---
title: Data360-MCP Visualization Engine Explorer
emoji: 📊
colorFrom: blue
colorTo: indigo
sdk: docker
app_port: 7860
pinned: false
---

# Data360-MCP Visualization Engine Explorer

This repository contains the deployment configuration for hosting the **Data360-MCP Visualization Engine Explorer** in Hugging Face Spaces.

## 🚀 Deployment Instructions

### 1. Create a Hugging Face Space
- Go to [Hugging Face Spaces](https://huggingface.co/spaces) and click **Create new Space**.
- Name the Space (e.g. `data360-mcp-explorer`).
- Select **Docker** as the SDK.
- Choose **Blank** template.

### 2. Configure Repository Secrets & Environment Variables
In your Space settings page, add the following under **Repository secrets**:
- `OPENAI_API_KEY`: Required for natural query question framing and running vision-based chart audits.
- `HF_TOKEN`: Your Hugging Face write access token (required to pull/push persistent data).

Under **Variables**:
- `HF_DATASET_ID`: Set to `rafmacalaba/data360-explorer-reports` (defaults to this if not set).

### 3. Sync Repository Files
Add the Hugging Face Space remote and push the branch:
```bash
# Add the remote and push
git remote add space https://huggingface.co/spaces/rafmacalaba/data360-mcp-explorer
git push space main
```

The Space will automatically build the container using the provided `Dockerfile` and boot the server on port `7860`.