THE INSTRUMENTS · THE CRAFT GATES · Nº 1340 · EDITION 1 ·Liquid error (snippets/rm-free-up-card line 27): comparison of String with 0 failed
LUND CARDS · THE INSTRUMENTSNº 1340
THE CARD AS TEXT · FOR EYES AND FOR MACHINES
THE FREE-UP CARD · the clear-room seat — frees up a full chat and carries the work into a clean one
WAKES WHEN a chat is getting long, slow or close to its limit and the work has to go on without losing anything · a person says 'free up this chat', 'we're running out of room', 'start a fresh session', 'it forgot what we decided' · a session is about to be handed to a new chat, a new day or another person, and every decision, file and open task must travel
THE LOOP
weigh · the machine weighs the chat: what is taking the room, the heaviest pieces, what was pasted twice, the blobs
clear · stop the bleed: raw output and blobs go to files with exact names; nothing is pasted twice; ask for the figure, not the dump
carry · write the tagged notes (goal, decisions, open tasks with next steps, files by name, rules); the machine builds the carry-forward
check · the carry-forward holds until every section is there, every open task has its next step, every file is named, no secret travels, and it fits its budget
move · open a fresh chat with the carry-forward as its first message and the named files attached; it starts on the first open task
NEVER
Drops a decision, an open task or a file name to make the carry-forward shorter — finished work is cut first
Lets a key, a password or a token travel in a carry-forward
Claims to delete or shrink a chat — a chat's room is freed only by moving the work to a fresh one
IT MAKES ; selftest — 17 checks that prove the machine.
THE LAW Nothing ships until it is proven. A craft gate checks the work against every representative case and every reading rule, records the evidence, and refuses to pass what it has not proven. The maker holds the standard; the gate holds the line.
PROOF SELFTEST 17/17 · witnessed in Claude 2026-10-21 on the studio’s own example inputs · the dated run in the box (RUN.md) · 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. LICENSE-NOTE.md rides in the packet and is the licence.
THE PUNCH · SHA-256 OF SKILL.MD · 6011C37E4791F2B4AD639DC0
CITE THIS CARDTHE FREE-UP CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=free-up-card. Punch 6011C37E4791F2B4AD639DC0. Retrieved 2026-10-21.
What it is.
Frees up a full chat and carries the work into a clean one. Use when a chat is getting long, slow or close to its limit and the work has to go on without losing anything. Not for summarising a document or deleting chats. Pairs with THE SESSION EXTENSION CARD.
It carries a machine that shows its working: ; selftest — 17 checks that prove the machine — 17 checks in its selftest, run in the open below. It never drops a decision, an open task or a file name to make the carry-forward shorter — finished work is cut first; never lets a key, a password or a token travel in a carry-forward; never claims to delete or shrink a chat — a chat's room is freed only by moving the work to a fresh one.
The law, written into the card.
It is in the card itself, so the AI follows it.
THE HOUSE LAW · THE CRAFT GATESNothing ships until it is proven. A craft gate checks the work against every representative case and every reading rule, records the evidence, and refuses to pass what it has not proven. The maker holds the standard; the gate holds the line.
THE LIMIT, ON THE TINIt estimates tokens at about four characters each; the real count depends on the model. It reads the transcript or export you give it and cannot reach into a chat to delete anything: the room is freed by starting a fresh chat with the carry-forward. What it carries is only as complete as the tagged notes.
The method.
Weigh. The machine weighs the chat: what is taking the room, the heaviest pieces, what was pasted twice, the blobs.
Clear. Stop the bleed: raw output and blobs go to files with exact names; nothing is pasted twice; ask for the figure, not the dump.
Carry. Write the tagged notes (goal, decisions, open tasks with next steps, files by name, rules); the machine builds the carry-forward.
Check. The carry-forward holds until every section is there, every open task has its next step, every file is named, no secret travels, and it fits its budget.
Move. Open a fresh chat with the carry-forward as its first message and the named files attached; it starts on the first open task.
What it does once your AI has it.
selftest 17 checks on the studio’s own example inputs; it must print selftest 17/17
What it will never do.
It never drops a decision, an open task or a file name to make the carry-forward shorter — finished work is cut first; never lets a key, a password or a token travel in a carry-forward; never claims to delete or shrink a chat — a chat's room is freed only by moving the work to a fresh one.
Where it hands off: A whole session carried over → THE SESSION EXTENSION CARD. A session budgeted from its start → THE UNLIMITED CARD. Every claim in the hand-off re-measured → THE DOUBLE CHECK CARD.
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. Names of other companies and standards are used only to say what the card works with.
What’s in the buy.
