FAQs about the Metricool MCP

The Metricool MCP is the official server that connects your Metricool account to clients that support the Model Context Protocol (MCP). It lets AI tools read and act on your Metricool data, such as posts, analytics, or advertising campaigns.

No. You connect by pasting a URL or copying a command into your AI client. After you connect, you interact with Metricool using natural language instead of writing code.

The Metricool MCP supports the following networks and advertising platforms:

• Instagram (posts, reels, stories)

• TikTok

• LinkedIn

• Facebook (posts, reels, stories)

• X (Twitter)

• YouTube

• Pinterest (pins and boards)

• Threads

• Bluesky

• Twitch

• Meta Ads

• Google Ads

• TikTok Ads

Documented clients include Claude Desktop, Claude Code, ChatGPT, Cursor, Mistral Vibe, Make, Zapier, Smithery, TypingMind, N8N, and Gemini CLI. Claude.ai is compatible as an MCP-capable client.

Start with How to connect Metricool’s MCP. For Claude-specific requirements and setup, see How to connect Metricool's MCP with Claude. For ChatGPT, see How to connect Metricool's MCP with ChatGPT.

MCP works with any Metricool plan, including Free. Your Metricool plan limits still apply after you connect. For Free-plan examples, client-side requirements, and the MCP vs API boundary, see MCP limits and plan requirements.

Remote MCP custom connectors are available on Claude Free, Pro, Max, Team, and Enterprise. Free users can use only one custom connector at a time. Claude Code requires a Claude Pro or Max subscription.

Full Claude compatibility and setup steps: How to connect Metricool's MCP with Claude.

You can authenticate in two ways:

• OAuth: sign in with your Metricool account and approve access from your AI client.

• Token and ID: provide the METRICOOL_USER_TOKEN and METRICOOL_USER_ID variables in your configuration.

Some clients, such as N8N, use header auth with X-Mc-Auth and your Metricool API key. See the client steps in How to connect Metricool’s MCP.

You choose. You can publish content immediately, schedule it for a specific date and time, or use the best-time publishing option recommended by Metricool.

Yes. The MCP includes the get_brands tool, which lets you list and select from the brands associated with your Metricool account.

Yes. You can modify posts that you have already scheduled using the update_schedule_post tool. See Schedule and edit posts through an LLM with Metricool's MCP.

The MCP uses the same permissions and authentication systems as your Metricool account. Revoke access from the AI client or platform where you authorized Metricool. Your data is not stored in the AI model.

MCP is an AI client connector for natural language. It works on any plan, including Free. The Metricool API is for programmatic HTTP access and is available only on Advanced and Custom.

See the full comparison in MCP vs API access: what is the difference? and plan limits in MCP limits and plan requirements.

Yes. The Metricool MCP server is public in the GitHub repository metricool/mcp-metricool and published as a package on PyPI.