# autocarver

This MCP server lets an AI assistant classify dataset columns and carve them against a target through local tool calls.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 11
- Author: mdefrance
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/mdefrance/AutoCarver
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/mcp/autocarver

## What it does

autocarver connects AutoCarver to an MCP-aware coding assistant. Give it a data file and target column; the server identifies feature types, proposes the carving, and returns a summary table. The workflow runs through tool calls, so you do not need to write the setup code by hand.

AutoCarver converts numeric, categorical, and ordinal columns into readable bins. It searches candidate groupings and checks them on held-out data; `cv` adds cross-validation views during fitting. The MCP process starts locally with `uvx autocarver`, and the dataset is not sent to AutoCarver or another external service.

The LLM affects the carving result. Have a person review the inferred feature definitions before production use.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add autocarver -- uvx autocarver
```
uvx runs the command named after the package; if the README names another one, use uvx --from autocarver <command>

### Codex

**Platform CLI**
```bash
codex mcp add autocarver -- uvx autocarver
```
uvx runs the command named after the package; if the README names another one, use uvx --from autocarver <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "autocarver": {
            "command": "uvx",
            "args": [
                "autocarver"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from autocarver <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "autocarver": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "autocarver"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from autocarver <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add autocarver uvx autocarver
```
uvx runs the command named after the package; if the README names another one, use uvx --from autocarver <command>

## When to use it
- classify columns in a dataset
- carve features against a target column
- return a summary of the resulting bins
- check groupings against held-out data

## Example prompts
- Qualify the columns in `titanic.csv` and carve them against `Survived`.
- Inspect `data.csv`, infer its feature types, and propose bins for `target`.
- Process the columns in `train.csv` against its target and show the summary table.

## Similar
- [World Monitor](https://affpapa.org/en/nevibecode/mcp/world-monitor.md) — MCP server · ★ 89408
- [Scrapling MCP Server](https://affpapa.org/en/nevibecode/mcp/scrapling-mcp-server.md) — MCP server · ★ 89433
- [MCP Toolbox for Databases](https://affpapa.org/en/nevibecode/mcp/googleapis-mcp-toolbox-for-databases.md) — MCP server · ★ 16717
- [PostHog MCP Server](https://affpapa.org/en/nevibecode/mcp/posthog-mcp-server.md) — MCP server · ★ 40409
- [dbx](https://affpapa.org/en/nevibecode/mcp/dbx.md) — MCP server · ★ 27669
- [Metabase](https://affpapa.org/en/nevibecode/mcp/metabase.md) — MCP server · ★ 49845

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md