How to Set Up Sapt MCP Integrations

Want your AI assistant to actually work with your business data? The Model Context Protocol (MCP) makes it possible. This tutorial walks you through setting up Sapt MCP integrations so tools like Claude, ChatGPT, and Cursor can connect directly to your marketing stack.

What You Will Need

• A free Sapt account (sign up here)
• Homebrew installed on your machine (brew.sh)
• An MCP-compatible AI tool (Claude Desktop, Cursor, VS Code, etc.)

Step 1: Create Your Sapt Account

Head to app.sapt.ai/sign-up and create a free account. You will need this to generate your MCP credentials.

Step 2: Navigate to Integrations

1. Log into your Sapt dashboard
2. Click your profile icon in the top right
3. Select Account from the dropdown
4. Click the Integrations tab

Step 3: Add MCP Servers

In the Integrations tab, you will see all available MCP servers. Click Add on any server you want to connect to your AI tools.

Available MCP Servers

Sapt provides 9 specialized MCP servers:

• CMS Server - Create, edit, and publish content directly from AI
• Email Server - Draft and send emails with AI assistance
• Meta Server - Manage Instagram and Facebook posts, DMs, and automation
• Leads Server - Access and enrich your lead database
• SERP Server - Search engine data and rankings
• GBP Server - Google Business Profile management
• Support Server - Handle tickets and customer inquiries
• Codebase Server - Search and explore your code repositories
• Project Server - Project settings and configuration

Step 4: Configure Your AI Tool

After adding a server, Sapt will provide you with the connection details. Copy these into your AI tools MCP configuration. The exact steps depend on your tool:

• Claude Desktop - Add to your claude_desktop_config.json
• Cursor - Add in Settings then MCP Servers
• VS Code - Configure via Copilot MCP settings

What Is MCP?

MCP is an open-source standard created by Anthropic that lets AI applications connect to external systems. Think of it like USB-C for AI - one universal connector that works everywhere.

Before MCP, every AI integration required custom code. MCP standardizes this into one protocol that works across all compatible tools.

Example: Publishing Content with AI

Once connected, you can tell Claude: "Draft a blog post about our new service and publish it to the CMS." Claude will use the CMS MCP server to actually do it - writing in your brand voice and publishing directly.

Or: "Check our Instagram DMs and draft replies to anyone asking about pricing." The Meta MCP server handles this without copy-paste workflows.