Charts 3d Skill for Claude

3D chart visualization with Swift Charts using Chart3D, SurfacePlot, interactive pose control, and surface styling — plus a 2D Swift Charts construction reference (marks, axes, selection, SectorMark, scrollable charts). Use when creating data visualizations with Swift Charts.

5k tokens

context cost

the whole folder, loaded on every use

files

instructions only

copies elsewhere

how many repositories repackaged it

585

stars on the repo

on the repository, not the skill itself

Install

one command, takes just this skill from the repository

npx skills add https://github.com/rshankras/claude-code-apple-skills --skill charts-3d

The instruction itself

42 sections, as written by the author

3D Charts with Swift Charts

Create 3D data visualizations using Chart3D and SurfacePlot. Covers math-driven surfaces, data-driven surfaces, interactive camera pose control, surface styling, and camera projection modes.

When This Skill Activates

Use this skill when the user:

Wants to create a 3D chart or 3D data visualization
Asks about Chart3D, SurfacePlot, or 3D surface plots
Needs to visualize a mathematical function as a 3D surface
Wants interactive drag-to-rotate on a 3D chart
Asks about 3D chart camera angles, pose, or projection
Needs to style 3D surfaces with gradients or height-based coloring
Wants to render multiple surfaces in a single 3D chart
Asks about data-driven 3D plots from an array of points
Needs the 2D Swift Charts API layer: marks, axes, series styling, selection, SectorMark pies/donuts, or scrollable charts

Decision Tree

What 3D chart feature do you need?
|
+-- Visualize a math function f(x, y) -> z
|   +-- Use SurfacePlot(x:y:z:function:)
|
+-- Visualize data points as a surface
|   +-- Use Chart3D(data) { point in SurfacePlot(...) }
|
+-- Interactive drag-to-rotate
|   +-- Bind pose: .chart3DPose($pose) with @State var pose: Chart3DPose
|
+-- Fixed viewing angle (no interaction)
|   +-- Read-only pose: .chart3DPose(Chart3DPose.front) or custom
|
+-- Style the surface color
|   +-- Solid color -> .foregroundStyle(Color.blue)
|   +-- Gradient -> .foregroundStyle(LinearGradient(...))
|   +-- Height-based -> .foregroundStyle(.heightBased(gradient, yRange:))
|   +-- Normal-based -> .foregroundStyle(.normalBased)
|
+-- Camera projection
|   +-- Perspective (depth) -> .chart3DCameraProjection(.perspective)
|   +-- Orthographic (flat) -> .chart3DCameraProjection(.orthographic)
|   +-- System default -> .chart3DCameraProjection(.automatic)
|
+-- Multiple surfaces in one chart
    +-- Place multiple SurfacePlot calls inside a single Chart3D { }

API Availability

|-----|----------------|--------|-------|

Quick Start

Math-Driven Surface

Render a surface from a function f(x, y) -> z:

import SwiftUI
import Charts

struct WaveSurfaceView: View {
    var body: some View {
        Chart3D {
            SurfacePlot(
                x: "X",
                y: "Height",
                z: "Z",
                function: { x, z in
                    sin(x) * cos(z)
                }
            )
            .foregroundStyle(.blue)
        }
    }
}

Data-Driven Surface

Render a surface from an array of data points:

import SwiftUI
import Charts

struct DataPoint: Identifiable {
    let id = UUID()
    let x: Double
    let y: Double
    let z: Double
}

struct DataSurfaceView: View {
    let points: [DataPoint]

    var body: some View {
        Chart3D(points) { point in
            SurfacePlot(
                x: .value("X", point.x),
                y: .value("Height", point.y),
                z: .value("Z", point.z)
            )
        }
    }
}

Interactive Rotation

Allow the user to drag to rotate the chart:

import SwiftUI
import Charts

struct InteractiveChartView: View {
    @State private var pose = Chart3DPose.default

    var body: some View {
        Chart3D {
            SurfacePlot(
                x: "X",
                y: "Height",
                z: "Z",
                function: { x, z in
                    sin(x) * cos(z)
                }
            )
            .foregroundStyle(.blue)
        }
        .chart3DPose($pose)
    }
}

Surface Styling Patterns

Solid Color

SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in x * z })
    .foregroundStyle(.blue)

Linear Gradient

SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in x * z })
    .foregroundStyle(
        LinearGradient(
            colors: [.blue, .green, .yellow],
            startPoint: .bottom,
            endPoint: .top
        )
    )

