/community/guides/connecting your ai
innernet
docsskillscommunitycompanydeveloperssecurityprivacytermscontact

connecting your ai · step 1 of 5

connect your ai tools

claude, chatgpt, cursor, claude code and the rest, each connected once, and how to check it worked.

/community/guides/starter-connect-your-ais
1 of 5
step 1 of 54 minconnect your ai toolsclaude, chatgpt, cursor, claude code and the rest, each connected once, and how to check it worked.
step 2 of 55 minstart in one commandconnect every ai tool on your machine to one memory, see what the command does at each step, and know what to try the moment it finishes.
step 3 of 54 minwhat to say to your aionce a tool is connected there is nothing to learn. the things people say most, what each one does, and what you'll see.
step 4 of 51 minbring what you already haveyour ai tools have been remembering things in their own files. bring that in once, and let netti sort it.
step 5 of 52 minthe chat windowthe small innernet line inside your ai chats, what it shows when you open it, and how to set it the way you like.

connecting a tool means telling it where your memory is. every tool is told the same thing, one link:

https://innernet.live/api/mcp

the first time a tool uses it, a browser window opens and asks you to sign in to innernet and allow it. that's the whole security model: you never paste a password or a key into a chat. if an ai ever asks you for one to "connect innernet", don't. give it the link instead.

where you do it

on your dashboard, open connections (the fifth word along the top). the mcp chip lists every tool innernet knows, grouped the way you use them.

the connections space. each tool shows its connections and whether they're live.
the connections space. each tool shows its connections and whether they're live.

a tool you haven't connected says connect →. press it and a small window walks you through that exact tool, in five tabs: name, connect, permissions, instructions, import.

claude (the app or claude.ai)

the connect window: name, connect, permissions, instructions, import.
the connect window: name, connect, permissions, instructions, import.
  1. 1name: call this connection something you'll recognise, like claude on my laptop.
  2. 2connect: in claude, open Customize → Connectors → Add. name it innernet, paste the link, press Connect, and sign in to innernet when the window opens. then press i've done this.
  3. 3permissions: let it work on its own, so it doesn't ask you before every read.
  4. 4instructions: copy the short instruction into claude's settings. it tells claude to keep you in mind. copy the first project prompt too; you'll use it in the next guide.
  5. 5import: bring what claude already remembers about you, or press done.
in claude itself: add a custom connector, name it innernet, paste the link, continue.
in claude itself: add a custom connector, name it innernet, paste the link, continue.

chatgpt

chatgpt connects through its plugins, which need developer mode, a feature of paid plans:

  1. 1in chatgpt, open Settings → Security and login and turn on Developer mode.
  2. 2open chatgpt.com/plugins, press +, and paste the link. tick I understand and press Create.
  3. 3sign in to innernet when it asks, and allow.
  4. 4in any chat, type @innernet (or press + and pick innernet) and allow it once.
in chatgpt: plugins, then + to add innernet by its link.
in chatgpt: plugins, then + to add innernet by its link.

if your workspace has developer mode switched off, you can still bring chatgpt's memory in with an export. see bring what you already have.

everything on your computer, at once

if you use tools that live on your computer (claude code, cursor, codex, windsurf, vs code, gemini cli, zed and others), one command connects all of them:

  1. 1open the terminal app. on a mac, press ⌘ space and type terminal.
  2. 2paste this and press enter:
npx innernet
one command: it signs you in, then writes the one line each tool needs.
one command: it signs you in, then writes the one line each tool needs.
  1. 1your browser opens. sign in and approve. then come back to the terminal: it lists every tool it found and what it wrote, and ends with what to say first.
  2. 2restart any tool that was already open, so it reads its new settings.

every step of that run, and what it wrote where, is in start in one command.

run it inside a code project and that project gets a memory of its own. that's step 10.

check that it worked

open a new chat in the tool you just connected and say:

the simplest test: ask it to load your projects.
the simplest test: ask it to load your projects.

it should answer with your projects, or tell you there aren't any yet. either answer means it's connected. what to say after that, with examples, is what to say to your ai.

back on your dashboard, the connection now reads live or active. one that says waiting for its first call is set up but hasn't spoken yet; send it a message.

when something doesn't connect

  • the tool doesn't mention innernet at all. start a brand-new chat. most tools only read their connections when a chat begins.
  • it asks for a key or a password. it's confused. give it the link, and let the browser sign-in run.
  • a tool is at the door. sometimes a tool asks to connect before you've set it up. your dashboard says {tool} is at the door: press let it in, or not now.
  • still stuck? every per-tool detail is in connect an ai in the docs.

what just happened

each tool you connected can now read your memory and add to it. none of them can see your private facts, ever; you'll see what that means in step 8.

next: find your way around the dashboard.



every chapter

the fundamentals5 guidesconnecting your ai5 guidesthe dashboard4 guidesfor builders3 guidessharing3 guides

one memory, every ai tool.

innernet is memory your ai tools read live: every project, every decision, every skill, in one place, connected once.

start your innernetevery guide