Skip to content
Back to skills

Layerchart

ASecurity

Expert guide for LayerChart, a Svelte component library for building diverse data visualizations (Cartesian, radial, hierarchical, geo, graph) with unopinionated building blocks, motion primitives, and advanced interactions.

  • 666 stars
  • 0 votes
  • 0 copies
  • 5 views
  • Added February 9, 2026
developmentgoreactapiperformance

Works with

  • cli
  • api

Security analysis

A100/100

Scanned February 12, 2026

npx -y skills add majiayu000/claude-skill-registry --skill layerchart --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Layerchart?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Layerchart
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majiayu000-layerchart/badge)](https://www.skillsdirectory.com/skills/majiayu000-layerchart)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: layerchart
description: Expert guide for LayerChart, a Svelte component library for building diverse data visualizations (Cartesian, radial, hierarchical, geo, graph) with unopinionated building blocks, motion primitives, and advanced interactions.
license: MIT
---

# LayerChart Skill

LayerChart is a comprehensive Svelte visualization framework built on Layer Cake, providing composable components for creating responsive, interactive charts across multiple visualization types.

## Core Architecture

LayerChart operates on a component-based, data-driven philosophy. The library provides three primary categories of components:

**Data-Driven Components** render visual marks directly from data (Area, Bars, Spline, Pie, Sunburst, Treemap, Sankey, etc.). These components automatically handle scale transformations through LayerCake's context.

**Motion-Enabled SVG Primitives** (Rect, Circle, Arc, Group, Line, Path) provide low-level drawing utilities with built-in Svelte transition support for animated data updates.

**Utility Components** handle legends, tooltips, pan/zoom interactions, annotations, and layout operations (hierarchy, geo projections).

## Visualization Types

- **Cartesian**: Bar, Area, Stack, Scatter, Histogram, ClevelandDotPlot, BarStack, BarGroup
- **Radial**: Pie, Arc, Sunburst, Threshold
- **Hierarchical**: Pack, Tree, Treemap, Partition
- **Graph**: Sankey, Link, Graph utilities
- **Geographic**: Choropleth, Spike Map, Bubble Map, GeoTile, GeoPath, StateMap, AnimatedGlobe, Globe projections (Mercator, Azimuthal, Equal Earth, etc.)

## Key Patterns

### Data Preparation

Use LayerChart's data transformation utilities before passing to visualizations:

- `stack()` - converts wide-format data into stacked series with baseline/top values
- `bin()` - groups data into histogram bins with x0/x1 bounds
- `groupLonger()` - pivots wide-format to long-format (one row per value)
- `flatten()` - flattens nested arrays one level, with optional accessor
- `calcExtents()` - calculates min/max across multiple fields, skipping nulls

### Component Composition

All LayerChart visualizations sit within a LayerCake wrapper that establishes scales and context. Child components access scales via Svelte's context API.

```svelte
<LayerCake x="date" y="value" data={data} padding={{...}}>
  <Svg>
    <Area />
    <Line />
    <AxisX />
  </Svg>
  <Canvas>
    <Points />  <!-- High-performance canvas rendering -->
  </Canvas>
  <Html>
    <Tooltip />
  </Html>
</LayerCake>
```

### Interaction Patterns

- **Tooltips**: Position over data with snap-to-data options
- **Pan/Zoom**: Built-in context utilities for interactive navigation
- **Highlighting**: Hover states trigger visual emphasis (opacity, stroke changes)
- **Selection**: Use reactive variables and event handlers for interactive filtering

### Responsive Design

LayerChart automatically handles responsive layouts via `padding` configuration and container dimensions. Components reactively update when data or scales change.

## Common Implementation Tasks

**Bar Charts**: Use `Bars` component with `x` as categorical field. Stack with `BarStack` or group with `BarGroup` for multi-series.

**Time Series**: Configure `xScale={scaleTime()}` with temporal data. Use `AxisX` with `tickFormat` for readable date labels.

**Geographic Visualizations**: Select appropriate projection (Mercator for web maps, Azimuthal for polar), use `GeoPath` for boundaries, `Choropleth` for value mapping.

**High-Volume Data**: Render marks via Canvas instead of SVG for 5000+ points. Layer SVG axes/legends with Canvas for hybrid rendering.

**Stacked/Grouped Series**: Use `stack()` utility to transform data, then render via `AreaStack`/`BarStack` components.

## Performance Considerations

- Canvas rendering for 5000+ points (~60fps on modern hardware)
- SVG for interactive elements and animations (<500 points recommended)
- Hybrid approach: Canvas for marks + SVG for axes/legends
- Scale calculations are reactive—only update scales when data/domain changes
- Memoize expensive data transforms outside component lifecycle

## Styling and Customization

All primitive components support standard SVG/Canvas attributes (stroke, fill, opacity, strokeWidth). Use Svelte's reactive statements for conditional styling based on interaction state or data values.

Gradient fills, patterns, and clipping available via `ClipPath`, `RectClipPath`, `CircleClipPath` components with SVG `<defs>`.

## Integration Notes

- Works seamlessly with D3 scales (linear, time, ordinal, log, threshold)
- Supports multiple render contexts in same chart (SVG + Canvas + HTML)
- Fully accessible with ARIA attributes on SVG elements
- SSR-compatible for server-side rendering in SvelteKit
- Zero external dependencies beyond Svelte and d3-array utilities

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.