> 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/mcp.md).

# Start here

Operate your Dapta workspace from Claude, ChatGPT, Codex, Cursor or your own scripts, using plain language. Sign in with your Dapta account or connect with an API key.

The **Dapta MCP** lets you operate your Dapta workspace from inside an AI client such as **claude.ai**, **Claude Desktop**, **Claude Code**, **ChatGPT**, **Codex** or **Cursor**, using plain language. Instead of clicking through the dashboard, you ask Claude to "create a voice agent", "show me last week's connection rate" or "add a step to this flow", and it does it for you.

It is built on the **Model Context Protocol (MCP)**, an open standard for connecting AI assistants to external systems. The server gives Claude two things:

* **Tools:** the actions Claude can take (read data, create agents, update prompts, patch flows, write rows).
* **Guides:** step-by-step playbooks, served by the server itself, that tell Claude which tools to use for a task, in what order, and how to confirm with you before anything changes.

{% hint style="info" %}
There is nothing to download or install. The guides live on the server and Claude loads them when a task needs them.
{% endhint %}

***

## What you can do

| Area                         | Examples                                                                      |
| ---------------------------- | ----------------------------------------------------------------------------- |
| **Voice agents**             | Create, configure and calibrate agents. Improve prompts from real calls       |
| **Voice analytics**          | Call stats, connection rates, best hours to dial, qualified leads, feedback   |
| **Voice sequences**          | Read and adjust outbound calling lists, contacts and retry rules              |
| **Text and WhatsApp agents** | Build chat agents, test them, manage versions, connect a WhatsApp line        |
| **Flow Studio**              | Design, validate, create, patch and run automation flows. Read execution logs |
| **Dapta databases**          | Create databases and tables, read and write the rows automations keep         |
| **Sales meetings**           | Meeting transcripts, action items and statistics                              |
| **Workspace and account**    | Phone numbers, knowledge bases, WhatsApp templates, balance, subscription     |
| **Support and docs**         | Search Dapta's documentation. FAQ and navigation help                         |

You do not need to know any tool names. You describe what you want, and Claude picks the right actions.

***

## Two ways to connect

* **Sign in with your Dapta account.** Recommended for people. Your AI client opens Dapta's sign-in page once and stays connected. There is no key to copy, and one connection reaches every workspace you can use. Works in claude.ai, Claude Desktop, Claude Code, ChatGPT, Codex and Cursor.
* **API key.** For scripts, automations, or a connection fixed to one workspace. The workspace owner creates keys in the Dapta admin panel.

{% content-ref url="/pages/mgrNs2Jy4B82jjw2FyzU" %}
[Setup & Connection](/dapta-mcp/setup.md)
{% endcontent-ref %}

{% content-ref url="/pages/nriNEe6uXqHeIL2IXLB5" %}
[Connect from the Claude connectors directory](/dapta-mcp/claude-connectors-directory.md)
{% endcontent-ref %}

***

## Safe by design

* **Workspace isolation.** Every call runs in one workspace: the one tied to your API key, or the one you chose after signing in.
* **Preview, then confirm.** Whenever Claude changes something, it shows you a preview first and waits for your yes. Reading your data does not need confirmation.
* **Keys stay on the server.** Your Dapta API key and your login session are never returned to Claude.

{% content-ref url="/pages/7Nwd3yZ8uoxBnjaqiVC6" %}
[Safety & Security](/dapta-mcp/safety-and-security.md)
{% endcontent-ref %}

***

## Who this is for

Anyone with a Dapta account who is an **owner or admin** of at least one workspace. Signing in needs nothing else and reaches each of those workspaces. API keys are created by the workspace owner.

{% content-ref url="/pages/LRJbKEAV5HB4G9Q0Ww1G" %}
[Key Concepts](/dapta-mcp/key-concepts.md)
{% endcontent-ref %}


---

# 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/mcp.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.