# DataFinder MCP

> DataFinder MCP is an open-source MCP server for analytics and data. Chat analytics on Volcengine DataFinder: 52 tools, metric dictionary, local jobs. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted
- **Repository:** https://github.com/cyuanxv/datafinder-mcp
- **MCP Registry name:** io.github.cyuanxv/datafinder-mcp
- **Categories:** [analytics and data](https://ai-powered.com/categories/analytics-and-data)
- **Listing updated:** 2026-11-18

## About

DataFinder MCP is written mainly in Chinese and targets Volcengine's DataFinder product-analytics service. Product managers, growth teams and analysts can ask in plain language for active users by channel, next-day retention against last week, or a one-day conversion funnel, then follow up with a filter such as iOS only and export the result as CSV.

The 52 tools cover queries, analysis, a metric dictionary, task records and controlled configuration. The dictionary stores names, definitions, sources and versions so later questions reuse the same definition, and SQL metrics can be bound to explicit tables and columns. Results keep their source, date range and quality notes. The standalone package ships an independent SQLite database, persistent jobs and a separate background process, and needs Node.js 24 or later.

Writes are off by default. Local changes and upstream metadata changes are separate switches, and each follows a preview, confirm and read-back flow. Project-wide queries are also disabled until enabled. Arbitrary SQL execution is not offered. Setup needs an access key, secret key and project, app and subject IDs.

## Features

- **Natural-language queries:** Asks for reports, events, retention and funnels without writing SQL.
- **Metric dictionary:** Saves definitions, sources and versions for reuse.
- **Follow-up questions:** Reuses earlier conditions to drill into channel or platform.
- **Report export:** Exports results as CSV or Excel-readable files.
- **Persistent jobs:** Keeps task progress and history in a local SQLite database.
- **Confirmed changes:** Metadata edits require preview, confirmation and read-back.
- **Scoped access:** Project-wide queries are off unless explicitly enabled.

## Good for

- Check yesterday's active users by channel and system
- Compare next-day retention week on week
- Analyse a conversion funnel with a one-day window
- Record an agreed metric definition for later reports

## Install

### Claude Code (Terminal)

```sh
claude mcp add datafinder-mcp --env DATAFINDER_AK='<DATAFINDER_AK>' --env DATAFINDER_SK='<DATAFINDER_SK>' --env DATAFINDER_PROJECT_ID='<DATAFINDER_PROJECT_ID>' --env DATAFINDER_APP_ID='<DATAFINDER_APP_ID>' --env DATAFINDER_SUBJECT_ID='<DATAFINDER_SUBJECT_ID>' -- npx -y @cyuanxv/datafinder-mcp
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "datafinder-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@cyuanxv/datafinder-mcp"
      ],
      "env": {
        "DATAFINDER_AK": "<DATAFINDER_AK>",
        "DATAFINDER_SK": "<DATAFINDER_SK>",
        "DATAFINDER_PROJECT_ID": "<DATAFINDER_PROJECT_ID>",
        "DATAFINDER_APP_ID": "<DATAFINDER_APP_ID>",
        "DATAFINDER_SUBJECT_ID": "<DATAFINDER_SUBJECT_ID>"
      }
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "datafinder-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@cyuanxv/datafinder-mcp"
      ],
      "env": {
        "DATAFINDER_AK": "<DATAFINDER_AK>",
        "DATAFINDER_SK": "<DATAFINDER_SK>",
        "DATAFINDER_PROJECT_ID": "<DATAFINDER_PROJECT_ID>",
        "DATAFINDER_APP_ID": "<DATAFINDER_APP_ID>",
        "DATAFINDER_SUBJECT_ID": "<DATAFINDER_SUBJECT_ID>"
      }
    }
  }
}
```

