---
name: connect-innernet-claude-desktop
title: connect innernet to Claude Desktop
kind: skill
version: 1.2.0
description: >
  the steps to connect innernet to Claude Desktop, one at a time, with a picture of each screen
  where there is one.
when_to_load: >
  when the person is connecting innernet to Claude Desktop and you are walking them through it.
authored_by: innernet
author_url: https://innernet.live
updated: 2026-10-24
parent: https://innernet.live/connect/ai
---

# connect innernet to Claude Desktop

you are helping someone connect innernet to Claude Desktop. innernet is their one memory for every ai tool, kept current by netti at innernet.live. it signs in through their browser, so there is no key or password in any of this.

the same screens on claude.ai in a browser. on a phone, it is added once on claude.ai and the phone app has it from then on.

give one step per message, with its picture, and wait for "done" before the next. the words carry each step; the picture is a bonus. use the link exactly as given.

## what they paste

the link:

```
https://innernet.live/api/mcp
```

it goes in: Customize → Connectors → Add custom connector → Remote MCP server URL

## steps

1. in Claude, open Customize → Connectors and press Add
![Claude, Customize → Connectors, with the Add button marked](https://innernet.live/connect/v2/claude-mcp-1.jpg)
if you cannot show the picture, say what it shows: Claude, Customize → Connectors, with the Add button marked.

2. name it innernet, paste this link, press Continue
![the Add custom connector dialog with the name and MCP server URL fields marked](https://innernet.live/connect/v2/claude-mcp-2.jpg)
if you cannot show the picture, say what it shows: the Add custom connector dialog with the name and MCP server URL fields marked.

3. press Connect, then allow access in the tab that opens

## so it stops asking

1. open Customize → Connectors → innernet

2. set both tool groups to Always allow
![the innernet connector page in Claude with the Tool permissions dropdowns marked](https://innernet.live/connect/v2/claude-perm-1.jpg)
if you cannot show the picture, say what it shows: the innernet connector page in Claude with the Tool permissions dropdowns marked.

## plan and admin limits

Claude Free allows one custom connector. on Team and Enterprise an owner adds it for the organisation first, then each person connects. (last checked 2026-10-22)

when one of these is what stops them, say so plainly, once, instead of looping.

## the standing instruction

offer this once. it lets the ai reach for innernet on its own.

1. open Settings → General → Instructions for Claude

2. paste the text below and save
![Claude settings, General, the Instructions for Claude box marked](https://innernet.live/connect/v2/claude-instr-1.jpg)
if you cannot show the picture, say what it shows: Claude settings, General, the Instructions for Claude box marked.

```
innernet is connected to this chat as an MCP server — my versioned memory across every AI tool (innernet.live). When a conversation touches a project innernet tracks (its product, strategy, roadmap, brand, design, code, or past decisions), call innernet_load_project before answering — innernet_list_projects finds the slug — and follow the returned capture_protocol. When work produces a decision or insight worth keeping, save it back with innernet_save_context.

When you learn something durable about me (a preference, a tool I rely on, a person in my life, how or when I work, a standing instruction), call innernet_self_capture at that moment, quietly. What you send is kept as your note, not my words: write it in your own voice, name who did what, never write your work or suggestion as mine, and pass my exact words in user_words whenever the note is about something I said, decided or prefer. When several captures pile up, call innernet_self_sync to fold them into facts. When personalising tone, defaults, or assumptions would help, read innernet_self_facts. Surface a relevant memory when it helps, and flag anything that looks stale or contradicted instead of asserting it.

When you save, write the present: if what you save supersedes something a dimension states, replace it in the body (never leave old and new side by side) and add one dated line under a "## history" heading at the end of that dimension — "- YYYY-MM-DD — <what changed>: was … → now …".

Keep all of this quiet and ambient — don't narrate the bookkeeping unless I ask.
```

## prove it

the sign-in finishes in the browser, and new tools load in a new chat or after a restart. once innernet's tools are in your list, call `innernet_list_projects` once and say in a line what you see. they will see Claude Desktop arrive on innernet's connections page.

if the tools are not there yet, ask them to open a new chat (or restart) and say: "innernet, what do you know about me?". until then, do not say it worked.

if nothing works, they can run `npx -y innernet connect` in their own terminal.

## never

- never ask for a password, a code from an email, or an api key. the sign-in happens in their browser, on innernet.live.
- never remove or rewrite other servers in a config file. add innernet beside them.
- never press a confirm or allow button for them in a settings page. they press it.
- never invent a menu that is not there. if a step does not match what they see, ask what they see and adapt.

pictures captured 2026-09-11. if what they see differs, trust what they see and ask.

every tool, and how to tell where you run: https://innernet.live/connect/ai