Build a Metabase dashboard with Claude

This is an end-to-end example: connect a BI tool over MCP, turn a question into a dashboard from live numbers, and close the review loop on one link. Total time, start to shared link, is under fifteen minutes. The same shape works for any of the other examples — only the connection step changes.

Here's the finished dashboard, published to a canvas — click any tile or chart to leave a comment, exactly as your team would:

The three moving parts

1. The Metabase MCP server gives Claude read access to your Metabase — tables, metrics, saved questions, dashboards, collections — through a controlled set of tools, all scoped to your own Metabase permissions. You can't see anything through Claude that you couldn't already see in Metabase.
2. Claude asks the questions, runs the queries, and writes a single self-contained HTML dashboard. You iterate on it in the artifact panel until it's right.
3. Drafty turns that HTML into a stable link your team reviews. Comments pin to the exact element; Claude ships the fix to the same URL.

The generation step is fast now. The part this example is really about is the third one — getting the answer in front of people without losing their feedback to a screenshot circled in Preview.

Step 1 — Connect the Metabase MCP server

Metabase ships an official MCP server built into the platform, served at /api/metabase-mcp on your own instance. An admin enables it once under Admin → AI → MCP (toggle the MCP server on). It uses Streamable HTTP and authenticates over OAuth 2.0 — Metabase runs its own embedded OAuth server, so no key is pasted into a config file.

In Claude Code (swap in your Metabase URL):

Then run /mcp inside Claude Code and follow the OAuth prompt to authorize. The connection inherits your Metabase permissions — you only need read access to the collections and tables this dashboard touches.

In Claude Desktop: open Settings → Connectors → Add custom connector, paste https://your-metabase.example.com/api/metabase-mcp, and authorize over OAuth the same way.

Step 2 — Pull the numbers

Ask Claude in plain language. It uses the MCP server's read tools (search, read_resource, construct_query, execute_query) to find the right tables and saved questions, then run them:

Claude searches Metabase, reuses your existing metrics where it can, runs the queries, and returns the figures. Sanity-check them against the matching saved questions in Metabase before going further. This is the moment to catch a wrong assumption — a stale definition of "active," a timezone offset, a filter you didn't expect — while it's cheap.

Step 3 — Build the dashboard

Once the numbers look right, ask for the artifact:

Claude renders it live in the artifact panel. Iterate in place — you're not regenerating from scratch:

• "Add last period's WAU as a faint line behind the current one."
• "Show conversion rate as the hero and move WAU into a tile."
• "Round large numbers and add a sparkline to each tile."

Step 4 — Publish to Drafty for review

A Claude artifact link is a preview, not a stable URL — iterate the artifact and the link you already sent now shows the old version. Ask Claude to publish it to a Drafty canvas instead, so the link you share always stays current:

Claude pushes the dashboard and hands back a drafty.im/canvas/… link that renders on any device. Send it — your team opens it in a browser, no login and no Claude account needed.

Step 5 — The review loop

This is the part that's not obvious until you've done it once.

A reviewer clicks the specific tile, chart, or number they want changed and leaves a pinned comment — "this WAU count looks low, are we filtering out internal accounts?" The comment is anchored to that element, not floating in a Slack thread. Claude reads the comments through the CLI, reruns the relevant Metabase query if needed, and pushes a revised dashboard to the same URL. The reviewer refreshes and sees the change; the thread stays attached to the element.

The mechanic matters because of what it removes. A Slack message about a chart produces "the number on the left looks wrong." A pinned comment on the actual tile produces "this — exclude internal accounts from the active-user count." One of those produces a correct revision; the other produces a guess.

Keeping it fresh

An MCP-generated dashboard is a snapshot — it holds the numbers Claude pulled when it built it; it doesn't re-query Metabase when someone opens the link. For a weekly review or a board-ready snapshot, that's fine.

To make it a live canvas that always shows today's figures, copy this prompt — Claude sets up the refresh for you and schedules it to run on its own:

The link stays stable while the content updates underneath it — see keeping a canvas updated automatically.

What to watch for

• Read-only, always. A reporting dashboard needs read access and nothing more. Connect with an account whose Metabase role is scoped to read the collections you need.
• Check the figures before you share. The MCP returns exactly what you ask for — if your "active users" definition double-counts sessions or includes test accounts, the dashboard will confidently show the wrong number. Reconcile against the matching saved question in Metabase once.
• The link is the deliverable, not the artifact. Share the Drafty URL, not the Claude artifact preview — that's the one you can update in place.