> For the complete documentation index, see [llms.txt](https://developer.frontitude.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.frontitude.com/mcp/authorization.md).

# Managing your connections

You always sign in on Frontitude, in your own browser, never inside the AI tool. So the tool never sees your password.

1. Your AI tool sends you to Frontitude in a browser window.
2. You sign in with your normal Frontitude account.
3. You approve the connection. The screen names the AI tool that's asking. You choose the workspace this connection works in, and select the tools it may use.
4. Frontitude issues your AI tool a token, tied to that one connection and that one workspace.

## What your AI tool asks for

Your AI tool picks a level: reading your content, or reading and changing it. That's its decision, not yours. What you do control is which tools the connection can use.

## Your connections

Your connections live on the ***Profile*** tab, under ***Connected agents (MCP)***. Each row covers one AI tool connected to one workspace. If you connect the same agent again you get a second row, not a replacement. The old row keeps its access until you disconnect it.

| Column    | What it shows                                      |
| --------- | -------------------------------------------------- |
| Agent     | The AI tool you connected                          |
| Workspace | The workspace this connection works in             |
| Access    | ***Read and write***, ***Read***, or ***Nothing*** |
| Last used | When this connection last made a call              |
| Connected | When you approved it                               |

## Change which tools an agent may use

Click ***Tools*** on the connection and select the tools you want the agent to have. Each change saves immediately. The picker groups the tools by what they do: reading your content, changing it, and removing or reorganizing it. The [tool reference](/mcp/tool-reference.md) says what each tool does.

If you select only the read tools, your agent can read and nothing more. Clear everything and it can't do anything at all. The picker can only narrow what you approved when you first connected. A read-only connection stays read-only no matter what you select here. To widen it, connect again.

## Disconnect an agent

Click ***Disconnect*** on the connection and confirm. It loses access straight away. To get it back, connect again from the agent.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://developer.frontitude.com/mcp/authorization.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.