GA4 MCP server for AI agents: schema discovery, server-side aggregation, analysis-ready defaults.
Metabase is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
What it does
Query and manage Metabase dashboards, cards, and data via Claude
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --env 'METABASE_URL=${METABASE_URL}' --env 'METABASE_API_KEY=${METABASE_API_KEY}' --transport stdio 1luvc0d3-metabase-mcp '--' npx -y @ai-1luvc0d3/metabase-mcp
# 2 — ask your agent for something these tools do
# batch_execute, ask_data, generate_insights and 27 more
It needs METABASE_URL, METABASE_API_KEY: this reads them from your environment, so export them first.
Metabase costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Metabase as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. Claude Code reads METABASE_URL (Your Metabase instance URL), METABASE_API_KEY (Metabase API key (Admin > Settings > API Keys)) from your environment each time it starts the server, so the value stays out of its config: export them in your shell profile (macOS, Linux), or run setx METABASE_URL …, setx METABASE_API_KEY … and open a new terminal (Windows). The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user --env 'METABASE_URL=${METABASE_URL}' --env 'METABASE_API_KEY=${METABASE_API_KEY}' --transport stdio 1luvc0d3-metabase-mcp '--' npx -y @ai-1luvc0d3/metabase-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
It reaches a system of yours, so it needs your own credential rather than the publisher’s: METABASE_URL — Your Metabase instance URL; METABASE_API_KEY — Metabase API key (Admin > Settings > API Keys). You set them in your own client’s config, on your machine — this marketplace never holds it.
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
30 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Execute up to 20 operations in parallel in a single call
| Parameter | Type | Required | Notes |
|---|---|---|---|
| operations | array | yes | Array of operations to execute in parallel |
Answer questions about your data using natural language
| Parameter | Type | Required | Notes |
|---|---|---|---|
| question | string | yes | Natural language question about your data |
| database_id | number | yes | Database ID to query |
| tables | array | no | Specific tables to consider |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
Auto-generate insights from an existing card/question
| Parameter | Type | Required | Notes |
|---|---|---|---|
| card_id | number | yes | Card ID to analyze |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
Compare two metrics or time periods
| Parameter | Type | Required | Notes |
|---|---|---|---|
| question | string | yes | Comparison question (e.g., "Compare sales in Q1 vs Q2") |
| database_id | number | yes | Database ID |
| tables | array | no | Specific tables to use |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
Identify trends in time-series data
| Parameter | Type | Required | Notes |
|---|---|---|---|
| question | string | yes | Question about trends (e.g., "What are the sales trends over the past year?") |
| database_id | number | yes | Database ID |
| tables | array | no | Specific tables to use |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
Convert a natural language question to SQL query
| Parameter | Type | Required | Notes |
|---|---|---|---|
| question | string | yes | Natural language question about the data |
| database_id | number | yes | Database ID to query |
| tables | array | no | Specific tables to consider (for large schemas) |
Explain a SQL query in plain English
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sql | string | yes | SQL query to explain |
Get suggestions to optimize a SQL query
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sql | string | yes | SQL query to optimize |
| database_id | number | no | Database ID (to run EXPLAIN if available) |
Check SQL query for syntax issues and security concerns
| Parameter | Type | Required | Notes |
|---|---|---|---|
| sql | string | yes | SQL query to validate |
List all dashboards in Metabase
No parameter schema on file.
Get dashboard details including its cards
| Parameter | Type | Required | Notes |
|---|---|---|---|
| dashboard_id | number | yes | Dashboard ID |
List questions/cards in Metabase. Returns up to 100 cards. Use search_content for discovery.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| collection_id | number | no | Filter by collection ID |
| limit | number | no | Max cards to return (default 100) |
Get details of a specific question/card
| Parameter | Type | Required | Notes |
|---|---|---|---|
| card_id | number | yes | Card ID |
Execute an existing question/card and get results
| Parameter | Type | Required | Notes |
|---|---|---|---|
| card_id | number | yes | Card ID to execute |
| parameters | object | no | Values for the card's SQL variables, keyed by variable name, e.g. { "state": "CA", "min_total": 100 } |
| fields | array | no | Column names to include in results (default: all) |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
| limit | number | no | Max rows to return (default: server maxRows setting) |
| offset | number | no | Row offset for pagination |
List all connected databases
No parameter schema on file.
Get tables and columns for a database
| Parameter | Type | Required | Notes |
|---|---|---|---|
| database_id | number | yes | Database ID |
| detail | string | no | Level of detail. "tables_only" returns table names without columns |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
| tables | array | no | Filter to specific table names |
Execute a SQL query (SELECT statements only)
| Parameter | Type | Required | Notes |
|---|---|---|---|
| database_id | number | yes | Database ID to query |
| sql | string | yes | SQL query (SELECT statements only) |
| fields | array | no | Column names to include in results (default: all) |
| format | string | no | Response format (default: compact). Use "default" for pretty-printed output |
| limit | number | no | Max rows to return (default: server maxRows setting) |
| offset | number | no | Row offset for pagination |
Search across Metabase content (dashboards, cards, collections)
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | yes | Search query |
| type | string | no | Filter by content type |
List all collections
No parameter schema on file.
