Guide

Building a second brain with AI: how each step of the CODE method changes

CODE builds a second brain in four steps, Capture, Organize, Distill and Express; once AI tools write to your memory, they capture and organise while you review.

A second brain is a place outside your head for what you know about your work, so you can find it again when you need it. For most of its history that place was a notes app you had to keep tidy yourself. AI tools change the deal. You already explain your work to ChatGPT or Claude every day, and that explanation is exactly what a second brain should hold. The missing piece is somewhere to keep it that every tool can read.

By Musab Alrakhis · Last reviewed: 27 October 2026

This article goes back to the CODE method the second brain is built on and walks through its four steps, one at a time, to show what changes in each once AI tools are part of your day. If you want the build steps themselves, our guide to the AI second brain covers them.

Where does the idea of a second brain come from?

The term was popularised by Tiago Forte, who describes a second brain as a personal system for capturing the ideas and information that matter to you and bringing them back when you do real work. His method is summed up as CODE: Capture, Organize, Distill, Express. The introductory guide on Forte Labs is the clearest overview.

The reasoning holds up. Your brain is good at thinking and poor at storing details for months. Write the details down somewhere reliable and your attention is free for the work itself.

The trouble is upkeep. Many people set up a second brain in a notes app, keep it going for a few weeks and then stop. Capturing, tagging and searching become another job stacked on top of the actual job.

What changed once AI tools arrived?

Think about what you type into an AI tool in a normal week. Who the client is. What you agreed last Tuesday. How the monthly report should be laid out. Which tone you don't want. That is high-value context, and you produce it without trying.

Most of it disappears. A new chat starts from zero, or you switch to a different tool that knows nothing about you. Several tools now offer their own memory, and that helps inside the tool. Claude, for example, keeps a separate memory for each project (Anthropic's help article explains how it works). But each vendor's memory lives with that vendor. What ChatGPT knows about you does not reach Claude, and neither reaches Cursor.

So the useful question is no longer "where should I take notes?" It is "how do I make what I explained once available to every tool I use, while still being able to read and fix it?"

How do the four CODE steps change when your tools do the work?

This is what separates a second brain built on AI tools from the classic version. The four steps stay; who does each one changes.

Capture: from copying by hand to saving inside the chat

In the classic method, you capture an idea yourself by copying it into a notes app. With a tool connected to your memory, capture happens inside the conversation: you and the tool settle a decision, and it saves it as a note. You still decide what gets kept, by asking for the save or approving it.

Organize: from tagging to projects

Forte recommends organising by active project rather than by topic. That's close to how Memory Bridge works: every note goes into a project, and the tool files it in the right one when you name it.

Distill: from highlighting to a short, clear note

Instead of highlighting the key lines yourself, the tool saves the gist: the decision, the reason, the date. You remain the editor and fix the summary if it's off.

Express: from searching the archive to an answer built on your notes

This is the big shift. In the classic method you search your notes, then write. With a second brain connected to your tools, you ask for the report or the email, and the tool searches your notes and writes from them directly.

So the steps that wore people out, capturing and organising, are now part of the work itself. What's left for you is reviewing and judgement.

How does a second brain work inside your tools?

In Memory Bridge your second brain is a set of plain-text Markdown notes, grouped into projects. Every note carries its source and date, so you can see that one came from ChatGPT in March and another from Claude last week.

You connect a tool from its own settings and sign in with a code sent to your email. After that, two things happen while you work as usual:

When you and your tool settle something important, such as a deadline or a project decision, the tool saves it as a note in the right project.
When you ask for something that depends on earlier context, the tool searches your notes by meaning, so it finds the right note even if you phrased it differently back then.

Supported tools today include Claude, ChatGPT, Codex, Cursor, VS Code and Gemini. The connection steps differ by tool, and sometimes by plan within a tool, so each one has its own page under supported tools.

A short example

Say you run reports for a client called Northwind. In the morning you tell Claude: "This client wants every report to open with a one-page summary in plain language." Claude saves that to the client's project.

In the afternoon you open a brand-new ChatGPT chat and type: "Write this month's report for Northwind." ChatGPT opens with a one-page summary, because it found the note. You didn't paste anything and you didn't repeat yourself.

This is an illustration of how it works, not a customer story.

What belongs in your second brain?

Less than you'd think. A crowded second brain is worse than a small one, because the tool keeps tripping over old or conflicting notes. Start with what you keep re-explaining:

Who you are and what you do: your role, your field, the clients or teams you work with.
Project decisions and the reasons behind them: what you agreed, what you turned down, and why.
How you like work done: the shape of your reports, the tone of your emails, the things that annoy you.
Dates and commitments: deadlines and agreed terms.

Keep passwords, access keys and other people's personal data out of it.

Two ways to get started

You don't have to type it all from scratch. Use either route, or both:

An introduction session inside your tool. The tool asks about your work, your clients and how you like to work. Answer by typing or talking, stop whenever you want, and pick it up later from any connected tool.
Import your past conversations. Upload your ChatGPT or Claude export, then choose what gets saved. The import page walks through it.

How is this different from a notes app?

In a notes app, you write, file and search. With a second brain connected to your AI tools, the tools write as you work and search when you ask. You open it when you want to review or correct something.

The notes stay yours and stay visible:

Read and edit any note whenever you like. Each edit keeps the previous version.
See which tool wrote each piece of information, and when.
Stop any tool's access at any time. Your notes stay as they are.
Download a project's notes as Markdown files in a ZIP.

Memory Bridge does not sell your data and does not use the text of your notes to train general-purpose AI models. When a connected tool such as ChatGPT reads from your memory, that tool handles what it reads under its provider's own policy. The details are on the privacy and control page.

What a second brain won't do

Knowing the limits makes it easier to trust:

It does not record every chat automatically. It keeps what you ask it to keep, what you agree on, what you import, and what you answer in the introduction session.
It does not stop a model from being wrong. Check the answers that matter.
It grows with your work. It gets more useful as the correct notes add up, so fix mistakes when you spot them.
It is built for personal use today, not for teams.

Do you need more than one AI tool?

No. One tool is enough. Your second brain is there in every new chat in that tool, so you stop starting from zero. If you try another tool later, connect it and the same notes are waiting.

Your next step

If you explain your work to an AI tool every day, you already have the raw material for a second brain. What's missing is a place that keeps it and shares it with every tool you use. Connect your first tool, start with the introduction session or an import, and carry on working.

Start your second brain with Memory Bridge

How this article was written

The Memory Bridge team drafted this article with help from AI tools and checked it against the live product and the sources linked above in October 2026. Menu names in other tools may change after that.

Common questions

What is a second brain, in one sentence?

A place outside your head for what you know about your work; in Memory Bridge your AI tools write to it and read from it for you, and you can open and edit it any time.

Does it save everything I type to my AI tool?

No. It saves what you ask it to save and what you agree on with your tool, plus anything you import and your answers in the introduction session.

Which tools does it work with?

Supported tools include Claude, ChatGPT, Codex, Cursor, VS Code and Gemini. Connection steps differ by tool, and some depend on your plan in that tool.

Can I take my notes elsewhere?

Yes. Download a project's notes as Markdown files in a ZIP and open them in any text editor.

Is it useful if I only use ChatGPT?

Yes. One tool is enough: your notes are there in every new chat, and any tool you add later sees the same notes.

One memory. Every tool you use.

Pay for a month or a year. Nothing renews automatically.