GitHub - andreagrandi/mb-cli: CLI client for MetaBase

A read-only CLI for the Metabase API. Designed for terminal users and AI coding agents (Claude, Codex) to explore databases, inspect schemas, and run ad-hoc queries.

Installation

Homebrew

brew tap andreagrandi/tap
brew install mb-cli

Binary download

Download the latest release from the releases page.

Go install

go install github.com/andreagrandi/mb-cli/cmd/mb-cli@latest

Configuration

Set MB_HOST plus one of the two authentication variables.

Option 1: API key (recommended)

export MB_HOST=https://your-metabase-instance.com
export MB_API_KEY=your-api-key

MB_API_KEY is a long-lived Metabase API key. Generating one requires Metabase admin access.

Option 2: Session token

export MB_HOST=https://your-metabase-instance.com
export MB_SESSION_TOKEN=your-session-token

Use this when you do not have admin access to mint an API key. Grab the value of the metabase.SESSION cookie from your browser (DevTools → Application → Cookies). Session tokens expire on logout or when the Metabase server's session timeout fires, so they are best for short-lived use.

MB_API_KEY and MB_SESSION_TOKEN are mutually exclusive — set one or the other, not both.

Optional

export MB_REDACT_PII=false  # Disable PII redaction (enabled by default)

Usage

Global flags

Database commands

# List all databases
mb-cli database list

# Get database details
mb-cli database get 1

# Full metadata (tables + fields)
mb-cli database metadata 1

# List all fields in a database
mb-cli database fields 1

# List schema names
mb-cli database schemas 1

# List tables in a specific schema
mb-cli database schema 1 public

The database command has an alias db:

Table commands

# List all tables
mb-cli table list

# Get table details
mb-cli table get 42

# Table metadata with field details
mb-cli table metadata 42

# Foreign key relationships
mb-cli table fks 42

# Raw table data
mb-cli table data 42

Field commands

# Get field details
mb-cli field get 100

# Summary statistics
mb-cli field summary 100

# Distinct values
mb-cli field values 100

SQL queries

# Run a SQL query by database ID
mb-cli query sql --db 1 --sql "SELECT * FROM users LIMIT 10"

# Run a SQL query by database name (case-insensitive substring match)
mb-cli query sql --db prod --sql "SELECT count(*) FROM orders"

# Append a LIMIT clause
mb-cli query sql --db 1 --sql "SELECT * FROM users" --limit 5

# Export results as CSV
mb-cli query sql --db 1 --sql "SELECT * FROM users" --export csv

The --db flag accepts a numeric database ID or a name substring. If the name matches multiple databases, you'll be asked to use the ID instead.

Saved questions (cards)

# List all saved questions
mb-cli card list

# Get card details
mb-cli card get 10

# Get the full card definition, including dataset_query and template tags
mb-cli card get 10 --full

# Execute a saved question
mb-cli card run 10

# Execute a saved question with parameters
mb-cli card run 10 --param timeframe_days=14

Dashboards

# List all dashboards
mb-cli dashboard list

# Get dashboard metadata, tabs, filters, and grouped cards
mb-cli dashboard get 298

# List the saved questions used by a dashboard
mb-cli dashboard cards 298

# Discover valid values for a dashboard filter
mb-cli dashboard params values 298 merchant_name

# Search dashboard filter values
mb-cli dashboard params search 298 merchant_name acme

# Execute a dashboard card with dashboard parameters applied
mb-cli dashboard run-card 298 1201 398 --param merchant_name="Acme Corp"

# Summarize tabs, parameter mappings, and source-card dependencies
mb-cli dashboard analyze 298

Search

# Search across all Metabase items
mb-cli search "users"

# Filter by type
mb-cli search "revenue" --models table,card

Version

Recipes

docs/RECIPES.md collects end-to-end recipes for common Metabase exploration tasks — schema discovery, saved-question inspection, dashboard analysis, parameterized card and dashboard runs, and PII-safe querying. Each recipe is a worked sequence of commands with sample output.

Agent integration

mb-cli is designed to be used by AI coding agents. When piped or used in scripts, output defaults to JSON for easy parsing.

Typical agent workflow:

# 1. Discover databases
mb-cli database list

# 2. Find relevant tables
mb-cli search "users" --models table

# 3. Inspect table schema
mb-cli table metadata 42

# 4. Query data
mb-cli query sql --db 1 --sql "SELECT id, email FROM users WHERE created_at > '2024-01-01' LIMIT 10"

# 5. Inspect a dashboard that depends on saved questions
mb-cli dashboard analyze 298

Dashboard analysis workflow

Example flow for dashboard 298:

# 1. Inspect dashboard structure
mb-cli dashboard get 298

# 2. List the saved questions behind the dashboard
mb-cli dashboard cards 298

# 3. Inspect a card's full SQL or MBQL definition
mb-cli card get 398 --full

# 4. Discover valid parameter values
mb-cli dashboard params values 298 merchant_name

# 5. Summarize dependencies and assumption-backed cards
mb-cli dashboard analyze 298

API coverage

The dashboard and parameter workflows use these Metabase endpoints:

PII Redaction

When AI agents use mb-cli directly (via shell commands), query results containing PII (emails, names, phone numbers) flow through stdout into the model's context. mb-cli prevents that by replacing PII column values with [REDACTED] before data leaves the client layer. Agents can still cross-reference records using IDs; for actual PII values, the user can check directly in Metabase.

Redaction is ON by default. Disabling it is a visible, auditable action: when the flag or env var is set to false, mb-cli prints Warning: PII redaction is disabled to stderr on every invocation.

What gets redacted

Columns whose Metabase field semantic type is one of:

type/Email, type/Name, type/Phone, type/Address, type/City, type/State, type/ZipCode, type/Country, type/Latitude, type/Longitude, type/Birthdate, type/AvatarURL, type/URL, type/ImageURL, type/Company.

For native SQL results, Metabase does not always populate the semantic type on result columns. mb-cli fills it in by fetching field metadata from the source database and matching by column name. When two fields share a name and disagree, mb-cli picks the PII type (err on the side of caution).

Where redaction applies

Schema and metadata commands (database metadata, table metadata, field get, etc.) return field definitions, not data, so there is nothing to redact.

Limits and known gaps

• Redaction relies on Metabase semantic types. A column holding PII without a semantic type (or with a custom type not in the list above) will not be redacted. Set semantic types in Metabase to close that gap.
• Derived columns from joins, aggregations, or CASE expressions may lose the upstream semantic type. Native SQL enrichment matches by column name only — alias derived columns to match a known PII field name if you want them caught.
• Free-form type/Description fields are not treated as PII; if you store names or emails in description-style columns, set their semantic type explicitly.
• Redaction operates on result rows, not column headers. Column names like email remain visible — values are replaced.

Export behavior

When redaction is enabled, the --export flag on query sql and query filter is blocked with this error:

export is not supported when PII redaction is enabled (use JSON or table format instead)

Raw export bytes (CSV/XLSX/JSON straight from Metabase) cannot be post-processed reliably for redaction, so the safe default is to refuse the operation. To export, explicitly opt out of redaction for that invocation:

MB_REDACT_PII=false mb-cli query sql --db 1 --sql "..." --export csv
# or
mb-cli query sql --redact-pii=false --db 1 --sql "..." --export csv

Opt-out precedence

The --redact-pii flag wins if set; otherwise MB_REDACT_PII is consulted; otherwise the default is true. Any value other than the literal string false is treated as enabled.

License

MIT