MCP server

Your data, in any AI client.

Q: What is the Basedash MCP server?

The Basedash MCP server lets any compatible AI client — Claude Code, Cursor, ChatGPT, Windsurf, and others — connect to your Basedash workspace through the open Model Context Protocol. Once connected, the client can ask data questions and look up which sources are available, with answers powered by Basedash's AI data analyst.

Q: Which clients work with the Basedash MCP server?

Any MCP client that supports remote, streamable-HTTP servers can connect. Common clients include Claude Code, Cursor, ChatGPT, and Windsurf. Other MCP-compatible tools work the same way — point them at the Basedash MCP URL and authenticate.

Q: Do I need an API key?

No. Connecting opens an OAuth flow in your browser the first time you use a new client. You sign in with your Basedash account, and the client receives a scoped token automatically. There are no API keys to copy, store, or rotate.

Q: What can the AI client actually do once connected?

It gets two tools: ask a question about your data and list the data sources in your workspace. Behind the scenes, those calls run through the same AI data analyst that powers Basedash chat — so the client can run lookups, generate charts, and analyze trends without leaving its own interface.

Q: How does it respect data access controls?

Every tool call enforces your workspace's access controls. If your account doesn't have access to a data source in Basedash, the MCP server won't expose that source either. Permissions for chats, dashboards, and sources all carry over.

Q: Does this count toward my Basedash usage?

Yes. Questions asked through the MCP server use the same AI engine as Basedash chat and count toward your workspace's AI usage. Workspace admins can review usage and plan limits in Basedash's billing settings.

Plug Basedash into Claude Code, Cursor, ChatGPT, or any other MCP client — and put a real data analyst inside the chat your team already uses.

Start free Read the docs

14-day trial. No credit card required.

Chats / Trial-to-paid by signup source

Basedash MCP

Which signup sources convert trials to paid most efficiently this quarter? Use Basedash to compare weekly conversion rates across the top three channels.

basedash.ask_question

Done 2.4s

question: "Trial-to-paid conversion rate by signup source, last 12 weeks, top 3 sources"

data_source: auto

Trial-to-paid conversion by signup source

Returned by Basedash

24.7% Referral

18.4% Organic

9.2% Paid ads

Referral leads at 24.7%, organic search has climbed to 18.4%, and paid ads has slipped to 9.2% as new spend chases lower-intent traffic.

Reply to AI assistant…

Compatible clients

Drop into whatever you already chat with.

One URL, one OAuth flow — Basedash shows up as a tool the moment you connect.

Claude Code

Terminal

Add Basedash to Anthropic's coding agent in a single command.

claude mcp add basedash --transport http https://charts.basedash.com/api/public/mcp

Cursor

Settings → MCP

Drop the URL into Cursor's MCP settings as a streamable HTTP server.

https://charts.basedash.com/api/public/mcp

ChatGPT

Settings → Connections

Add Basedash from ChatGPT's Connections panel — no setup file required.

https://charts.basedash.com/api/public/mcp

Windsurf

MCP catalog

Connect from Windsurf's MCP catalog using the streamable HTTP endpoint.

https://charts.basedash.com/api/public/mcp

Any MCP client

Streamable HTTP

Anything that speaks remote MCP can connect — refer to your client's docs.

https://charts.basedash.com/api/public/mcp

Tools

Two tools, full coverage.

The Basedash MCP server exposes a small surface — but it goes deep into your data.

ask_question

Ask question

A back-and-forth with your data analyst.

Pose a question in plain English and get a real answer from Basedash's data analyst — quick lookups or deeper trends, comparisons, and strategic analysis.

Continues the same chat across follow-ups
Generates validated SQL behind the scenes
Returns numbers, charts, and reasoning

get_data_sources

Get data sources

What's available, before you ask.

Lists every connected database, warehouse, and SaaS source in your workspace so the client knows what data exists and where to point a question.

Direct databases like Postgres and MySQL
Warehouses like BigQuery and Snowflake
750+ SaaS apps via Fivetran or MCP connectors

Access

Same permissions, every client.

Whatever an account can see in Basedash is exactly what flows through MCP — nothing more.

Workspace permissions, enforced

Every tool call respects the same access controls as your Basedash workspace — nothing more.

OAuth-authenticated

First-time connection opens an OAuth flow in your browser. No API keys to share or rotate.

Source-level visibility

If you can't see a data source in Basedash, neither can the AI client connected to it.

Server URL

Add this as a remote MCP server in any compatible client.

https://charts.basedash.com/api/public/mcp

MCP server FAQ

What is the Basedash MCP server?

Which clients work with the Basedash MCP server?

Do I need an API key?

What can the AI client actually do once connected?

How does it respect data access controls?

Does this count toward my Basedash usage?

Get started in under 30 minutes

We can help you migrate your data and dashboards from any other tool.

Start free Book a demo