広告掲載
MCP.so
ログイン
L

Lightdash MCP Server

@poddubnyoleg

リポジトリ

基本情報

カテゴリ

データと分析

トランスポート

stdio

作者

poddubnyoleg

投稿者

pd

Lightdash MCP Server について

MCP server for Lightdash analytics platform. Enables AI assistants like Claude to discover data, execute queries, create charts, and manage dashboards programmatically. Supports Lightdash Cloud and self-hosted instances.

#lightdash#analytics#mcp#claude#business-intelligence#data-visualization#dashboards#python

Lightdash MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "lightdash": {
      "command": "uvx",
      "args": [
        "lightdash-mcp"
      ],
      "env": {
        "LIGHTDASH_TOKEN": "<YOUR_TOKEN>",
        "LIGHTDASH_URL": "https://app.lightdash.cloud"
      }
    }
  }
}

Lightdash MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Lightdash Mcp?

Lightdash Mcp is a Model Context Protocol (MCP) server that connects AI assistants like Claude and Cursor to Lightdash analytics. It enables LLMs to discover data, create charts, and manage dashboards programmatically.

How to use Lightdash Mcp?

Install via pip install lightdash-mcp, uvx lightdash-mcp, or pipx run lightdash-mcp. Set environment variables LIGHTDASH_TOKEN and LIGHTDASH_URL, then run lightdash-mcp. Optionally configure Cloudflare Access or Google Cloud IAP. Add the server to MCP clients (e.g., Claude Desktop) by providing the command and environment variables.

Key features of Lightdash Mcp

  • Discovery: explore data catalogs, tables, and schemas
  • Querying: execute queries with filters, metrics, and aggregations
  • Chart management: create, read, update, delete charts
  • Dashboard management: build and manage dashboards with tiles and filters
  • Resource organization: create and manage spaces for content

Use cases of Lightdash Mcp

  • Allow an AI assistant to explore available tables and schemas in Lightdash
  • Automate chart creation and configuration from natural language prompts
  • Duplicate and modify existing dashboards programmatically
  • Run ad-hoc metric queries against any explore in a project
  • Organize analytics content by creating spaces and moving resources

よくある質問

1

How do I obtain a Lightdash token?

Go to **Settings** → **Personal Access Tokens** in your Lightdash instance, click **Generate new token**, and copy the token (starts with `ldt_`).

2

What are the prerequisites?

Python 3.10+, a Lightdash instance (Cloud or self-hosted), and a Lightdash Personal Access Token.

3

How do I handle authentication errors?

Verify your `LIGHTDASH_TOKEN` is correct and starts with `ldt_`, check the token hasn’t expired, and ensure you have the necessary permissions in Lightdash.

4

How do I set up Google Cloud IAP support?

Install with the `[iap]` extra (`pip install lightdash-mcp[iap]`), set `IAP_ENABLED=true`, and ensure the runtime service account has `roles/iam.serviceAccountTokenCreator` on itself and `roles/iap.httpsResourceAccessor` on the Cloud Run service.

5

How can I add a new tool to the server?

Create a new Python file in `lightdash_mcp/tools/` with a `TOOL_DEFINITION` and a `run` function. The server automatically discovers and registers the tool on restart.

コメント

「データと分析」の他のコンテンツ

T

Texas RRC Wellbore Intelligence

TEOS

Texas RRC oil & gas wellbores: API numbers, completion dates, depth, plug status

Statable

Key Arg B.V.

Privacy-first, cookieless web analytics hosted in the EU. Query visitors, pages, traffic sources, countries, goals, funnels and live traffic for your sites, and create sites, goals and funnels — 25 tools over OAuth.

Zillow MCP Server

HasData

HasData's hosted Zillow MCP server lets AI agents search homes for sale, rentals and sold listings, then retrieve property details, price history and Zillow estimates as structured JSON.

4

Subtext

Subtext by Fullstory

Session replay, built for agents. Subtext is agentic session review: it captures production sessions of your app and connects them to your coding agent — Claude Code, Cursor, Codex, Devin, your own harness — so it can

9

Tracetify — Competitor Ads, Growth Traces & Search Console

tracetify

17 tools for marketing and SEO research inside Claude Code, Cursor, Codex or ChatGPT: the ads any domain runs on Meta and Google, how a product grew from 12 public sources with every claim linked, your own Search Console

1
広告掲載