# WarpGBM

WarpGBM trains gradient boosting models and runs inference directly from your assistant, exporting results to ONNX or joblib.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 678 installs
- Author: jefferythewind
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/jefferythewind/warpgbm-mcp
- Updated: 2025-10-17
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/warpgbm

## What it does

The server connects to your coding assistant over HTTP to handle gradient boosting workloads. It accepts tabular data in CSV or Parquet formats via `upload_data`, converting raw files into feature and target matrices ready for training. The `train` tool outputs portable `ONNX` or `joblib` artifacts using either the `warpgbm` or `lightgbm` backend.

Fresh training runs yield a temporary `artifact_id` that remains valid for 5 minutes, so immediate predictions with `predict_from_artifact` must run before that window closes. Other utilities include `get_agent_guide` for usage tips and `submit_feedback` for reporting issues.

## Requirements
- Remote server: `https://warpgbm-mcp--jefferythewind.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http warpgbm-mcp https://warpgbm-mcp--jefferythewind.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add warpgbm --url https://warpgbm-mcp--jefferythewind.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "warpgbm": {
            "url": "https://warpgbm-mcp--jefferythewind.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "warpgbm": {
            "type": "http",
            "url": "https://warpgbm-mcp--jefferythewind.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http warpgbm https://warpgbm-mcp--jefferythewind.run.tools
```

## When to use it
- train gradient boosting models on lightgbm or warpgbm engines
- parse CSV or Parquet files into training matrices
- export models into portable ONNX or joblib formats
- run quick predictions using temporary artifact identifiers

## Example prompts
- Parse data from `train.parquet` into feature arrays
- Train a gradient boosting model with the lightgbm engine and export to ONNX
- Generate inference using the new `artifact_id`

## Similar
- [xgboost-lightgbm](https://affpapa.org/en/nevibecode/skills/xgboost-lightgbm.md) — skill · 501 installs
- [lightgbm](https://affpapa.org/en/nevibecode/rules/lightgbm.md) — rule · ★ 3589 repo stars
- [signal-classification](https://affpapa.org/en/nevibecode/skills/signal-classification.md) — skill · 583 installs
- [nlp-engineer](https://affpapa.org/en/nevibecode/agents/nlp-engineer.md) — subagent · 927 installs
- [deep-learning](https://affpapa.org/en/nevibecode/skills/pluginagentmarketplace-deep-learning.md) — skill · 813 installs
- [vector-embed](https://affpapa.org/en/nevibecode/skills/vector-embed.md) — skill · 732 installs

---
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
