# Exploratory Server

This MCP server for Claude Code reads notes, dashboards, charts and data published on Exploratory. It can search content and inspect datasets.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: exploratory-io
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/exploratory-io/exploratory
- Updated: 2026-11-15
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/exploratory-server

## What it does

Exploratory Server finds published material and reads notes, dashboards and slides as Markdown. For charts, it can return the underlying figures or an image; for analytics, it provides results such as coefficients and p-values.

The server can show a dataset’s schema and rows, then filter or aggregate its data. MCP tools are read-only, apart from exporting a result as CSV or Parquet; the download link stays valid for 15 minutes.

An Exploratory account is needed. On the first call, the client opens a browser for OAuth sign-in and access approval. No API key is required.

## Requirements
- Remote server: `https://exploratory.io/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http exploratory-server https://exploratory.io/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add exploratory-server --url https://exploratory.io/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "exploratory-server": {
            "url": "https://exploratory.io/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "exploratory-server": {
            "type": "http",
            "url": "https://exploratory.io/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http exploratory-server https://exploratory.io/mcp
```

## When to use it
- find a published note or dashboard
- read a note or slide
- inspect dataset columns and rows
- filter data and export a result

## Example prompts
- Find my published content on Exploratory.
- Show the columns in the “Sales” dataset and identify the regions with the highest sales.
- Filter a published dataset and export the result as CSV.

## Similar
- [exploratory-testing-advanced](https://affpapa.org/en/nevibecode/skills/exploratory-testing-advanced.md) — skill · 1 install
- [build-dashboard](https://affpapa.org/en/nevibecode/skills/build-dashboard.md) — skill · 8 installs
- [data-story-outliner](https://affpapa.org/en/nevibecode/skills/data-story-outliner.md) — skill · 1 install
- [data-driven-feature](https://affpapa.org/en/nevibecode/commands/seth-hobson-data-driven-feature.md) — command · ★ 126 repo stars
- [autario](https://affpapa.org/en/nevibecode/mcp/autario.md) — MCP server · 5089 installs
- [data-pipeline](https://affpapa.org/en/nevibecode/skills/data-pipeline.md) — skill · 5100 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
