# Analytics MCP server

> Generates citable, re-runnable statistical reports and data interpretations from datasets through an AI chat interface.

- Repository: https://github.com/embeddedlayers/mcp-analytics
- Page: https://hostedmcp.org/servers/embeddedlayers/mcp-analytics
- Transport: remote
- Stars: 9
- License: MIT
- Latest version: 1.0.8

## Install

### Claude Code

```sh
claude mcp add --transport http mcp-analytics https://api.mcpanalytics.ai/mcp/api-key
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "mcp-analytics": {
      "url": "https://api.mcpanalytics.ai/mcp/api-key"
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "mcp-analytics": {
      "type": "http",
      "url": "https://api.mcpanalytics.ai/mcp/api-key"
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"mcp-analytics","type":"http","url":"https://api.mcpanalytics.ai/mcp/api-key"}'
```

### Codex

```sh
codex mcp add mcp-analytics --url https://api.mcpanalytics.ai/mcp/api-key
```

### Codex config.toml (`~/.codex/config.toml`)

```toml
[mcp_servers.mcp-analytics]
url = "https://api.mcpanalytics.ai/mcp/api-key"
```

### Gemini CLI

```sh
gemini mcp add -s user --transport http mcp-analytics https://api.mcpanalytics.ai/mcp/api-key
```

### Gemini settings.json (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "mcp-analytics": {
      "url": "https://api.mcpanalytics.ai/mcp/api-key",
      "type": "http"
    }
  }
}
```

## FAQ

### Can I try the server before signing up?

Yes, you can connect in discovery mode without a key to list the available tools, though you cannot execute them.

### What kind of output does the server provide?

The server provides reports containing charts, tables, written interpretations, the named statistical method used, and the code that executed the analysis.

### Is my data kept private?

Uploaded data is used exclusively to answer your specific questions.
