THE METHODS · THE STUDIO ROOM · Nº 23 · EDITION 2 · LAUNCHING SOON

LUND CARDS · THE METHODSNº 23

THE CARD AS TEXT · FOR EYES AND FOR MACHINES

THE PRIOR ART CARD · The publication seat: what was searched and found, set against each claim; a record, not a verdict.

WAKES WHEN Someone says 'nobody has done this' and no search was ever logged. · A filing is coming and the record of what was searched must stand on its own. · Findings are piling up and nobody has tied them to the claims they touch. · A reviewer asks what was searched, where and when.

THE LOOP

Log every search: where, what, when, hits
List every finding with what it shows
Tie findings to claims: anticipates, overlaps
Report the record; never the word novel

NEVER

Never prints the word novel as a verdict
Never reports a claim with no search logged
Never drops a finding that overlaps
Never claims a certainty it does not have

IT MAKES log — append one search to the searches sheet (created with its header when missing) and print the row back; a bad day, hits below 0 or no claim id is bad input (exit 2); matrix — claims down, findings across, the relation in each cell (anticipates, overlaps, differs, - for no pair); every ANTICIPATES and OVERLAPS listed first with its why; a pair naming an id not on the sheets is UNKNOWN ID, listed apart, exit 1; check — UNSEARCHED for every claim no logged search names; UNKNOWN ID for a search or pair naming a claim (or, with --found, a finding) not on the sheets; exit 1 on either; report — the record on one page: the header that it is not a legal opinion, the claims, the searches, the findings, the matrix, every tie with its why, the UNSEARCHED claims; printed, and written to --out when given.

THE LAW Prior art is a record of what was looked for and what was found; novelty is a word for a professional. Every search is logged with where, what and when; every finding is tied to the claims it touches as anticipates, overlaps or differs; a claim with no search behind it is UNSEARCHED. The report says so at the top: not a legal opinion.

IN THE BOX SKILL.md · CARD.md · ▣ scripts/priorart.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 14/14 · 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 · 16A3547A4315CE8A3A6DC051

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

THE PRIOR ART CARD

The publication seat: what was searched and found, set against each claim; a record, not a verdict.

LOG · MATRIX · CHECK · REPORT

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/priorart.py · 4 commands · selftest 14/14
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: 16A3547A4315CE8A3A6DC051 · sha-256 of SKILL.md · check any file against it →

CITE THIS CARDTHE PRIOR ART CARD, Edition 2. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=prior-art-card. Punch 16A3547A4315CE8A3A6DC051. Retrieved 2026-10-14.

What it is.

Use when a claim or invention needs a prior-art record before publishing or filing: searches logged, findings listed, each tied to each claim. Not for a legal opinion on novelty.

It carries a machine that shows its working: log — append one search to the searches sheet (created with its header when missing) and print the row back; a bad day, hits below 0 or no claim id is bad input (exit 2); matrix — claims down, findings across, the relation in each cell (anticipates, overlaps, differs, - for no pair); every ANTICIPATES and OVERLAPS listed first with its why; a pair naming an id not on the sheets is UNKNOWN ID, listed apart, exit 1; check — UNSEARCHED for every claim no logged search names; UNKNOWN ID for a search or pair naming a claim (or, with --found, a finding) not on the sheets; exit 1 on either; report — the record on one page: the header that it is not a legal opinion, the claims, the searches, the findings, the matrix, every tie with its why, the UNSEARCHED claims; printed, and written to --out when given — 14 checks in its selftest. It never prints the word novel as a verdict; never reports a claim with no search logged; never drops a finding that overlaps; 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 LAWPrior art is a record of what was looked for and what was found; novelty is a word for a professional. Every search is logged with where, what and when; every finding is tied to the claims it touches as anticipates, overlaps or differs; a claim with no search behind it is UNSEARCHED. The report says so at the top: not a legal opinion.

The method.

Log each search as it is run: the day, the place searched, the query, the number of hits, a note.
List each finding: an id, title, year, where it lives, and in plain words what it shows.
Tie findings to claims in a pairs sheet: anticipates, overlaps or differs, with a line of why; a claim with no search logged is UNSEARCHED and the matrix says so.
Build the matrix: claims down, findings across, the relation in each cell; the overlaps and anticipations listed first.
Report for the professional: the searches, the findings, the matrix, the open claims; the header says it is a record and not an opinion.

What you bring.

