How to connect Metricool’s MCP

The screenshots in this guide are in English.

This guide explains the available methods to connect the MCP (Metricool Connector Protocol) with different platforms and AI tools. The MCP is an intelligent tool that integrates with your Metricool account to optimize your social media management, allowing you to schedule content with the best posting times, offer personalized recommendations based on your data, analyze your metrics automatically, and help you save time while improving your results.

What You Need Before Getting Started

Before connecting the MCP, make sure you have the following:

• A Metricool account: The MCP works with any plan, including the free account. Your plan's limitations apply to MCP usage (e.g., free accounts can't query data older than 3 months or schedule more than 20 posts).

• Active account on the tool you want to connect (ChatGPT, Make, Zapier, etc.), with access to advanced features, such as the ability to connect to external sources or perform actions via URL. 👉 If you have questions about the specific requirements for each tool, please check directly with the AI client provider.

• Production URL of the Metricool MCP: https://ai.metricool.com/mcp

Available Connection Methods

Additionally, you can connect Claude Desktop using mcp-remote.

What Permissions the MCP Grants

When you connect the MCP, the AI tool will be able to access the following data and features of your Metricool account:

• View and manage your brands and social network connections

• View and download metrics and analytics

• View, create, and publish posts (both scheduled and already published)

How to connect Metricool’s MCP step by step

To help you connect the Metricool MCP using the method that best fits your workflow, the following sections walk you through each step for all supported clients — including ChatGPT, Make, Zapier, Smithery, Mistral Vibe, Cursor, N8N, Claude Code, and Gemini CLI — whether you're using URL-based connections, header token authentication, or command-line tools.

ChatGPT

Metricool is available as a verified connector in ChatGPT's Apps section. Open ChatGPT, go to Apps, search for Metricool, connect, and authorize access to make Metricool's MCP available in your chats.

See the full guide →

Claude

You can connect Metricool's MCP to Claude Desktop (via mcp-remote) and Claude Code (via the CLI). Each path includes authorization in the browser and a check that the connector is active.

See the full guide →

Make

1. Go to a new scenario in Make.com

2. Add new module and search MCP

3. Crete a connection

4. Add new MCP Server and add the Metricool URL. Click Save

5. Automatically, a new window will open to log in to Metricool and authorize the access

6. Now, you can use the Metricool MCP in both ways: selecting a tool and using it, or adding the task as a prompt, and the MCP will do the actions.

Zapier

1. Go to Zapier.com and start a new Zap

2. Add new Action and search "MCP" and use "MCP Client"

3. Go to Sign In. Add the Metricool MCP URL and continue

4. Login into Metricool and authorize the access

5. Now, you can use the Metricool MCP inside Zapier.

Smithery

1. Go to the Smithery page (free plan) → https://smithery.ai/playground

2. At the bottom, in the text box, click on 'Add tool'.

3. Enter the URL in the search bar and click on 'Connect to MCP server.

4. The next step is to grant the MCP access to use your Metricool account.

5. Once the connection has been established, you will be able to see the available tools and it will be ready for interaction.

Mistral Vibe

Connect Metricool’s MCP in Mistral Vibe by adding a custom connector with OAuth 2.1.

1. Go to the Mistral Vibe page (free plan) → Vibe.

2. Go to Vibe → Connectors (left-hand panel) → + Add Connector.

   

3. You will see featured connectors. To add the Metricool connector, click the Add connector button.

   

4. In Custom MCP Connector, add the following: connector name, connector server (URL), description (optional), and the authentication method, which is OAuth2.1, then click Connect.

   

5. Grant the MCP access to use your Metricool account.

6. At the bottom, you will have your available connectors, ready to use in a chat or to view the available tools.

   

   

Cursor

To be able to add MCPs to Cursor you need the Cursor Desktop app, you can download it here: Download.

After logging into your Cursor account, you have two options:

Metricool MCP install with link

1. Copy any of these links in your browser:

   Deep link:

   cursor://anysphere.cursor-deeplink/mcp/install?name=metricool-mcp&config=eyJ1cmwiOiJodHRwczovL2FpLm1ldHJpY29vbC5jb20vbWNwIiwidHlwZSI6Imh0dHAifQ%3D%3D

   Web link: Auth | Cursor - The AI Code Editor

2. These links will open your Cursor Desktop App and ask if you want to add the MCP server:

   

3. After clicking Install, the MCP server will appear in your Installed MCP tab. However, it shows that it needs authentication, so click Connect:

   

4. It will ask you to open an external URL in your browser to authorize the connection with Metricool:

5. After granting access, the Cursor application will open again, and the MCP will be connected.

Metricool MCP Install without Cursor link

1. The first step is to click on the settings gear at the top right and go to the Tools & MCP section:

   

   

2. Once inside the section, click on Add Custom MCP to open the mcp.json file:

   

   Originally, if no MCP is connected, the mcp.json file has the following structure:

   {
     "mcpServers": {}
   }

   To add an MCP server, in this case Metricool's, you need to insert the following code snippet inside mcpServers:

   "metricool-mcp": {
     "url": "https://ai.metricool.com/mcp",
     "type": "http"
   }

   The final result should be as follows:

   {
     "mcpServers": {
         "metricool-mcp": {
           "url": "https://ai.metricool.com/mcp",
           "type": "http"
         }
     }
   }

   Remember to save the file by clicking Files > Save.

3. Back in the Tools & MCP section, the MCP should appear as unauthenticated, so follow the same process by clicking Connect:

   

4. It will ask you to open an external URL in your browser to authorize the connection with Metricool:

5. After granting access, the Cursor application will open again, and the MCP will be connected.

   

N8N

1. Go to N8N and create new workflow

2. Add new AI Agent node

3. Inside the tools of the AI Agent search MCP Client Tool

4. Add the Metricool URL to the endpoint field
5. Select in Authentication → Header Auth

6. Create new credential.

7. Add the Name → X-Mc-Auth

8. And the value → Your Metricool API Key (You can find in your account settings)

9. Save the Credential and you will have something like this

10. Now you can use the Metricool MCP inside N8N

Although its use is more focused on specific tasks and a more specialized audience, there are also command-line clients capable of using Metricool's MCP.

Gemini CLI

1. To download Gemini CLI you need to open a terminal and run the following command:

npm install -g @google/gemini-cli

2. Once installed, you need to access the configuration file, located in one of the following paths:

• /etc/gemini-cli/system-defaults.json (Linux)

• C:\\ProgramData\\gemini-cli\\system-defaults.json (Windows)

• /Library/Application Support/GeminiCli/system-defaults.json (macOS).
  The path can be overridden using the GEMINI_CLI_SYSTEM_DEFAULTS_PATH environment variable.

3. In the configuration file, add this code snippet:

{
  "mcpServers": {
    "metricool": {
      "httpUrl": "https://ai.metricool.com/mcp"
    }
  },
  "security": {
    "auth": {
      "selectedType": "oauth-personal"
    }
  }
}

4. Save the file and open a Gemini CLI terminal with the gemini command. Run the following command:

/mcp auth metricool

5. A browser tab will open requesting authorization:

6. After granting access, you can return to your terminal and the MCP will be connected.

Related articles

• How to connect Metricool's MCP with ChatGPT

• How to connect Metricool's MCP with Claude

• FAQs about the Metricool MCP

• Schedule and edit posts through an LLM with Metricool's MCP

• Understanding MCPs: What They Are and How They Can Help Your Workflow

• MCP vs API access: what is the difference?