THE CALL NOTES CARD · the call, as a record the team can act on
WAKES WHEN someone finishes a sales call or demo and needs the record, the CRM entry, the follow-up
THE LOOP
the template open during the call
callnotes.py extract · within the hour
fix what it sorted wrong; add the owner
check · complete enough to act on?
the follow-up from the record, their words
NEVER
Write a quote the buyer did not say
Promise in the follow-up what was not agreed
Skip the consent question on a recording
IT MAKES template — the note-taking sheet for the call; extract — the record and the follow-up skeleton; check — complete enough to act on? (next step, date, owner, pain, buyer).
THE STANDARD their words, not ours; a call has a next step or it did not happen; within the hour; the machine sorts, the person checks; a recording needs consent from every party; the follow-up promises nothing new.
PROOF SELFTEST 15/15 · witnessed in Claude 2026-10-09 · 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 · 8006AC0AD687879E252314AE
CITE THIS CARDTHE CALL NOTES CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=call-notes-card. Punch 8006AC0AD687879E252314AE. Retrieved 2026-10-10.
What it is.
Turns the notes or transcript of a sales call into a structured record and a follow-up draft: who was there, what hurts in the buyer's words, what they asked for, what they pushed back on, what was committed, the numbers and dates mentioned, who decides, and the next step with its date and owner — then checks whether the record is complete enough to act on, and asks about consent when a recording is mentioned.
It carries a machine that shows its working: template — the note-taking sheet for the call; extract — the record and the follow-up skeleton; check — complete enough to act on? (next step, date, owner, pain, buyer) — 15 checks in its selftest, every command on the record. It will never write a quote the buyer did not say; never promise in the follow-up what was not agreed; never skip the consent question on a recording.
Six laws, written into the card.
They are in the card itself, so the AI follows them.
THEIR WORDS, NOT OURSThe record quotes the buyer; it never improves on them.
A CALL HAS A NEXT STEP OR IT DID NOT HAPPENWhat, by when, who — before the record is complete.
WITHIN THE HOURNotes a day old are fiction; extract, correct and file the same hour.
THE MACHINE SORTS, THE PERSON CHECKSA sentence can land in the wrong column; the rep who was there fixes it before the CRM sees it.
A RECORDING NEEDS CONSENT FROM EVERY PARTYA recording needs consent from every party, under the law where each of them was. The machine asks; it cannot know.
THE FOLLOW-UP PROMISES NOTHING NEWIt carries what was said and what was agreed, and marks what is still owed.
What rides in the card.
In the card’s own voice, with a machine that shows its working.
THE GUIDEreferences/TAKING-NOTES-ON-A-CALL.md
THE MACHINEcallnotes.py · three commandstemplate — the note-taking sheet for the call; extract — the record and the follow-up skeleton; check — complete enough to act on? (next step, date, owner, pain, buyer). Every command prints its working and refuses a bad input; 15 checks in its selftest, every command on the record.
On the record The template, then an example call sorted into pain, asks, objections, a commitment, a question, the numbers and the Friday next step; the check named the missing owner and the recording to confirm consent on; with the owner named and the consent written down, the record was complete, and the follow-up skeleton drafted from it.
What it will never do.
It will never write a quote the buyer did not say; never promise in the follow-up what was not agreed; never skip the consent question on a recording.
It reads words, not tone, and a sentence can be sorted wrong — the person who was on the call checks the record. It cannot know whether a recording was consented to; it asks. It drafts a follow-up skeleton, never the final email. 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
TAKING-NOTES-ON-A-CALL.md — the guide in references/, in the card's own voice
scripts/callnotes.py — the machine: template · extract · check, standard library only, selftest 15/15
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 15/15 · 3 COMMANDS ON THE RECORD · WITNESSED IN CLAUDE 2026-10-09 · EDITION 1 · PUNCH 8006AC0AD687879E252314AE
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 840 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 2026-10-09
callnotes.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.
