Quickchat AI

The Quickchat AI MCP Server lets you expose your Quickchat AI Agent to popular AI apps so users can interact with your AI through their preferred interfaces. It provides a simple, secure way to connect via MCP and manage how your AI is invoked, described, and controlled from client apps.

How to use

Set up your MCP in each client you want to support. You provide a name and a description for your MCP so AI apps know when to contact your AI and what capabilities it offers. You can also specify an optional command to start the local MCP runtime.

How to install

Prerequisites: you need the UV tooling to run MCP clients. Install UV with the following command.

curl -LsSf https://astral.sh/uv/install.sh | sh

Configuration and usage tips

For Claude Desktop and Cursor, you will provide a configuration snippet that tells the client how to reach your Quickchat AI MCP. This snippet includes the MCP name, a command to run the MCP, and environment variables to provide your scenario and API key.

{
  "mcpServers": {
    "< QUICKCHAT AI MCP NAME >": {
      "command": "uvx",
      "args": ["quickchat-ai-mcp"],
      "env": {
        "SCENARIO_ID": "< QUICKCHAT AI SCENARIO ID >",
        "API_KEY": "< QUICKCHAT AI API KEY >"
      }
    }
  }
}

Security and sharing tips

Do not publish your Quickchat API key to end users. When you are ready to share your MCP with others, disable the API key requirement and provide a version of the configuration snippet that omits the API key.

{
  "mcpServers": {
    "< QUICKCHAT AI MCP NAME >": {
      "command": "uvx",
      "args": ["quickchat-ai-mcp"],
      "env": {
        "SCENARIO_ID": "< QUICKCHAT AI SCENARIO ID >"
      }
    }
  }
}

Available tools

configureMCP

Configure an MCP for an AI app by defining its name, description, and startup command so clients know how to reach your AI.

testMCPConnection

Test connection between an AI app and your Quickchat MCP to verify setup and environment variables are passed correctly.