Examples

Each example is a small dbt project with seeds, models, ggsql visualisations, and dashboard YAML.

Gallery

• simple_dbt: minimal revenue dashboard and chart syntax sampler.
• sales_dashboard: sales performance by month, channel, and region, two maps (revenue by country and store locations) and a treemap of store revenue.
• product_analytics: product usage and activation by plan, with a histogram and a boxplot over per-account rows and a weekday-by-hour heatmap.
• finance_metrics: bookings, expenses, margin and collections over twelve months. Uses all eight chart types.
• clanker_insights: what one customer's AI agents did on Clanker, a made-up agent platform: runs, success, spend by model, run time, failures. Built with export.embed for the glyf-js customer-facing demo app.

Run an example

From an example directory:

uv run dbt seed --profiles-dir . --full-refresh --no-partial-parse
uv run dbt build --profiles-dir .
uv run glyf render
uv run glyf dashboard
uv run glyf export --clean --zip

Each example keeps its local DuckDB file under that example's target/ directory.

Open target/glyf/site/index.html.

Seed data

The seeds of finance_metrics, product_analytics and clanker_insights are synthetic and are written by seed_data.py. The generator is seeded, so it rewrites the same files byte for byte:

uv run python examples/seed_data.py

Change the shapes at the top of that file, run it, and commit the CSVs to change what the examples show. The metric tiles in the dashboards quote figures from the data, so check them after regenerating.