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

```json
{
  "servers": {
    "datafinder-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@cyuanxv/datafinder-mcp"
      ],
      "env": {
        "DATAFINDER_AK": "<DATAFINDER_AK>",
        "DATAFINDER_SK": "<DATAFINDER_SK>",
        "DATAFINDER_PROJECT_ID": "<DATAFINDER_PROJECT_ID>",
        "DATAFINDER_APP_ID": "<DATAFINDER_APP_ID>",
        "DATAFINDER_SUBJECT_ID": "<DATAFINDER_SUBJECT_ID>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "datafinder-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@cyuanxv/datafinder-mcp"
      ],
      "env": {
        "DATAFINDER_AK": "<DATAFINDER_AK>",
        "DATAFINDER_SK": "<DATAFINDER_SK>",
        "DATAFINDER_PROJECT_ID": "<DATAFINDER_PROJECT_ID>",
        "DATAFINDER_APP_ID": "<DATAFINDER_APP_ID>",
        "DATAFINDER_SUBJECT_ID": "<DATAFINDER_SUBJECT_ID>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "datafinder-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@cyuanxv/datafinder-mcp"
      ],
      "env": {
        "DATAFINDER_AK": "<DATAFINDER_AK>",
        "DATAFINDER_SK": "<DATAFINDER_SK>",
        "DATAFINDER_PROJECT_ID": "<DATAFINDER_PROJECT_ID>",
        "DATAFINDER_APP_ID": "<DATAFINDER_APP_ID>",
        "DATAFINDER_SUBJECT_ID": "<DATAFINDER_SUBJECT_ID>"
      }
    }
  }
}
```

## FAQ

### What do I need to run it?

Node.js 24 or later and your own DataFinder access key, secret key, project ID, app ID and subject ID.

### Can it change my DataFinder setup?

Only if you enable upstream writes, and each change needs preview, confirmation and read-back.

### Can it run any SQL?

No. The README says unrestricted arbitrary SQL is not provided.

### Is it open source?

Yes, under the MIT licence.

### Is DataFinder MCP free?

Yes. DataFinder MCP is open source, so you can use and self-host it for free.

### How do I add DataFinder MCP to Claude Code?

Run this command in your terminal: claude mcp add datafinder-mcp --env DATAFINDER_AK='<DATAFINDER_AK>' --env DATAFINDER_SK='<DATAFINDER_SK>' --env DATAFINDER_PROJECT_ID='<DATAFINDER_PROJECT_ID>' --env DATAFINDER_APP_ID='<DATAFINDER_APP_ID>' --env DATAFINDER_SUBJECT_ID='<DATAFINDER_SUBJECT_ID>' -- npx -y @cyuanxv/datafinder-mcp

### What are the best alternatives to DataFinder MCP?

The closest alternatives to DataFinder MCP are Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, dashboard-icons, Tapetide - Indian Stock Market MCP and World Monitor.

## Where this comes from

- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [Google Ads + Meta Ads + SEO/GEO MCP](https://ai-powered.com/mcp/google-ads-meta-ads-seo-geo-mcp): Google Ads + Meta Ads + SEO/GEO MCP is a freemium MCP server for marketing and seo.
- [dashboard-icons](https://ai-powered.com/mcp/dashboard-icons): dashboard-icons is an open-source MCP server for analytics and data.
- [Tapetide - Indian Stock Market MCP](https://ai-powered.com/mcp/tapetide-indian-stock-market-mcp): Tapetide - Indian Stock Market MCP is a freemium MCP server for finance and payments.
- [World Monitor](https://ai-powered.com/mcp/world-monitor): World Monitor is a freemium MCP server for analytics and data.
- [DataArcus for Power BI](https://ai-powered.com/mcp/dataarcus-for-power-bi): DataArcus for Power BI is a free MCP server for analytics and data.
- [CompanyData](https://ai-powered.com/mcp/companydata): CompanyData is a paid MCP server for analytics and data.
- [Pypi Stats](https://ai-powered.com/mcp/pypi-stats): Pypi Stats is a freemium MCP server for analytics and data.

---
Source: https://ai-powered.com/mcp/datafinder-mcp
