---
title: "Other MCP clients"
description: "Any client that speaks the Model Context Protocol over stdio can use the Loggerhead server."
source: "https://getloggerhead.com/docs/connect-your-ai-tool/other-mcp-clients/"
---

# Other MCP clients

> Any client that speaks the Model Context Protocol over stdio can use the Loggerhead server.

Loggerhead ships an MCP server. Its tools let your coding tool read your logs, metrics, traces, and lint findings, mark a finding fixed, and change the lint rules with a reason that you can undo in the app. The server talks to the Loggerhead engine on your Mac over `127.0.0.1`. Nothing leaves the machine.

## Add it with one click

1. Open Loggerhead and click **Settings** in the sidebar.
2. Open the **AI** tab.
3. In the **AI coding tools** card, click **Add** next to **Claude Desktop**.

Loggerhead writes the server into `~/Library/Application Support/Claude/claude_desktop_config.json` and keeps every other entry in that file. It leaves a copy of the previous file beside it with the extension `.bak`. Restart Claude Desktop, and the Loggerhead tools appear.

The server finds the engine by itself. You do not set a port, and the entry survives a restart.

## Add it by hand

The **AI coding tools** card covers Claude Desktop too. For any other client, use the shape below.

## The general shape

Every MCP client needs the same two facts: the command to start, and the engine address.

| Field | Value |
| --- | --- |
| Command | `/Applications/Loggerhead.app/Contents/Resources/loggerhead-mcp` |
| Arguments | none |
| Transport | stdio |

## Claude Desktop

Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "loggerhead": {
      "command": "/Applications/Loggerhead.app/Contents/Resources/loggerhead-mcp"
    }
  }
}
```

## Gemini CLI

Add to `~/.gemini/settings.json` under `mcpServers` with the same command and arguments. Restart the CLI.

## VS Code

Add the server to the workspace `.vscode/mcp.json` with type `stdio`, the same command, and the same arguments.

To send the project's own logs, metrics, and traces to Loggerhead, paste the prompt from [Connect your project with one prompt](/docs/connect-your-ai-tool/connect-your-project-with-one-prompt/).

See [What the server exposes](/docs/mcp-server/what-the-server-exposes/) for the tools and the one resource.