a malformed --now or a status outside open|waiting|closed is bad input (exit 2)
selftest 18/18

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 tickets, with the hour each arrived. The house's written answers. The reply policy: the window and the promises allowed. 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 (18/18), 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 legal advice on a complaint.

What will it never do?

It never promises what the policy does not allow; never invents an answer not on file; never closes a ticket without a reply logged; never claims a certainty it does not have.

← THE METHODS · THE FLOOR →