THE PARTNER SKILL CARD · Drafts the skill a software company ships beside its connector, so the model uses its tools well and safely.
WAKES WHEN A company is building a connector for Claude · A connector needs a companion skill · A partner listing needs its skill checked
THE LOOP
Take the tool list and jobs.
Run `draft`.
Edit the description and jobs with the product team.
Run `check`.
Fix until it passes.
NEVER
Lets a write act without a yes
Invents a tool
Calls the tools itself
IT MAKES draft — draft SKILL.md and references/TOOLS.md from the tool; check — check a skill against the tool list.
THE STANDARD every write waits for the person's yes; unknown tools are treated as writes; no invented tools or IDs; tools are named with their server prefix; verify by execution; say the limit out loud in the work; nothing hidden; the person decides.
PROOF SELFTEST 16/16 · witnessed 2026-10-13 on material and live on Claude · Edition 1.
THE SEAL 96 of 100 · LEGENDARY · WITNESSED · Trigger 4 · 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 · EE298932B389B3A74D228647
CITE THIS CARDTHE PARTNER SKILL CARD, Edition 1. Lund Studio LLC, Westminster, Colorado. https://lundr.tech/?view=partner-skill-card. Punch EE298932B389B3A74D228647. Retrieved 2026-10-13.
What it is.
Drafts and checks the companion skill a software company ships beside its connector (MCP server): reads the tool list, sorts tools into reads and writes, writes when to use which tool, the order of calls for common jobs, and the rule that every write waits for the person's yes, then checks no tool is invented or left out.
Use when a company is building or listing a connector and wants the model to use it well. Do not use to build the connector itself or to call the tools.
It carries a machine that shows its working: draft — draft SKILL.md and references/TOOLS.md from the tool; check — check a skill against the tool list — 16 checks in its selftest. It will never lets a write act without a yes; never invents a tool; never calls the tools itself.
Eight laws, written into the card.
They are in the card itself, so the AI follows them.
EVERY WRITE WAITS FOR THE PERSON'S YESEvery write waits for the person's yes.
UNKNOWN TOOLS ARE TREATED AS WRITESUnknown tools are treated as writes.
NO INVENTED TOOLS OR IDSNo invented tools or IDs.
TOOLS ARE NAMED WITH THEIR SERVER PREFIXTools are named with their server prefix.
VERIFY BY EXECUTIONVerify by execution, not by reasoning: run the machine and read what it prints before saying a thing is done.
SAY THE LIMIT OUT LOUD IN THE WORKSay the limit out loud in the work: every result names what it did not check.
NOTHING HIDDENNothing hidden: two readers, the person and the AI, see the same words.
THE PERSON DECIDESThe person decides.
What rides in the card.
In the card’s own voice, with a machine that shows its working.
THE GUIDEreferences/READS-AND-WRITES.md
THE MACHINEpartner.py · two commandsdraft — draft SKILL.md and references/TOOLS.md from the tool; check — check a skill against the tool list. Every command prints its working and refuses a bad input; 16 checks in its selftest.
What it does once your AI has it.
draft Draft SKILL.md and references/TOOLS.md from the tool
check Check a skill against the tool list
On the record the 12 tools of the official open-source MCP git server (modelcontextprotocol/servers @ bbb701e7), rebuilt from its source with their real input schemas — real material (not made up for the test). Drafted a companion skill: 7 READ, 5 WRITE, git_reset marked destructive, three jobs in order with Server:tool names. Check: CLEAN, no tool left out, none invented. git_branch was sorted READ from its description ('List Git branches') and labelled for the product team to review. Live on Claude Sonnet, 2026-10-13: 7 of 7 passed; the card fired on 2 of 3 trigger cases and on none of its hand-off cases.
What it will never do.
It will never lets a write act without a yes; never invents a tool; never calls the tools itself.
It sorts tools by their names and reads their schemas. A tool named like a read that changes data will be sorted wrong, so the product team reviews every row. 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: its boot, laws, loop, when it says no, its seals, ledger, term and score · SKILL.md — the door your AI reads first · START-HERE.md — the first five minutes, for you
READS-AND-WRITES.md — the guide in references/, in the card's own voice
scripts/partner.py — the machine: draft · check, standard library only, no network, selftest 16/16
WITNESS.md — the runs on real material and live on Claude, with what was not run · SELFTEST.md — the machine's own test, as run at the cut · MANIFEST.md, every file hashed · CHANGELOG.md
tests/evals.json + EVALS.md — the cases it must take, hand off and refuse
MY-LEDGER.md — the card's memory, on your own machine · TIN.md, the nine lines on the tin · LICENSE-NOTE.md
card.svg + card-machine.svg + card.json · card.html, the card as one offline file · doors/ — the rules files for Cursor, Codex, Gemini, Copilot and Claude Code, the system text for any other AI, a local-model file and an MCP door
PROOF · SELFTEST 16/16 · WITNESSED ON MATERIAL · LIVE ON CLAUDE SONNET 2026-10-13 · 7 OF 7 PASSED · EDITION 1 · PUNCH EE298932B389B3A74D228647
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 1202 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 CARD’S OWN WITNESS.MD
partner.py, every command showing its working.
A small Python program, standard library only. Your AI runs it; you can too. The blocks below are pasted from the card’s own WITNESS.md, as printed.
draft
draft SKILL.md and references/TOOLS.md from the tool
check
check a skill against the tool list
selftest
16 checks that prove the machine; it must print 16/16
THE COMMANDS, IN ORDER, ON THE 12 TOOLS OF THE OFFICIAL OPEN-SOURCE MCP GIT SERVER (MODELCONTEXTPROTOCOL/SERVERS @ BBB701E7), REBUILT FROM ITS SOURCE WITH THEIR REAL INPUT SCHEMAS
THE RUNS ARE PASTED FROM THE CARD’S OWN WITNESS.MD AS PRINTED. FIXTURES MARKED FICTIONAL ARE THE STUDIO’S OWN EXAMPLES, INVENTED FOR THE RECORD; NO PRIVATE PERSON IS NAMED.
QUESTIONS AT THE TABLE
Asked before the wire.
What does it need from me?
The connector's tool list, its server name, the product name and common jobs. Take the tool list and jobs, and it goes from 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 self-test (16/16), a run on real or invented material with every command and verdict pasted into WITNESS.md as printed, and live runs on Claude Sonnet: 7 of 7 passed; the card fired on 2 of 3 trigger cases and on none of its hand-off cases.; 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 sorts tools by their names and reads their schemas. A tool named like a read that changes data will be sorted wrong, so the product team reviews every row.
What will it never do?
It will never lets a write act without a yes; never invents a tool; never calls the tools itself.
