Metabase MCP Server

What it does

Metabase MCP lets compatible AI clients search analytics content, browse data, run queries and create supported content in a Metabase instance. Results can be returned as text or interactive charts when the client supports them.

An administrator enables the MCP server, then each user authenticates with their existing permissions. The AI client supplies the model, so a separate Metabot model connection is not required. Query results are sent to the client and may reach its model provider, making the client’s handling policies part of the deployment decision.

What it can do

• Analytics search
• Data browsing
• Query execution
• Content creation
• Interactive chart results

Before you choose

• Client support determines whether inline charts render.
• Returned data enters the client’s environment.
• MCP calls do not consume the Metabase AI-provider connection.

Listed under

• Coding
• MCP and agent tools

Get started

1. Have an administrator enable AI features and the MCP server.
2. Register the instance endpoint in the AI client.
3. Authenticate and review tool permissions before running queries or creating content.

The connector requires an account with the documented product access. No separate numerical connector rate is established here; the connected AI client has its own requirements and charges.

We couldn’t find published prices for Metabase MCP Server. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 30 Oct 2026. Each source supports part of the overview; availability and prices can change.

• MCP server | Metabase Documentation ↗www.metabase.com

Company website

metabase.com: product, pricing and policy pages

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 30 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.