# MCP

## Overview

The **MCP Server Registry** in Karini AI allows users to register, manage, and configure various MCP servers within the platform. This feature enables seamless integration of custom or pre-existing servers, offering full control over their configuration. Users can easily reference and manage these servers across different modules, enhancing system flexibility and interoperability.

The registry provides:

* Centralized visibility into server configurations
* Search and listing capabilities
* Operational controls such as **enable/disable**, **test execution**, **edit**, and **removal**

Karini’s MCP Registry supports the following types of MCP servers:

1. [**Local MCP Servers**](/karini-ai-documentation/mcp/local-mcp-servers.md)
2. [**Remote MCP Servers**](/karini-ai-documentation/mcp/remote-server.md)

Once the MCP servers are successfully registered, tested and enabled, the workflow proceeds as follows:

* The MCP servers become available as tools within [**Agent prompts** ](/karini-ai-documentation/prompt-management/agentic-prompts/create-agent-prompt.md#configuring-mcp-server)in the Prompt Playground.
* These tools are selected and used during agent prompt testing to validate functionality and behavior.
* After successful validation, the agent prompts are published.
* The published prompts are then used to configure and build [**recipes**](/karini-ai-documentation/recipes/workflow-recipe.md) and [**copilots**](/karini-ai-documentation/copilots.md), enabling end-to-end workflow execution across the platform.
* In **Recipes**, MCP server access is governed by authorization and recipe type. Users can authorize an MCP server when required and proceed with testing once access is granted. Key behaviors include:
  * In **Chat Recipes**, users can authorize the MCP server if it has not been previously authorized, and then test the recipe.
  * For **Webhook Recipes**, only **service-based remote servers** are supported. 
  * If a user attempts to use a **Remote MCP Server** without authorization, the system displays a **authorization pop-up** to complete the access approval.
*

```
<figure><img src="/files/RyJyWvELrdHH81VFTEN9" alt=""><figcaption></figcaption></figure>
```

* <br>

<br>


---

# Agent Instructions: 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:

```
GET https://karini-ai.gitbook.io/karini-ai-documentation/mcp.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.