The Model Context Protocol (MCP) allows you to connect AI assistants, like Claude or OpenClaw, directly to your Post for Me account. By running our MCP server, you can empower AI agents to search documentation, write code, and execute social media posts on your behalf using natural language commands.
Here is a breakdown of how to set up the server and what to expect regarding AI access.
Installation & Setup
The Post for Me MCP Server is designed to run locally or remotely, acting as a bridge between your AI client and our API.
How it works:
You can run the server directly using npx. This eliminates the need for a complex installation process. You simply need to provide your API Key as an environment variable to authenticate the requests.
The Commandexport POST_FOR_ME_API_KEY="My API Key"npx -y post-for-me-mcp@latest
Requirements
You must have a valid API Key from your Post for Me dashboard before starting the server.
Client Integration
Once the server is running, you need to configure your AI client (such as Claude Desktop, Cursor, or VS Code) to recognize the new tools.
How it works
Most clients use a configuration file (like mcp.json or .claude.json) to define available servers. You will add "post_for_me_api" to your mcpServers list, specifying the command and your API key in the environment settings.
The User Experience
Once connected, you can ask your AI to "Schedule a post for Instagram next Tuesday" or "Check the analytics for my last YouTube video." The AI will utilize the Docs Search tool to understand the API and the Code tool to generate and execute the necessary TypeScript SDK code in a sandboxed environment.
Security & Access Control
Granting an AI agent access to your API key is a powerful capability that requires careful management.
Full Access
It is important to note that by setting up this MCP server, you are giving the AI full access to any accounts connected to your Post for Me project.
The AI can query all social data.
The AI can publish to any connected account.
The AI can delete scheduled content.
Note: Because the agent executes code using your administrative API key, there is no "user-level" restriction. Ensure you trust the environment and the prompts you are providing to the chatbot.