Height-Based Surface Style

Color the surface based on height values, mapping a gradient across the y-axis range:

SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in sin(x) * cos(z) })
    .foregroundStyle(
        Chart3DSurfaceStyle.heightBased(
            Gradient(colors: [.blue, .cyan, .green, .yellow, .red]),
            yRange: -1...1
        )
    )

Normal-Based Surface Style

Color based on surface normals, giving a lighting-aware appearance:

SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in sin(x) * cos(z) })
    .foregroundStyle(Chart3DSurfaceStyle.normalBased)

Surface Roughness

Control surface shininess (0 = reflective, 1 = matte):

SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in sin(x) * cos(z) })
    .foregroundStyle(.blue)
    .roughness(0.3)

Interactive Pose Control

Preset Poses

Chart3DPose provides built-in presets for common viewing angles:

.chart3DPose(.default)   // Standard 3/4 angle
.chart3DPose(.front)     // Viewing from front
.chart3DPose(.back)      // Viewing from back
.chart3DPose(.top)       // Top-down view
.chart3DPose(.bottom)    // Bottom-up view
.chart3DPose(.right)     // Right side view
.chart3DPose(.left)      // Left side view

Custom Pose

Specify exact azimuth (horizontal rotation) and inclination (vertical tilt):

.chart3DPose(
    Chart3DPose(azimuth: .degrees(45), inclination: .degrees(30))
)

Anti-Patterns

// ❌ Passing a literal where a binding is needed for interactivity
.chart3DPose(.default) // This is read-only; drag gestures will not work

// ✅ Use a @State binding for interactive rotation
@State private var pose = Chart3DPose.default
// ...
.chart3DPose($pose)

Camera Projection

Control how 3D depth is rendered:

Chart3D {
    SurfacePlot(x: "X", y: "Y", z: "Z", function: { x, z in sin(x) * cos(z) })
        .foregroundStyle(.blue)
}
.chart3DCameraProjection(.perspective)   // Objects farther away appear smaller
// .chart3DCameraProjection(.orthographic)  // No perspective distortion
// .chart3DCameraProjection(.automatic)     // System decides

Multiple Surfaces

Render multiple surfaces in a single chart for comparison:

import SwiftUI
import Charts

struct ComparisonChartView: View {
    @State private var pose = Chart3DPose.default

    var body: some View {
        Chart3D {
            SurfacePlot(
                x: "X",
                y: "Wave A",
                z: "Z",
                function: { x, z in sin(x) * cos(z) }
            )
            .foregroundStyle(.blue.opacity(0.8))

            SurfacePlot(
                x: "X",
                y: "Wave B",
                z: "Z",
                function: { x, z in cos(x) * sin(z) }
            )
            .foregroundStyle(.red.opacity(0.8))
        }
        .chart3DPose($pose)
        .chart3DCameraProjection(.perspective)
    }
}

Complete Example

A full-featured 3D chart with height-based coloring, interactive rotation, and perspective projection:

import SwiftUI
import Charts

struct TerrainView: View {
    @State private var pose = Chart3DPose(
        azimuth: .degrees(30),
        inclination: .degrees(25)
    )

    var body: some View {
        VStack {
            Text("Terrain Visualization")
                .font(.headline)

            Chart3D {
                SurfacePlot(
                    x: "Longitude",
                    y: "Elevation",
                    z: "Latitude",
                    function: { x, z in
                        let distance = sqrt(x * x + z * z)
                        return sin(distance) / max(distance, 0.1)
                    }
                )
                .foregroundStyle(
                    Chart3DSurfaceStyle.heightBased(
                        Gradient(colors: [
                            .blue, .cyan, .green, .yellow, .orange, .red
                        ]),
                        yRange: -0.5...1.0
                    )
                )
                .roughness(0.4)
            }
            .chart3DPose($pose)
            .chart3DCameraProjection(.perspective)
        }
        .padding()
    }
}

Chart Design Fundamentals (WWDC22)

These apply to every chart you build — 2D or 3D. The pillars: focused, approachable, accessible.