15 checks that prove the machine; it must print 15/15
template · THE NOTE-TAKING SHEET FOR THE CALL
$ python3 scripts/callnotes.py template
# Call notes · (account) · (date) · (who from our side)
Attendees (one per line, Name: role):
Name: role
Agenda agreed:
-
Notes (one speaker per line, Name: what they said — write their words, not yours):
Name:
Pain (in their words):
-
Asks / requirements:
-
Objections / concerns raised:
-
Numbers mentioned (team size, volume, budget range, timeline):
-
Who decides / signs off:
-
Next step · what · by when · owner:
-
Recording: (none | recorded with consent from everyone on the call)
extract · THE RECORD AND THE FOLLOW-UP SKELETON
$ python3 scripts/callnotes.py extract call-2026-10-08.txt --out record.json --md followup.md
CALL RECORD · call-2026-10-08.txt · 11 sentences · 2 people
PEOPLE Dana (VP Support) | Sam (us)
PAIN Dana (VP Support): Our team spends hours every week manually tagging tickets. | Dana (VP Support): It's frustrating and error-prone.
ASK Dana (VP Support): We need it to integrate with our help desk, and we'd want an export to CSV. | Dana (VP Support): How many seats would we need for 40 agents?
OBJECTION Dana (VP Support): My concern is the budget this quarter, and we already have a tool that does part of this. | Sam (us): The director signs off on tools over a certain size, so we should loop in procurement.
COMMITMENT Sam (us): I'll send the integration guide and a pricing range by Friday.
NEXT Sam (us): I'll send the integration guide and a pricing range by Friday. | Sam (us): Let's schedule a demo with your admins next week. | Sam (us): The director signs off on tools over a certain size, so we should loop in procurement.
QUESTION Dana (VP Support): How do we know you'll still be around in two years? | Dana (VP Support): How many seats would we need for 40 agents?
NUMBERS 40 agents
DATES this quarter | Friday | next week
ROLES Dana (VP Support): I'd need to run it by my CFO. | Sam (us): The director signs off on tools over a certain size, so we should loop in procurement.
NEXT STEP Sam (us): I'll send the integration guide and a pricing range by Friday. · by Friday · owner (no owner)
GAP · the next step has no owner
GAP · RECORDING · the notes mention a recording or transcript: confirm every party consented, under the law where each of them was, and write it in recording_consent
record written · record.json
follow-up skeleton written · followup.md
$ python3 scripts/callnotes.py check record.json
CHECK · record.json · 2 gap(s)
GAP · the next step has no owner
GAP · RECORDING · the notes mention a recording or transcript: confirm every party consented, under the law where each of them was, and write it in recording_consent
(the rep adds the owner of the next step, Sam, and writes down that every party agreed to the recording at the
$ (the rep adds the owner of the next step, Sam, and writes down that every party agreed to the recording at the start)
$ python3 scripts/callnotes.py check record.json
CHECK · record.json · COMPLETE — ready for the CRM
head -14 followup.md
$ head -14 followup.md
# Follow-up · draft from the call notes
Thank you for the time today. From our conversation:
**What is getting in the way**
- Our team spends hours every week manually tagging tickets.
- It's frustrating and error-prone.
**What you asked for**
- We need it to integrate with our help desk, and we'd want an export to CSV.
- How many seats would we need for 40 agents?
**What you raised, and what I owe you on it**
- My concern is the budget this quarter, and we already have a tool that does part of this. — (your answer here; promise nothing you cannot show)
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?
The template open during the call. It asks first for the notes or transcript as a text file, who was on the call, and whether it was recorded.
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 (15/15) 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. It reads words, not tone, and a sentence can be sorted wrong — the person who was on the call checks the record. It cannot know whether a recording was consented to; it asks. It drafts a follow-up skeleton, never the final email.
What will it never do?
It will never write a quote the buyer did not say; never promise in the follow-up what was not agreed; never skip the consent question on a recording.
