How to Use Model Context Protocol (MCP)

What is MCP?

MCP (Model Context Protocol) is a standard that lets AI assistants connect to external tools and services. With it, an assistant can do more than chat—read and write files, drive apps, query data—depending on which MCP servers you attach.

Why it matters

Most AI assistants are limited to the conversation itself. If you want them to:

• Edit documents
• Drive apps like Photoshop
• Fetch weather or query a database
• Control other software on your machine

you need a bridge. MCP is that bridge: a common interface for exposing tools to AI, with operations typically running on your device, and the freedom to swap servers as needed.

How MCP works

Take Mind Elixir as an example. You want the AI to add a branch to a mind map, but Claude or ChatGPT has no built-in idea of Mind Elixir’s UI. The flow looks like this:

1. You say: “Add a branch to the mind map”
2. The AI understands the intent and sends a standardized request through an MCP client
3. Mind Elixir’s MCP server receives it and adds the branch
4. The result goes back over MCP, and the new branch appears in the map

Three pieces:

• AI assistant — parses natural language and decides which tool to call
• MCP protocol — defines how the two sides talk
• MCP-enabled tools (e.g. Mind Elixir) — perform the actual work

Clients that support MCP

Claude Desktop

• Desktop app for Claude
• One of the first clients with MCP support; docs and ecosystem are relatively mature

Claude Code

• CLI-oriented, aimed at developers
• Flexible config; good for scripting and automation

Cline

An editor extension for developers:

• Works with VS Code and other common editors
• Supports multiple AI providers
• Can attach custom tools via MCP

More clients: https://modelcontextprotocol.io/clients

Using MCP with Mind Elixir

Mind Elixir App ships with an MCP server. Once connected, you can inspect and edit mind maps in natural language.

What you can do

• Let the AI read the structure of the current map
• Add, edit, or remove nodes with plain language (e.g. “add a branch”)
• Ask the AI to reorganize hierarchy and logic
• Get suggestions based on existing content

How to connect

Step 1: Open Mind Elixir App

• Launch Mind Elixir App
• Create or open the mind map you want to work on

Step 2: Add the MCP server in your AI client

Standard setup:

1. Open your MCP client
2. Go to Settings → MCP Servers
3. Add a server:
   • Name: Mind Elixir
   • Transport: SSE
   • Address: http://localhost:6595/sse

Claude Code:

claude mcp add --transport http mind-elixir http://127.0.0.1:6595/mcp

Codex:

codex mcp add --url http://127.0.0.1:6595/mcp mind-elixir

Gemini:

{
  "mcpServers": {
    "mind-elixir": {
      "httpUrl": "http://localhost:6595/mcp"
    }
  }
}

Notes:

• Mind Elixir App listens on fixed port 6595
• SSE (Server-Sent Events) is used for real-time communication
• A port conflict will prevent the connection

Step 3: Start using it

After the connection succeeds, keep the target mind map open in the app, then talk to the AI, for example:

• “Help me add three branches: Ideas, Plans, Actions”
• “How can this mind map be improved?”
• “Help me reorder this content”
• “Give me some new ideas on this topic”

Tips

• Keep Mind Elixir App running while you use MCP
• Specific instructions work better than vague ones
• Break large edits into smaller steps
• Save after important changes

Connection failed?

1. Check that port 6595 is free
2. Confirm Mind Elixir App is running
3. Make sure the firewall is not blocking local connections
4. Restart Mind Elixir App and try again