THE HOME GROUND · THE TRAILHEAD · Nº 1210 · EDITION 1 · LAUNCHING SOON
LUND CARDS · THE HOME GROUNDNº 1210
THE CARD AS TEXT · FOR EYES AND FOR MACHINES
THE COURSE BOOK · Race courses kept one fact at a time, each with the day it was walked and who said so.
WAKES WHEN Someone is writing up a race course and wants every fact dated and sourced. · A course page needs checking for facts nobody has walked or that have gone stale. · Two courses need comparing on the facts both actually have.
THE LOOP
Log each fact with its source and walk date
Check for UNWALKED, STALE and MISSING
Write the course page, flags inline
Compare courses on shared facts only
NEVER
Never calls a director's or map's fact walked
Never fills a missing fact by guessing
Never drops a STALE or UNWALKED flag
IT MAKES add — append one fact after checking its type, unit, source and date; check — per-race table of UNWALKED, STALE and MISSING facts (exit 1 if anything is flagged); page — Markdown course page: every fact with its date and source, flags inline; compare — two races side by side (latest edition of each), only the facts both have; export — every fact, sorted, as CSV, or as JSON with --json.
THE LAW A course book is only as good as its freshest walk. Every fact carries where it came from and the day someone last checked it. A fact from the director or a map is useful and printed as such; it is never promoted to walked.
PROOF SELFTEST 14/14 · witnessed in Claude 2026-10-13 on fictional fixtures · the test prompts in fresh chats owed · Edition 1.
THE SEAL 96 of 100 · LEGENDARY · WITNESSED · Trigger 5 · Machinery 5 · Law 5 · Portability 5 · Proof 4
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 · 2092D50C2631392033575097
CITE THIS CARDTHE COURSE BOOK, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=course-book-card. Punch 2092D50C2631392033575097. Retrieved 2026-10-13.
What it is.
Use when building or checking a race course book from facts walked on foot: distance, climbs, aid, cutoffs, parking. Not for picking a race to enter or for training plans.
It carries a machine that shows its working: add — append one fact after checking its type, unit, source and date; check — per-race table of UNWALKED, STALE and MISSING facts (exit 1 if anything is flagged); page — Markdown course page: every fact with its date and source, flags inline; compare — two races side by side (latest edition of each), only the facts both have; export — every fact, sorted, as CSV, or as JSON with --json — 14 checks in its selftest. It never calls a director's or map's fact walked; never fills a missing fact by guessing; never drops a STALE or UNWALKED flag.
The law, written into the card.
It is in the card itself, so the AI follows it.
THE LAWA course book is only as good as its freshest walk. Every fact carries where it came from and the day someone last checked it. A fact from the director or a map is useful and printed as such; it is never promoted to walked.
The method.
Keep one row per fact, not one row per race: distance, climb, high and low points, surface, aid stations, cutoffs, parking, start time, restrooms, hazards.
Give every fact a source — walked, director, map or other — and, only for walked facts, the date it was walked.
Run the check before anything is published: facts never walked print UNWALKED, walks older than the stale window print STALE, and a course missing distance, climb or cutoff prints MISSING.
Write the course page from the ledger. Every fact keeps its date and source on the page, and flags stay inline where a reader will see them.
Compare two courses only on the facts both books hold; say what one has and the other lacks.
What you bring.
Your own walks: the facts you checked on foot, with dates.
Facts from directors and maps, labelled as such.
The races you keep, by name and year.
Planned as a separate seat, not in this box and not yet on sale: a course book the studio walks and refreshes each season for the races it covers.
What it does once your AI has it.
add Append one fact after checking its type, unit, source and date
check Per-race table of UNWALKED, STALE and MISSING facts (exit 1 if anything is flagged)
page Markdown course page: every fact with its date and source, flags inline
compare Two races side by side (latest edition of each), only the facts both have
export Every fact, sorted, as CSV, or as JSON with --json
What it will never do.
It never calls a director's or map's fact walked; never fills a missing fact by guessing; never drops a STALE or UNWALKED flag.
Where it hands off: Choosing which race to enter → THE FIRST RACE; Race-day weather → THE WEATHER CALL; Running the event itself → THE RACE DIRECTOR.
It is not safety advice. 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/coursebook.py — the machine, Python standard library only; `selftest` proves it on its own fixtures.
tests/EVALS.json and scripts/evals.py — 8 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: a course book the studio walks and refreshes each season for the races it covers). Also not in the box: any data you must bring, and advice of any kind (not safety advice).
PROOF · SELFTEST 14/14 · WITNESSED IN CLAUDE 2026-10-13 ON FICTIONAL FIXTURES · THE TEST PROMPTS IN FRESH CHATS OWED · EDITION 1 · PUNCH 2092D50C2631392033575097
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 1210 — 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 CARD’S OWN SELFTEST.MD
coursebook.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.
add
append one fact after checking its type, unit, source and date
check
per-race table of UNWALKED, STALE and MISSING facts (exit 1 if anything is flagged)
page
Markdown course page: every fact with its date and source, flags inline
compare
two races side by side (latest edition of each), only the facts both have
export
every fact, sorted, as CSV, or as JSON with --json
selftest
14 checks that prove the machine; it must print selftest 14/14
THE SELFTEST, AS PRINTED
ok 1 add appends a valid fact
ok 2 add rejects a fact_type outside the vocabulary (exit 2)
ok 3 add rejects a unit not allowed for the fact (exit 2)
ok 4 add rejects a bad walked_on date (exit 2)
ok 5 add refuses a walk date on a director fact (exit 1)
ok 6 check marks blank walked_on UNWALKED and skips # lines
ok 7 check marks a walk older than 400 days STALE
ok 8 check marks a race without cutoff MISSING
ok 9 --stale-days override clears the STALE flag
ok 10 page opens with the keeper line and keeps director and walked distance apart
ok 11 page never labels a director fact walked
ok 12 compare shows only facts both races have
ok 13 export --json is valid and byte-identical across runs
ok 14 page refuses a race with no facts on file (exit 1)
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?
Your own walks: the facts you checked on foot, with dates. Facts from directors and maps, labelled as such. The races you keep, by name and year. 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 2026-10-13 on fictional fixtures and pasted into SELFTEST.md as printed; the twelve-check audit (AUDIT.md). Owed, and said so in WITNESS.md: The 8 test prompts asked in fresh chats with the card installed. Until they are run and scored, the tests are owed. 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 safety advice.
What will it never do?
It never calls a director's or map's fact walked; never fills a missing fact by guessing; never drops a STALE or UNWALKED flag.