Run a multi-step pipeline with output references between steps
| Parameter | Type | Required | Notes |
|---|---|---|---|
| steps | array | yes | Ordered pipeline steps. Each step can reference results from previous steps. |
Create a new question/card in Metabase
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Card name |
| database_id | number | yes | Database ID |
| sql | string | yes | SQL query for the card |
| collection_id | number | no | Collection to save the card to |
| description | string | no | Card description |
| display | string | no | Visualization type |
Update an existing question/card
| Parameter | Type | Required | Notes |
|---|---|---|---|
| card_id | number | yes | Card ID to update |
| name | string | no | New card name |
| description | string | no | New description |
| sql | string | no | New SQL query |
| collection_id | number | no | Move to collection |
| display | string | no | Change visualization type |
Permanently delete a question/card. This cannot be undone: the card does not go to the Metabase trash
| Parameter | Type | Required | Notes |
|---|---|---|---|
| card_id | number | yes | Card ID to delete |
Create a new dashboard
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Dashboard name |
| description | string | no | Dashboard description |
| collection_id | number | no | Collection to save to |
Update an existing dashboard
| Parameter | Type | Required | Notes |
|---|---|---|---|
| dashboard_id | number | yes | Dashboard ID to update |
| name | string | no | New dashboard name |
| description | string | no | New description |
| collection_id | number | no | Move to collection |
Permanently delete a dashboard. This cannot be undone: the dashboard does not go to the Metabase trash
| Parameter | Type | Required | Notes |
|---|---|---|---|
| dashboard_id | number | yes | Dashboard ID to delete |
Add a card/question to a dashboard
| Parameter | Type | Required | Notes |
|---|---|---|---|
| dashboard_id | number | yes | Dashboard ID |
| card_id | number | yes | Card ID to add |
| row | number | no | Row position |
| col | number | no | Column position |
| size_x | number | no | Width (in grid units) |
| size_y | number | no | Height (in grid units) |
Remove a card from a dashboard
| Parameter | Type | Required | Notes |
|---|---|---|---|
| dashboard_id | number | yes | Dashboard ID |
| dashcard_id | number | yes | Dashboard card ID (not the card ID) |
Create a new collection
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Collection name |
| description | string | no | Collection description |
| color | string | no | Collection color (hex code) |
| parent_id | number | no | Parent collection ID |
Move a card or dashboard to a different collection
| Parameter | Type | Required | Notes |
|---|---|---|---|
| item_type | string | yes | Type of item to move |
| item_id | number | yes | ID of the item to move |
| collection_id | number | yes | Target collection ID |
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Metabase or to call it, at any volume.
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tools were statically extracted from the published source (30 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
4 of the 30 recovered tools are registered only when !ctx.llmService is false (dist/tools/insight-tools.js:9) — typically when a key or option is configured. They are part of the server's surface and were analysed; a server started without that configuration exposes only the other 22.
4 of the 30 recovered tools are registered only when !ctx.llmService is false (dist/tools/nlq-tools.js:9) — typically when a key or option is configured. They are part of the server's surface and were analysed; a server started without that configuration exposes only the other 22.
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
22 tools are registered unconditionally · (1) 4 more only when !ctx.llmService is false (dist/tools/insight-tools.js:9) · (2) 4 more only when !ctx.llmService is false (dist/tools/nlq-tools.js:9) — typically when a key or option is configured. All 30 are in the table below and were analysed; the conditional ones are marked.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| batch_execute | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| ask_data conditional (1) | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| generate_insights conditional (1) | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| compare_metrics conditional (1) | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| trend_analysis conditional (1) | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| nlq_to_sql conditional (2) | sensitive-sourceuntrusted-input | sensitive-source: keyword "sql_query"; untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| explain_sql conditional (2) | sensitive-sourceuntrusted-input | sensitive-source: keyword "sql_query", parameter "sql"; untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| optimize_sql conditional (2) | sensitive-sourceuntrusted-input | sensitive-source: keyword "sql_query", parameter "sql"; untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| validate_sql conditional (2) | sensitive-source | sensitive-source: keyword "sql_query", parameter "sql" |
| list_dashboards | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_dashboard | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| list_cards | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_card | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| execute_card | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| list_databases | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_database_schema | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| execute_query | sensitive-sourceuntrusted-input | sensitive-source: keyword "sql_query", parameter "sql"; untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| search_content | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_collections | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| run_workflow | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| create_card | sensitive-sourceuntrusted-input | sensitive-source: parameter "sql"; untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| update_card | sensitive-sourceexternal-sink | sensitive-source: parameter "sql"; external-sink: declares destructiveHint: true, openWorldHint: true |
| delete_card | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| create_dashboard | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| update_dashboard | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| delete_dashboard | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| add_card_to_dashboard | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| remove_card_from_dashboard | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
| create_collection | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| move_to_collection | external-sink | external-sink: declares destructiveHint: true, openWorldHint: true |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
Supply chain and provenance
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 1.8.0. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Any signed-in account that does not publish Metabase can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.