THE ISSUE TREE CARD · the problem, broken down clean
WAKES WHEN a problem is being broken down, or a team is planning which analyses to run
THE LOOP
write the question; split it
tree · every split checked
leaves · a hypothesis and its test each
rank · do first, plan, quick, park
run the analyses in that order
NEVER
Pass a split with one branch
Call a leaf a fact before its test
Rank a leaf no one has tagged
IT MAKES tree — the numbered tree and every split that is not clean; leaves — each leaf as a hypothesis with its test, or NO TEST; rank — leaves by impact and ease: DO FIRST, PLAN, QUICK, PARK.
THE STANDARD start with the question; every split is clean; leaves are hypotheses; no test, no analysis; impact and ease decide the order; the tree is the team's thinking.
PROOF SELFTEST 9/9 · 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 · DE441BE4097D6487C7197550
CITE THIS CARDTHE ISSUE TREE CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=issue-tree-card. Punch DE441BE4097D6487C7197550. Retrieved 2026-10-10.
What it is.
Breaks a client's problem down so the parts do not overlap and nothing is left out: tree reads an indented outline — the question at the top — as a numbered tree and names every split that is not clean: a top that is not a question, a branch with one child, siblings that say the same thing, more than five branches, deeper than four levels; leaves lists the leaves as hypotheses with the analysis that would test each, and names the ones with no test; rank orders the leaves tagged by impact and ease — DO FIRST, PLAN, QUICK, PARK — and names the untagged.
It carries a machine that shows its working: tree — the numbered tree and every split that is not clean; leaves — each leaf as a hypothesis with its test, or NO TEST; rank — leaves by impact and ease: DO FIRST, PLAN, QUICK, PARK — 9 checks in its selftest, every command on the record. It will never pass a split with one branch; never call a leaf a fact before its test; never rank a leaf no one has tagged.
Six laws, written into the card.
They are in the card itself, so the AI follows them.
START WITH THE QUESTIONThe top of the tree is the client's question, ending in a question mark.
EVERY SPLIT IS CLEANTwo to five branches that do not overlap; a single child is not a split, and overlapping siblings are named with the words they share.
LEAVES ARE HYPOTHESESEach leaf is something that might be true, with the analysis that would prove or disprove it.
NO TEST, NO ANALYSISA leaf without a test is named; the team writes one before work starts.
IMPACT AND EASE DECIDE THE ORDERHigh impact first; between equals, the easier.
THE TREE IS THE TEAM'S THINKINGThe card checks its shape; the team owns what is in it.
What rides in the card.
In the card’s own voice, with a machine that shows its working.
THE GUIDEreferences/THE-CONSULTANCY.md
THE MACHINEissuetree.py · three commandstree — the numbered tree and every split that is not clean; leaves — each leaf as a hypothesis with its test, or NO TEST; rank — leaves by impact and ease: DO FIRST, PLAN, QUICK, PARK. Every command prints its working and refuses a bad input; 9 checks in its selftest, every command on the record.
What it does once your AI has it.
tree The numbered tree and every split that is not clean
leaves Each leaf as a hypothesis with its test, or NO TEST
rank Leaves by impact and ease: DO FIRST, PLAN, QUICK, PARK
On the record A clean tree numbered with no faults; its four leaves read as hypotheses with the one lacking a test named; ranked — flour and butter first; a muddled draft named for a top that is not a question, a one-child split and two overlapping branches.
What it will never do.
It will never pass a split with one branch; never call a leaf a fact before its test; never rank a leaf no one has tagged.
Overlap is read from shared words; two branches can overlap in meaning with no word in common, and share words without overlapping. Whether the tree is exhaustive — whether anything is missing — the card cannot know; the team asks "what else could it be?" at every split. 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/issuetree.py — the machine: tree · leaves · rank, standard library only, selftest 9/9
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 9/9 · 3 COMMANDS ON THE RECORD · WITNESSED IN CLAUDE 2026-10-10 · EDITION 1 · PUNCH DE441BE4097D6487C7197550
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 1102 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
issuetree.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.
tree
the numbered tree and every split that is not clean
leaves
each leaf as a hypothesis with its test, or NO TEST
rank
leaves by impact and ease: DO FIRST, PLAN, QUICK, PARK
selftest
9 checks that prove the machine; it must print 9/9
(the team's issue tree written)
$ (the team's issue tree written)
tree · THE NUMBERED TREE AND EVERY SPLIT THAT IS NOT CLEAN
$ python3 scripts/issuetree.py tree tree.md
Q Why have the bakeries' profits fallen this year?
1 Revenue is down
1.1 Fewer customers come in
1.2 Each customer spends less
2 Costs are up
2.1 Flour and butter cost more
2.2 More bread is thrown away
7 nodes · 4 leaves · every split is clean
leaves · EACH LEAF AS A HYPOTHESIS WITH ITS TEST, OR NO TEST
$ python3 scripts/issuetree.py leaves tree.md
1.1 Fewer customers come in
test: footfall by month, this year against last
1.2 Each customer spends less
test: average ticket by month
2.1 Flour and butter cost more
test: supplier invoices, price per kilo by quarter
2.2 More bread is thrown away
NO TEST — write the analysis that would prove or disprove it
4 hypotheses · 3 with a test · 1 without
rank · LEAVES BY IMPACT AND EASE: DO FIRST, PLAN, QUICK, PARK
$ python3 scripts/issuetree.py rank tree.md
DO FIRST 2.1 Flour and butter cost more (impact H · ease H)
DO FIRST 1.1 Fewer customers come in (impact H · ease M)
PLAN 2.2 More bread is thrown away (impact H · ease L)
QUICK 1.2 Each customer spends less (impact M · ease H)
(a muddled first draft)
$ (a muddled first draft)
tree · THE NUMBERED TREE AND EVERY SPLIT THAT IS NOT CLEAN
$ python3 scripts/issuetree.py tree muddled.md
Q Profit is down
1 Revenue
1.1 Customers
2 Costs
3 Costs of goods
QUESTION the top of the tree is a question; end it with ?
OVERLAP 2 and 3 say much the same: cost
ONE CHILD 1 Revenue — a split needs two branches or none
5 nodes · 3 leaves · 3 splits to mend
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?
Write the question; split it. It asks first for the question, and the outline as the team has it.
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 (9/9) 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. Overlap is read from shared words; two branches can overlap in meaning with no word in common, and share words without overlapping. Whether the tree is exhaustive — whether anything is missing — the card cannot know; the team asks "what else could it be?" at every split.
What will it never do?
It will never pass a split with one branch; never call a leaf a fact before its test; never rank a leaf no one has tagged.
