THE INTERVIEW CARD · stakeholder interviews, read one way
WAKES WHEN stakeholder or customer interviews are planned, run or synthesised
THE LOOP
guide --topics · one guide for everyone
interview; notes under a code
agree the themes and their words
code · each theme counted and quoted
matrix · who raised what; what no one did
NEVER
Keep notes under a person's name
Quote without file and line
Count a theme by a word inside another
IT MAKES guide — the guide, each part with its minutes; code — each theme: how many raised it, and quotes by file and line; matrix — themes by interviewee; themes no one raised.
THE STANDARD one guide; codes, not names; themes are agreed before coding; counted by people, not mentions; every count has its quotes; silence is a finding.
PROOF SELFTEST 8/8 · witnessed in Claude 2026-10-10 · 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 · ED3A3A0575B8C981E028A9F9
CITE THIS CARDTHE INTERVIEW CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=interview-card. Punch ED3A3A0575B8C981E028A9F9. Retrieved 2026-10-10.
What it is.
Runs the stakeholder interviews that start an engagement and reads them all the same way: guide writes an interview guide timed to the meeting — an opening that says notes are kept under a code and never a name, an open question and a probe per topic, a closing; code reads a folder of interview notes (one file per interview, named by its code) against the themes the team agreed and gives each theme the number of interviewees who raised it with quotes by file and line; matrix shows themes by interviewee and names themes no one raised; both name any line that looks like an address, a number or a link so it can be taken out.
It carries a machine that shows its working: guide — the guide, each part with its minutes; code — each theme: how many raised it, and quotes by file and line; matrix — themes by interviewee; themes no one raised — 8 checks in its selftest, every command on the record. It will never keep notes under a person's name; never quote without file and line; never count a theme by a word inside another.
Six laws, written into the card.
They are in the card itself, so the AI follows them.
ONE GUIDEEvery interviewee hears the same opening, the same questions and the same probes, so their answers can be read together.
CODES, NOT NAMESNotes are kept under codes (I-01, I-02); the opening says so, and a line that looks like an address, a number or a link is named to take out.
THEMES ARE AGREED BEFORE CODINGThe team writes each theme and its words; the card counts, it does not invent themes.
COUNTED BY PEOPLE, NOT MENTIONSA theme raised five times by one person counts once.
EVERY COUNT HAS ITS QUOTESEach theme is quoted with its file and line so the reader can check.
SILENCE IS A FINDINGA theme no one raised is named.
What rides in the card.
In the card’s own voice, with a machine that shows its working.
THE GUIDEreferences/THE-CONSULTANCY.md
THE MACHINEinterviews.py · three commandsguide — the guide, each part with its minutes; code — each theme: how many raised it, and quotes by file and line; matrix — themes by interviewee; themes no one raised. 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.
guide The guide, each part with its minutes
code Each theme: how many raised it, and quotes by file and line
matrix Themes by interviewee; themes no one raised
On the record One guide timed to forty-five minutes; three interviews coded — waste, ordering and staff each raised by two people and quoted by file and line; the matrix named price as raised by no one.
What it will never do.
It will never keep notes under a person's name; never quote without file and line; never count a theme by a word inside another.
Coding by words finds what the team named; a theme said in other words is missed, and a word can appear where the theme is not meant. A count of interviewees is not a measure of a population. Consent to interview and to keep notes is the team's to ask for and record. 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-CONSULTANCY.md — the guide in references/, in the card's own voice
scripts/interviews.py — the machine: guide · code · matrix, 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 2026-10-10 · EDITION 1 · PUNCH ED3A3A0575B8C981E028A9F9
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 1103 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 2026-10-10
interviews.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.
guide
the guide, each part with its minutes
code
each theme: how many raised it, and quotes by file and line
matrix
themes by interviewee; themes no one raised
selftest
8 checks that prove the machine; it must print 8/8
guide · THE GUIDE, EACH PART WITH ITS MINUTES
$ python3 scripts/interviews.py guide --topics ordering,waste,staffing --minutes 45
0′ opening Thank them; say what the interview is for, how long it takes, that notes are kept under a code and never their name; ask if that is all right.
5′ ordering Tell me about ordering — how does it work today, from where you sit?
probe: Can you give me the last time that happened? What would you change first?
17′ waste Tell me about waste — how does it work today, from where you sit?
probe: Can you give me the last time that happened? What would you change first?
29′ staffing Tell me about staffing — how does it work today, from where you sit?
probe: Can you give me the last time that happened? What would you change first?
40′ closing What should we have asked and did not? Who else should we speak to? Thank them; say what happens next.
3 topics · 45 minutes
(three interviews' notes under codes, and the agreed themes)
$ (three interviews' notes under codes, and the agreed themes)
code · EACH THEME: HOW MANY RAISED IT, AND QUOTES BY FILE AND LINE
$ python3 scripts/interviews.py code notes --themes themes.txt
waste · raised by 2 of 3
I-01:1 "Too much bread is thrown away on Mondays."
I-02:2 "Waste is worst after rain."
ordering · raised by 2 of 3
I-02:1 "Customers would pre-order if they could."
I-03:2 "We never know the order numbers until the morning."
staff · raised by 2 of 3
I-01:2 "The rota changes too often."
I-03:1 "The shift handover is rushed."
price · raised by 0 of 3
3 interviews · 4 themes; raised by no one: price
matrix · THEMES BY INTERVIEWEE; THEMES NO ONE RAISED
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?
Guide --topics · one guide for everyone. It asks first for the topics, the length of the interviews, and afterwards the folder of notes and the agreed themes.
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. Coding by words finds what the team named; a theme said in other words is missed, and a word can appear where the theme is not meant. A count of interviewees is not a measure of a population. Consent to interview and to keep notes is the team's to ask for and record.
What will it never do?
It will never keep notes under a person's name; never quote without file and line; never count a theme by a word inside another.
