metabase/agent-skills/metabase-cli

metabase-cli

Load the workflow content from the CLI:

mb skills get core      # start here — auth, flag conventions, every command group
mb skills list          # enumerate specialized skills bundled with this CLI version
mb skills get <name>    # load a specialized skill (transform, data-workflow, dashboard, git-sync, data-app, …)

Command groups, in full: authenticate with named profiles; inspect databases (list, get, full metadata rollup, schemas, tables in a schema) and trigger a manual schema sync or field-values rescan; inspect tables and fields; list/get/create/update/archive cards (questions, models, metrics) and run them as JSON/CSV/XLSX; list/get/create/update dashboards and patch dashcards; list/get/create collections and traverse the hierarchy by id, entity_id, or "root"/"trash" (with items and recursive tree); list/get/create/update/archive native query snippets, segments, and measures; author/update/run transforms and schedule transform-jobs; publish tables to the Library; create data actions; scaffold data apps, generate their schema, questions and actions, and turn them off; read/update settings; search content (cards, dashboards, collections, transforms, metrics); git-sync to/from a git remote (status, dirty, import, export, branches, stash, add/remove a collection from sync).

Don't drive Metabase by curling /api/... directly — the CLI handles auth profiles, retries, schema validation, and credential redaction.

Doing a whole job, not one command? Higher-level data-engineering workflows — raw data to clean tables, the semantic layer, dashboards — live in the metabase-rde skill, which drives this CLI:

npx skills add metabase/agent-skills --skill metabase-rde

When it is installed, follow it; it loads the bundled skills it needs.