# Model Context Protocol (MCP)

### What is MCP?

MCP (Model Context Protocol) allows you to connect external tools hosted on third-party platforms (such as Zapier or similar providers) directly with your BotPenguin AI agents.

It acts as a bridge between your AI agent and external capabilities, enabling the bot to perform advanced actions, fetch real-time data, and execute workflows beyond native platform limitations.

***

### Benefits of MCP

* 🚀 Enables your AI agents to interact with external tools and systems seamlessly.
* ⚙️ Extends bot capabilities beyond built-in features using third-party integrations.
* 🔗 Centralizes tool management through a single MCP connection.
* 🧠 Enhances automation by enabling real-time actions and data retrieval.
* 📈 Improves scalability by allowing integration with multiple external services.

***

### How to Use MCP

#### Prerequisites

Before setting up MCP, ensure:

* You have created an AI Engine.
* MCP feature is included in your current plan.
* You have valid MCP credentials from your third-party provider (e.g., Zapier).
* Your MCP tools are already configured on the provider’s platform.

***

#### Part 1: Connecting MCP

1. Go to the “More” section from the left sidebar.
2. Navigate to “Integrations” and select “MCP”.

<figure><img src="/files/20OrHpnBgwj4SXsOQO2i" alt=""><figcaption></figcaption></figure>

3. Click on the “Connect” button to start the setup.

<figure><img src="/files/uLbbO9LwgA871gWey7e4" alt=""><figcaption></figcaption></figure>

4. Enter the following details:
   1. **Configuration Name** – A unique name to identify this MCP connection.
   2. **MCP URL** – Provided by your MCP provider.
   3. **Authentication Method** – Choose one of the following:
      1. **None** – No authentication required.
         1. **Access Token** – Enter the token provided by your MCP provider.
         2. **Custom Headers** – Add key-value pairs required by your provider (multiple entries supported).

<figure><img src="/files/5zELc7zWppQjP4E5NFjh" alt=""><figcaption></figcaption></figure>

5. Click on “Connect” to establish the connection.

***

#### Part 2: Adding MCP Tools to an AI Agent

6. Go to the **“Agents”** module and select your desired agent.

<figure><img src="/files/bSEAKqupj0QODu1HCKiw" alt=""><figcaption></figcaption></figure>

7. Navigate to “Personality & Settings” → “Prompts”.
8. Click inside the text editor and type “/” to open the actions modal.
9. Select “MCP” from the available options.
10. If MCP tools are not added yet, click on “Add Tools”.

<figure><img src="/files/tHfZAuwu8RLAVvyYx0uj" alt=""><figcaption></figcaption></figure>

11. From the dropdown, select your MCP Configuration Name.
12. Choose the tools you want to integrate into your agent.

***

#### Part 3: Using MCP Tools in Prompts

13. Inside the prompt editor, type “/” again.
14. Select “MCP” and choose the required tool.
15. The selected MCP tool will be inserted into your prompt and ready for use.

***

#### Important Notes

* A tool cannot be added multiple times from the same MCP configuration.
* The same tool from different MCP configurations will be treated as separate tools.
* Ensure all credentials and configurations are valid to avoid connection failures.

🎉 Congratulations, you have successfully completed this section! You're one step closer to mastering BotPenguin.

***

### MCP FAQs

<details>

<summary>What is required before setting up MCP in BotPenguin?</summary>

You need an active AI Engine, MCP feature enabled in your plan, and valid MCP credentials (URL, authentication details) from your third-party provider.

</details>

<details>

<summary>Which authentication methods are supported in MCP?</summary>

MCP supports three authentication methods: None (no authentication), Access Token, and Custom Headers (key-value pairs).

</details>

<details>

<summary>Can I connect multiple MCP configurations?</summary>

Yes, you can create and manage multiple MCP configurations, each with its own tools and credentials.

</details>

<details>

<summary>Why am I unable to see my MCP tools in the agent?</summary>

This usually happens if tools have not been added after connecting MCP. Use the “Add Tools” option in the prompt editor to import them.

</details>

<details>

<summary>Can I add the same tool multiple times?</summary>

No, duplicate tools from the same MCP configuration are not allowed. However, the same tool from different MCP configurations will appear separately.

</details>

<details>

<summary>Where can I use MCP tools within the agent?</summary>

MCP tools can be used inside both Core Instructions and Conversation Flow instructions within the prompt editor.

</details>

<details>

<summary>What happens if my MCP credentials are incorrect?</summary>

The connection will fail, and the tools will not be accessible. Ensure your MCP URL and authentication details are accurate.

</details>

<details>

<summary>Do I need to configure tools inside BotPenguin?</summary>

No, tools must be configured on the MCP provider’s platform. BotPenguin only connects and utilizes them.

</details>

***

If you still have questions for our team, write to us at [**support@botpenguin.com**](mailto:support@botpenguin.com). We’ll get back to you within 48 hours.


---

# 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://help.botpenguin.com/settings/integrations/model-context-protocol-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.