如何把個人第二大腦擴展為公司大腦

TL;DR

A Company Brain is a shared notebook that people and AI agents both read. A team needs four things your notes never needed alone: an owner, a reader rule (a check on who may read what), a waiting area and a date on every fact.
The free Second Brain Starter is a one-person start.

I write Code Meet AI, the newsletter where I document how I actually use AI to ship real products, receipts included. New guides land there first.

Your Second Brain works because one person trusts it. Add a second person and nobody checks the notes.

Alone, you know which note is old and which file is the real one. A teammate knows neither, and neither does their agent. Box, the file storage company, says in its Company Brain tutorial that a team needs "a shared source of truth that multiple people and multiple agents can use safely."

My vault is the folder of notes my AI works from. I am its only reader and only owner. I have not run it with a team yet.

A folder that works for one person has no owner

With two people, each of you can believe the other one updates the pricing page.

In my vault each project keeps its facts in one file, FACTS.md, and nowhere else. A skill, my saved instructions for the AI, points to that file and never copies a fact out of it.

For a team, put a name next to each area of work. Box puts it as "assign every domain an accountable owner."

A shared login lets the agent read what the asker may not

I have no reader rule yet.

Picture a new hire asking the company chat for the wifi password. The example is made up. The chat's agent was set up with one shared login to all the documents, so it can read the salary sheet too.

Maxi Contieri, a developer who writes about AI coding, named this in July 2026. Naive integrations, he writes, "flatten access control when they index everything under one service account." Access control means who may see what. Indexing reads every document in advance so the agent can search it. A service account is a program's login.

His fix is that the step which finds documents for an answer "checks permissions against the requester's own identity." The agent answers only from what the person asking may see.

New facts should wait before agents read them

Saved links land in a 06-Dump inbox first. My own /dump command sorts them into notes later. I counted 75 saved reels older than 14 days that I never processed. The waiting area works. I am the part that does not.

Box puts that wait in front of agents too: "New files land in a _staging area; only approved files move to the approved area that agents are pointed at."

Approved facts still age

When a rule in my vault changes, I add a dated note at the top that overrides the text below it.

Say you change the refund window from 30 days to 14. The old page stays approved. The agent keeps promising 30 days. Forty customers a month take it up on a 59-euro refund. With made-up numbers, the sum is 40 times 59, or 2,360 euros a month.

Facts that quietly stop matching reality are drift. Box calls it "the default failure mode of any knowledge base" and says to "pair this with a scheduled review per domain." A review-by date is the day a page stops being trusted until someone checks it again.

The why and the parts each get a longer piece

Pick one area and give it an owner and a date

Before you hand yours to anyone, do the boring step first:

Open the main note of one area, like pricing or the refund policy, and write an owner's name and a review-by date at the top.

This one came from Code Meet AI, the newsletter where I write up the setups I actually run on my own projects.