make-me-a-dashboard
Made by JeffBuild and deploy one new dashboard page on an existing dashboard hub: a data-source interview, then read-only wiring per panel, either snapshot mode (a publish script rebuilds a Supabase cache table) or live-warehouse mode (fsnow on a read-only Snowflake role), composed from shadcn/ui. Use when someone says "make me a dashboard", "dashboard for <company or dataset>", "add a dashboard to the hub", "visualize this data on the hub", or asks for a KPI or metrics page.
From Jeffrey's Skills.md: professionally crafted skills for any AI agent, installed and kept in sync by the jsm CLI. Subscription only, $20/month.
Works With
Quality and Utility come from canonical skill metadata. Community ratings appear separately as stars when subscriber reviews exist.
Subscribe to download
Public previews are available, but downloads, CLI sync, and premium content require an active subscription.
Sign In to Subscribe- Full skill content with deterministic versioning
- CLI sync + local search index
- Access to Jeffrey’s premium library
Save to your account
Saved skills live in your cloud list. Run jsm sync on each machine to pull them locally.
What's New with this Skill
1Updates, enhancements, and release notes for this skill.
New skill: make-me-a-dashboard, adding one read-only dashboard page to an existing hub
An agent adds one dashboard to the hub that dashboard-hub-onboarding set up. It asks where the data lives, then wires each panel read-only: snapshot mode, where a publish script rebuilds Supabase cache tables from files or tool output, or live-warehouse mode, where a hub host with fsnow queries curated Snowflake views on a read-only role. Every panel shows how fresh its numbers are, and the page never writes back to a source.
- Checks for the hub first and sends the user to dashboard-hub-onboarding if there isn't one.
- The interview asks where the data lives (a CLI or data tool, Snowflake, local files, an API, or a mix), plus credentials, size, change rate and sensitivity for each source.
- Each panel is snapshot or live, and pages that mix the two are normal.
- Snapshot mode: one publish script rebuilds every dash_<slug> table from source in one transaction. A bad source file aborts the run and the old snapshot stays.
- Live mode: curated Snowflake views or dynamic tables through fsnow on a read-only role and a capped warehouse, cached for 15 to 60 minutes and checked against a monthly credit budget. It needs a hub host with fsnow.
- Pages use shadcn/ui stat cards, tables and the hub's existing chart component, with no second chart library.
- Every panel has loading, empty and error states and shows how fresh its numbers are.
- Display only: no mutations and no fix-it-here buttons. When a number disagrees with the source, the fix goes in the publish script or the query.
- Personal or regulated data stays in the system that owns it instead of being snapshotted into Supabase.
The skill checks for the hub first: a src/lib/registry.ts, a working deploy, and the dash_reader and dash_publisher roles. If there is no hub, it sends the user to dashboard-hub-onboarding. Then it asks where the data lives: a CLI or data tool with JSON or CSV output, a Snowflake warehouse, local CSV, JSON or SQLite files, an API or website, or a mix. For each source it also asks what credentials exist, how big the data is, how often it changes and how sensitive it is. DATA_INTERVIEW.md has the follow-ups for each answer, including the licensing questions.