The claims, one per line.
The searches run, with where and when.
The findings, with where each lives.

Planned as a separate seat, not in this box and not yet on sale: the studio's own prior-art records for its filings.

What it does once your AI has it.

log Append one search to the searches sheet (created with its header when missing) and print the row back; a bad day, hits below 0 or no claim id is bad input (exit 2)
matrix Claims down, findings across, the relation in each cell (anticipates, overlaps, differs, - for no pair); every ANTICIPATES and OVERLAPS listed first with its why; a pair naming an id not on the sheets is UNKNOWN ID, listed apart, exit 1
check UNSEARCHED for every claim no logged search names; UNKNOWN ID for a search or pair naming a claim (or, with --found, a finding) not on the sheets; exit 1 on either
report The record on one page: the header that it is not a legal opinion, the claims, the searches, the findings, the matrix, every tie with its why, the UNSEARCHED claims; printed, and written to --out when given

What it will never do.

It never prints the word novel as a verdict; never reports a claim with no search logged; never drops a finding that overlaps; never claims a certainty it does not have.

Where it hands off: Whether to file, and where → a patent attorney; The claims and their sources for the paper itself → THE SOURCE CARD; Whether the claim as written overpromises → THE HONEST CARD.

It is not a legal opinion. 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/priorart.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 prior-art records for its filings). Also not in the box: any data you must bring, and advice of any kind (not a legal opinion).

PROOF · SELFTEST 14/14 · 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 16A3547A4315CE8A3A6DC051

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 23 — 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

priorart.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.

`log`	append one search to the searches sheet (created with its header when missing) and print the row back; a bad day, hits below 0 or no claim id is bad input (exit 2)
`matrix`	claims down, findings across, the relation in each cell (anticipates, overlaps, differs, - for no pair); every ANTICIPATES and OVERLAPS listed first with its why; a pair naming an id not on the sheets is UNKNOWN ID, listed apart, exit 1
`check`	UNSEARCHED for every claim no logged search names; UNKNOWN ID for a search or pair naming a claim (or, with --found, a finding) not on the sheets; exit 1 on either
`report`	the record on one page: the header that it is not a legal opinion, the claims, the searches, the findings, the matrix, every tie with its why, the UNSEARCHED claims; printed, and written to --out when given
`selftest`	14 checks that prove the machine; it must print selftest 14/14

THE SELFTEST, AS PRINTED

ok   1  log appends one search row to the sheet (the # line and the earlier rows untouched) and prints it back with its day, place, query, hits and claims
ok   2  log creates a missing sheet with its header, then the row
ok   3  log refuses a malformed day, hits below 0, no claim id and a blank place as bad input (exit 2) and writes nothing
ok   4  matrix: claims down, every finding across, the relation in each cell and - where no pair is written, exit 0
ok   5  matrix lists ANTICIPATES then OVERLAPS first, each with the finding's title, year and the why, before the grid; the differs stay in the grid
ok   6  a finding that overlaps is never left out: with fewer pairs r2 still stands in the OVERLAPS list and in its grid cell
ok   7  matrix lists a pair naming an id not on the sheets as UNKNOWN ID apart, with its why kept, fills no cell from it and exits 1
ok   8  check names UNSEARCHED for the claim no logged search names, with its text, and UNKNOWN ID for the pairs, exit 1
ok   9  check counts a zero-hit search as a search (c3 searched once), names a search for a claim not on the sheet as UNKNOWN ID (exit 1), and says when findings are not checked
ok   10  check with every claim searched and every id known prints no flags and exits 0; searches are listed oldest first
ok   11  report: the header sentence is the line under the title, every section is present, the UNSEARCHED claim and the UNKNOWN ID pair are listed, and --out writes the same text
ok   12  the report never prints the word novel as a verdict (no 'novel:', 'is novel' or 'NOVEL' anywhere in it); only the header's 'novelty is a word for a professional'
ok   13  the same sheets give the same bytes twice, for the report and the matrix
ok   14  a relation outside anticipates|overlaps|differs, the same pair twice, a claim id twice, a sheet without its columns and hits that are not a whole number are bad input (exit 2)
selftest 14/14

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 claims, one per line. The searches run, with where and when. The findings, with where each lives. 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 (14/14), 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 legal opinion.

What will it never do?

It never prints the word novel as a verdict; never reports a claim with no search logged; never drops a finding that overlaps; never claims a certainty it does not have.

← THE METHODS · THE FLOOR →