Fire in da houseTop Tip:Paying $100+ per month for Perplexity, MidJourney, Runway, ChatGPT and other tools is crazy - get all your AI tools in one site starting at $15 per month with Galaxy AI Fire in da houseCheck it out free

mcp-server

MCP.Pizza Chef: integration-app

The mcp-server is an implementation of the Model Context Protocol (MCP) server that exposes tools powered by Integration App. It enables developers to integrate and manage Actions defined in an Integration App workspace as callable tools within AI workflows. This server facilitates real-time interaction between LLMs and external tools, enhancing automation and context-aware task execution. Setup requires Node.js, configuration of Integration App actions, and authentication via tokens or workspace keys. The mcp-server is ideal for developers looking to extend AI capabilities with custom, managed tools in a secure and scalable manner.

Use This MCP server To

Expose Integration App actions as callable AI tools Enable AI-driven automation with custom workflows Manage and update tools dynamically via Integration App Authenticate and secure tool access with tokens Integrate external APIs into AI context via MCP Facilitate multi-step reasoning with real-time tool calls

README

Integration App MCP Server

Overview

This is an implementation of the Model Context Protocol (MCP) server that exposes tools powered by Integration App.

Managing Tools

This server uses Actions defined in an Integration App workspace as tools. To understand how this works and how to effectively manage tools for each application, please refer to the Using Tools guide.

Running the server

  1. Install node.js
  2. Configure some actions in your Integration App workspace
  3. Get Integration App token from your Workspace Settings page or generate using your Workspace Key and Secret (Authentication Guide).

You need to provide two environment variables to the server:

  • INTEGRATION_APP_TOKEN - token for accessing Integration App API
  • INTEGRATION_KEY - key of the integration you want to use tools for

This server exposes tools from one integration at a time. If you want to expose tools from multiple integrations, you can create multiple servers or modify the code to iterate over multiple integrations.

Here is an example of claude_desktop_config.json file with the server configured:

{
  "mcpServers": {
    "integration-app-hubspot": {
      "command": "npx",
      "args": ["-y", "@integration-app/mcp-server"],
      "env": {
         "INTEGRATION_APP_TOKEN": "<your-integration-app-token>",
         "INTEGRATION_KEY": "hubspot"
      }
    }
  }
}

Testing

To understand if everything works as expected, you can ask Claude what tools are available:

Claude Test

mcp-server FAQ

How do I configure tools for the mcp-server?
Tools are configured as Actions within your Integration App workspace. Refer to the 'Using Tools' guide for detailed instructions.
What environment variables are required to run the mcp-server?
You need to provide your Integration App token or workspace key and secret as environment variables for authentication.
Which runtime environment is needed for the mcp-server?
The server requires Node.js to be installed on your system.
How does the mcp-server authenticate requests?
It uses tokens generated from your Integration App workspace settings or via workspace key and secret for secure authentication.
Can I update tools without restarting the mcp-server?
Yes, since tools are managed in the Integration App workspace, updates there reflect dynamically without server restarts.
Is the mcp-server compatible with multiple LLM providers?
Yes, it is provider-agnostic and works with models like OpenAI, Anthropic Claude, and Google Gemini.
Where can I find documentation for managing tools?
The 'Using Tools' guide on the Integration App documentation site provides comprehensive instructions.
How do I get the Integration App token?
Tokens can be obtained from your workspace settings page or generated using your workspace key and secret.