SKILL.md — the door your AI reads first: when it wakes, when it does not, the loop, the never-lines, the house law
CARD.md — the whole card in ten sections, §0 to §9 · TIN.md — what is on the tin, in nine lines
scripts/freeup.py — the machine, Python standard library only, offline; selftest 17/17
RUN.md — the dated run: the selftest as it printed when the card was cut, with the date and the host
references/NOTES-TEMPLATE.md — the tagged notes form the carry-forward is built from
references/THE-FREE-UP-LAW.md — what fills a chat, how to keep it clear, and when to move: seventeen rules
references/THE-METHOD.md — the method in plain words, in the card’s own voice
references/THE-WORKED-EXAMPLE.md — the studio's own long session weighed: 58% attached context, 185 repeated blocks, one 10,000-token export, and the nine lines that would have carried it
START-HERE.md and system/short-prompt.md — the doors, for any AI with or without skills
MY-LEDGER.md — the card’s memory, on your own machine · LICENSE-NOTE.md — the licence
PROOF · SELFTEST 17/17 · AUDIT 12/12 · WITNESSED IN CLAUDE 2026-10-21 ON THE STUDIO’S OWN EXAMPLE INPUTS · THE TEST PROMPTS IN FRESH CHATS OWED · EDITION 1 · PUNCH 6011C37E4791F2B4AD639DC0
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 1340 —Liquid error (snippets/rm-free-up-card line 116): comparison of String with 0 failed
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.
AFTER YOU PAY · SIGN IN WITH THE EMAIL YOU PAID WITH · THE CARD OPENS ON MY CARDS → · CHECKED AGAINST ITS FINGERPRINT · NOTHING TO WAIT FOR
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 STUDIO’S RUN OF 2026-10-21
freeup.py, showing its working.
A small Python program, standard library only. Your AI runs it; you can too. The block below is the machine’s selftest as it printed when the studio ran it from inside the packet, in Claude, on 2026-10-21.
selftest
17 checks that prove the machine; it must print selftest 17/17
selftest · AS PRINTED
$ python3 scripts/freeup.py selftest
FREE IT UP NOW · about 11,003 tokens of 12,000 (92%) · 10,808 can go · 1 repeats · wrote /tmp/freeup-kqu69zo2/FREE-UP.md
ROOMY · about 11,003 tokens of 1,000,000 (1%) · 10,808 can go · 1 repeats · wrote /tmp/freeup-kqu69zo2/FREE-UP.md
ROOMY · about 77 tokens of 1,000 (8%) · 0 can go · 0 repeats · wrote /tmp/freeup-kqu69zo2/FREE-UP.md
CARRY-FORWARD · about 143 tokens of 4,000 · 1 open · 1 files · wrote /tmp/freeup-kqu69zo2/CARRY-FORWARD.md
CLEAR · about 143 tokens · every section there · ready to open the fresh chat with it
REFUSED: an OPEN task without its next step (write it after -> or →): the bundles
REFUSED: a FILE line must carry the file's exact name, not 'zip 1'
REFUSED: a secret is in the DECISION lines: take it out and keep it where secrets belong
HOLD · missing section ## DECISIONS · missing section ## FILES · missing section ## RULES · missing section ## FIRST MESSAGE · open task without a next step: the bundles
HOLD · about 6,987 tokens, over the 2,000 budget: shorten DONE first, then DECISIONS · wrote /tmp/freeup-kqu69zo2/CARRY-FORWARD.md
REFUSED: no file or folder /tmp/freeup-kqu69zo2/nothing.jsonl
selftest 17/17
THE SELFTEST RUNS ON THE STUDIO’S OWN EXAMPLE INPUTS, INVENTED FOR THE TEST; NO REAL COMPANY, PERSON OR ACCOUNT IS NAMED.
QUESTIONS AT THE TABLE
Asked before the wire.
What does it need from me?
Your own files and figures, on your own disk. It starts by asking for them: the machine weighs the chat: what is taking the room, the heaviest pieces, what was pasted twice, the blobs. If something it needs is missing, it marks it and says so — it never fills a gap with a guess.
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 (17/17), run from inside the packet in Claude on 2026-10-21 on the studio's own example inputs and printed below as it ran; the twelve-check audit (AUDIT.md, 12/12); MANIFEST.md with every file hashed. RUN.md in the box is that run, dated. Still to come, as the next receipt: the card's test prompts asked in fresh chats with the card installed. 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. It estimates tokens at about four characters each; the real count depends on the model. It reads the transcript or export you give it and cannot reach into a chat to delete anything: the room is freed by starting a fresh chat with the carry-forward. What it carries is only as complete as the tagged notes.
What will it never do?
It never drops a decision, an open task or a file name to make the carry-forward shorter — finished work is cut first; never lets a key, a password or a token travel in a carry-forward; never claims to delete or shrink a chat — a chat's room is freed only by moving the work to a fresh one.
