# Data Profiler MCP server

> Turns CSV, Parquet, JSON, and Excel files into compact structured profiles containing types, ranges, and data-quality summaries for AI agents.

- Repository: https://github.com/Ridadata/mcp-data-profiler
- Page: https://hostedmcp.org/servers/ridadata/mcp-data-profiler
- Transport: stdio
- Stars: 1
- License: MIT
- Latest version: 0.1.1

## Install

### Claude Code

```sh
claude mcp add mcp-data-profiler -- uvx mcp-data-profiler
```

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

```json
{
  "mcpServers": {
    "mcp-data-profiler": {
      "command": "uvx",
      "args": [
        "mcp-data-profiler"
      ]
    }
  }
}
```

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

```json
{
  "servers": {
    "mcp-data-profiler": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "mcp-data-profiler"
      ]
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"mcp-data-profiler","type":"stdio","command":"uvx","args":["mcp-data-profiler"]}'
```

### Claude Desktop (`claude_desktop_config.json`)

```json
{
  "mcpServers": {
    "mcp-data-profiler": {
      "command": "uvx",
      "args": [
        "mcp-data-profiler"
      ]
    }
  }
}
```

### Codex

```sh
codex mcp add mcp-data-profiler -- uvx mcp-data-profiler
```

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

```toml
[mcp_servers.mcp-data-profiler]
command = "uvx"
args = ["mcp-data-profiler"]
```

### Gemini CLI

```sh
gemini mcp add -s user mcp-data-profiler uvx -- mcp-data-profiler
```

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

```json
{
  "mcpServers": {
    "mcp-data-profiler": {
      "command": "uvx",
      "args": [
        "mcp-data-profiler"
      ]
    }
  }
}
```

## Tools (5)

- `path`
- `sample_rows`
- `max_columns`
- `top_k`
- `sheet`

## FAQ

### What file formats does the server support?

It supports CSV, TSV, TXT, Parquet, JSON, JSONL, NDJSON, and Excel formats, including Gzip-compressed variants.

### Can I restrict the server to a specific directory?

Yes, you can use the --root argument to confine profiling to a single directory for path safety.

### What kind of data quality issues does it detect?

It identifies columns that are entirely empty, constant values, probable identifiers, mixed types, and numbers or dates stored as text.
