Dev48
Language
  • About
  • Services
  • Industries
  • Technologies
  • Articles
  • Contacts
Book a call
    Home/Articles/Updated claude skill for learning metabase
Dev48
info@dev48.org+375 29 320-40-96WhatsApp@isky_by

© 2026 · All rights reserved.

Updated Claude skill for learning Metabase

Source: Metabase | Open source Business Intelligence and Embedded Analytics•November 19, 2026 at 12:36 AM

Updated Claude skill for learning Metabase

Our updated Learn Metabase skill is a personalized guide to Metabase. It now tracks your progress directly in Metabase and can check your work, all while talking less.

All posts

Nov 18, 2026 in Product

We’ve updated our Claude skill for learning Metabase. It’s still your personalized guide to Metabase, but it now tracks your progress directly in Metabase, can check your work – and it also talks less.

The most important change is that the skill uses the improved MCP server we introduced in Metabase 64. This lets it pick meaningful examples from your own Metabase, and it can now also check what you’ve created. You can ask it for help in the context of your questions or dashboards, and it can even make them for you.

The new MCP server also allows Claude to create a Document in your personal collection to track your progress instead. When you ask Claude to teach you more about Metabase, it knows where you left off last time. This document is only visible to you and the admins of your instance.

Finally, we’ve made the skill more concise. Claude tends to be a little verbose, especially when explaining concepts. We’ve added instructions to keep explanations short. The result is fewer words that get you to the meaningful content faster.

Install the skill and try it out today!

You might also enjoy

Nov 5, 2026 in Product

From Toronto Hackathon to Stockholm meetup: Metabase data app world tour

From a Toronto hackathon to a Stockholm data meetup: how we built data apps for asteroids, F1 telemetry, and a pumpkin ERP on Metabase and ClickHouse.

Oct 19, 2026 in Product

The bottleneck moved: how to build a semantic layer for AI Analytics

AI removed the querying bottleneck and created a verification one. Here's the semantic layer that makes AI-generated answers traceable, and how to build it.

← All articles

What this article says

More in Data & Analytics

All →
Java4Shell Remote Code Execution (RCE) Vulnerability - CVE-2026-114938
Elastic

Java4Shell Remote Code Execution (RCE) Vulnerability - CVE-2026-114938

A bridge to modernization: What IDC’s Market Note says about Cameyo by Google
Looker

A bridge to modernization: What IDC’s Market Note says about Cameyo by Google

Google Cloud Highlights EMEA Startup Momentum at Slush 2026
Looker

Google Cloud Highlights EMEA Startup Momentum at Slush 2026

ObservabilityCON on the Road: New sessions coming to you in 2027
Grafana Labs

ObservabilityCON on the Road: New sessions coming to you in 2027

Java4Shell Remote Code Execution (RCE) Vulnerability - CVE-2026-114938
Elastic

Java4Shell Remote Code Execution (RCE) Vulnerability - CVE-2026-114938

A checklist for trusting AI-generated carbon footprints
Watershed

A checklist for trusting AI-generated carbon footprints

More from Metabase

From Toronto Hackathon to Stockholm meetup: Metabase Data App world tour
Metabase

From Toronto Hackathon to Stockholm meetup: Metabase Data App world tour

'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse
Metabase

'Wait, you don't have a semantic layer yet?' AI analytics readiness pulse

We don't use PHP
Metabase

We don't use PHP

As humans write less code: workloads shift from human to agent verification with iterative test suites
Metabase

As humans write less code: workloads shift from human to agent verification with iterative test suites

Making pivot tables faster with GROUPING SETS
Metabase

Making pivot tables faster with GROUPING SETS

How we made pivot tables three times faster
Metabase

How we made pivot tables three times faster