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

# Setup & Connection

Connect the Dapta MCP in a few minutes. Option A: sign in with your Dapta account from Claude, ChatGPT, Codex or Cursor. Option B: use an API key for scripts and automations.

There are two ways to connect. Both reach the same server, `https://mcp.dapta.ai/mcp`, and give your AI client the same tools and guides. There is nothing to download or install.

***

## Prerequisites

* A **Dapta account**. Signing in reaches the workspaces where you are an **owner or admin**.
* An AI client: **claude.ai** (web or mobile), **Claude Desktop**, **Claude Code**, the **ChatGPT** desktop app, **Codex**, **Cursor**, or another MCP client.

***

## Option A: sign in with your Dapta account

Recommended for people. No key to copy: one connection reaches every workspace you can use.

| Client                                   | How                                                                                                                                                        |
| ---------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **claude.ai, Claude Desktop and Cowork** | Add a custom connector with the URL `https://mcp.dapta.ai/mcp` and sign in. Once Dapta is listed, you can also add it from the Claude connectors directory |
| **Claude Code**                          | Already there if Dapta is connected in your Claude account. Otherwise, one command and `/mcp`                                                              |
| **ChatGPT desktop app**                  | Create an MCP app with the same URL, sign in and install it                                                                                                |
| **Codex**                                | Two commands in the terminal: `codex mcp add` and `codex mcp login`                                                                                        |
| **Cursor**                               | One entry in `mcp.json`, then **Authenticate**                                                                                                             |

Your Dapta sign-in uses your email and a 6-digit code we email you.

**Choosing a workspace.** If your account has one workspace, the AI client uses it. If it has several, it lists them by name and account and asks which one to work in. Tell it by name. To change later, say "switch to \<workspace>".

***

## Option B: API key

For scripts, automations, or a connection fixed to one workspace. The key fixes the workspace, so the client never asks which one to use.

**Step 1:** Create a key in the Dapta admin panel.

{% content-ref url="/pages/zhvtvDJJrqsIADXWzT7U" %}
[Mint your API key](/dapta-mcp/setup/mint-your-api-key.md)
{% endcontent-ref %}

**Step 2:** Connect using the form that matches your client.

| Client                                          | How the key travels                                                                                               |
| ----------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| **claude.ai / Claude Desktop** custom connector | Paste the connector URL the admin panel shows: `https://mcp.dapta.ai/<your-key>/mcp`                              |
| **Claude Code**                                 | `claude mcp add --transport http --scope user dapta-ai https://mcp.dapta.ai/mcp --header "x-api-key: <your-key>"` |
| **Cursor, Codex, custom scripts**               | Header `x-api-key: <your-key>` (or `Authorization: Bearer <your-key>`) on `https://mcp.dapta.ai/mcp`              |

***

## Pick your client

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>claude.ai &#x26; Claude Desktop</strong></td><td>Add a connector and sign in.</td><td><a href="/dapta-mcp/setup/connect-claude-desktop.md">Connect claude.ai &amp; Claude Desktop</a></td></tr><tr><td><strong>Claude Code</strong></td><td>Reuse your Claude connector, or one command.</td><td><a href="/dapta-mcp/setup/connect-claude-code.md">Connect Claude Code</a></td></tr><tr><td><strong>ChatGPT</strong></td><td>Create an MCP app and sign in.</td><td><a href="/dapta-mcp/setup/connect-chatgpt.md">Connect ChatGPT</a></td></tr><tr><td><strong>Codex</strong></td><td>Two commands, or a config block with a key.</td><td><a href="/dapta-mcp/setup/connect-codex.md">Connect Codex</a></td></tr><tr><td><strong>Cursor</strong></td><td>One config entry, then sign in.</td><td><a href="/dapta-mcp/setup/connect-cursor.md">Connect Cursor</a></td></tr></tbody></table>

***

## Verify

Ask **"Who am I in Dapta?"** in a new chat.

{% content-ref url="/pages/y0N0ytQ0C9FIgnF9BZoY" %}
[Verify your connection](/dapta-mcp/setup/verify.md)
{% endcontent-ref %}

***

## What changed from the old setup

* **Every supported client can sign in.** Claude, ChatGPT, Codex and Cursor all connect with your Dapta account. API keys are now the advanced option.
* **The Claude plugin and the skills bundle are retired.** There is nothing to download or upload. Old instructions that mention "Download skills bundle" or "Upload local plugin" no longer apply.
* **Several keys per workspace,** each with its own name, instead of one.
* **Web search tools were removed from the server.** When a guide needs to read a company website, for example to create an agent, your AI client uses its own web tools, or asks you for the details.


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.dapta.ai/dapta-mcp/setup.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.