Context and operations for the smolsqls ALPHA deployment (https://alpha.smolsqls.com): the auth model, provisioning a tenant/database via the management API, and safety when touching live shared data. To actually RUN SQL against alpha, use the `query-db` skill with `--db alpha` — this skill covers everything around that. Triggers on: "query alpha", "run this on alpha", "check the live db", "hit alpha.smolsqls.com", "create a database on alpha", "provision an alpha db".
The shared Elixir tool for querying ANY smolsqls database over the HTTP query API (POST /v1/databases/:id/query). This is the low-level primitive the query-alpha-db and smolsqls-pm skills build on. Use it directly to run SQL against an arbitrary smolsqls database given its URL / id / auth_token, or when you need the query CLI's mechanics (positional args, --args-file, applying a .sql file, JSON output). Triggers on: "query a smolsqls db", "run SQL against a smolsqls database", "use the query tool".
Track this repo's own work — projects, plans, and tasks — in a live smolsqls database on alpha (dogfooding the product). Use whenever you plan or manage smolsqls work: creating/updating a project, writing or reading a plan (markdown design doc), adding/moving tasks, or answering "what's in progress / what's next / what's the plan for X". This is where plans live for the smolsqls repo (superseding local .plans/ markdown). Triggers on: "add a task", "what's in progress", "track this", "store the plan", "project tracker", "what's next", "mark done".