The guide, shipped as code.
Words are easy to agree with and easy to ignore. These artifacts are the same standards in the formats your tools already read. Paste in, ship today.
Drop in, not opt in.
Every artifact is a file you can paste into a workspace. No vendor lock, no install step, no dependency.
Versioned with the guide.
When the visual design section changes the chart vocabulary, the Omni vis configs change too. Single source of truth.
AI-readable by default.
Descriptions, synonyms, and example queries are first-class. A chatbot dropped in tomorrow will already know what to say.
Three to start. More as the stack evolves.
Omni
Theme YAML, semantic-layer field templates, and AI-context snippets.
- Spicy theme YAML (8 hero hues, light + dark)
- Topic + view field-description templates
- AI context blocks for the chatbot
- Vega-Lite vis configs for the chart vocabulary
dbt
Schema-doc templates, grain macros, and test conventions.
- Model + column description templates
- Macros: grain check, freshness assertion, golden-row test
- Recommended `dbt_project.yml` defaults
- A starter `schema.yml` for a metric mart
Hex
Notebook templates, chart presets, and app-mode patterns.
- Spicy chart presets (matplotlib + Plotly)
- Notebook template: explore → narrative → app
- App-mode layout patterns
- SQL cell conventions for cohort + funnel analysis
A different tool you want covered?
Tableau, Looker, Sigma, Mode, Metabase, Streamlit, Lightdash. If your team lives in it, we will write artifacts for it.