Pick marks by goal: bars for patterns, ranges, and individual values; points for spotting outliers; lines for rates of change.
Test with real, noisy data early — placeholder sine waves hide layout failures. Design the edge cases explicitly (e.g., gaps in a line chart where data is missing).
~4 horizontal gridlines is a good baseline; pick intuitive intervals — multiples of 20 for counts, multiples of 7 for day-based axes.
Descriptive takeaway title ("Total Sales: 1,234 Pancakes"), not a generic label ("Sales") — and give the chart surrounding context.
Touch targets stretch to full chart height — a tap anywhere in a bar's column selects it, not just the drawn pixels.
Support touch, keyboard, Voice Control, Switch Control, and VoiceOver equally — every interaction needs a non-pointer path.
Color enhances, never solely conveys — differentiate series with symbols/shapes first, color second. Verify contrast in Dark, Light, and Increase Contrast, and balance saturation so no series visually outweighs the others.
Every visual encoding needs a non-visual representation — per-mark VoiceOver labels and an Audio Graph descriptor; Swift Charts provides both by default, so don't break them with custom drawing. Custom-drawn charts implement them by hand — see "Audio Graphs: AXChartDescriptor" below.

Swift Charts Construction Reference (2D)

The design fundamentals above apply to every chart; this is the API layer for standard 2D charts (iOS 16+ unless noted; selection, SectorMark, and scrolling are iOS 17+).

Marks Compose

A Chart is a composition of marks — BarMark, LineMark, PointMark, AreaMark, RuleMark, RectangleMark. The .value("Label", v) factory arguments do double duty: they bind data AND drive the automatic axes and legend, so label them meaningfully:

Chart(salesData) { sale in          // Identifiable data — no explicit ForEach needed
    BarMark(
        x: .value("Day", sale.day, unit: .day),   // unit: .day buckets temporal values per day
        y: .value("Sales", sale.count)
    )
}

Transpose a chart by swapping the x/y value pairs — horizontal bars need no other change.
unit: (.day, .month, .hour) controls temporal bucketing; omit it and every timestamp is its own position.

Series: Style + Symbol, Never Color Alone

.foregroundStyle(by:) splits marks into series; pair it with .symbol(by:) so series stay distinguishable without color (WWDC22):

Chart(data) { point in
    LineMark(x: .value("Day", point.day), y: .value("Sales", point.sales))
        .foregroundStyle(by: .value("City", point.city))
        .symbol(by: .value("City", point.city))
}

Remap series colors with .chartForegroundStyleScale(["Cupertino": .indigo, "San Francisco": .teal]).
.position(by: .value("City", point.city)) converts stacked bars into grouped bars.

Pin Scales for Stability

Pin scales so a filtered dataset doesn't make the whole chart jump:

.chartYScale(domain: 0...maxExpectedSales)
.chartXScale(domain: startDate...endDate)

Composing Statistics into a Chart

Marks compose freely, so summary statistics are just more marks:

Chart {
    ForEach(data) { point in
        AreaMark(                                    // min–max band
            x: .value("Day", point.day),
            yStart: .value("Min", point.min),
            yEnd: .value("Max", point.max)
        )
        .opacity(0.3)

        LineMark(x: .value("Day", point.day), y: .value("Average", point.average))
    }

    RuleMark(y: .value("Overall", overallAverage))   // overall average line
        .foregroundStyle(.secondary)
        .annotation(position: .top, alignment: .leading) {
            Text("Avg: \(overallAverage, format: .number.precision(.fractionLength(0)))")
                .font(.caption)
                .foregroundStyle(.secondary)
        }
}

Axes Are Builders

.chartXAxis {
    AxisMarks(values: .stride(by: .month)) { value in
        AxisGridLine()
        AxisTick()
        AxisValueLabel(format: .dateTime.month(.narrow))
    }
}
.chartYAxis {
    AxisMarks(position: .leading)   // move the value axis to the leading edge
}

Conditional styling belongs *inside* the builder — check value.as(Date.self) in the closure to, say, bold only the first month of each quarter.
.chartXAxis(.hidden) removes an axis entirely; .chartPlotStyle { $0.frame(height: 200).background(.gray.opacity(0.05)).border(.quaternary) } sizes and styles the plot area itself.

Interactivity: Selection First

Prefer the built-in selection binding over hand-rolled overlay gestures (iOS 17):

@State private var selectedDay: Date?

Chart(data) { ... }
    .chartXSelection(value: $selectedDay)

Render the selection as marks — a RuleMark with zIndex(-1) so it draws behind the data, and an annotation that stays inside the plot:

if let selectedDay {
    RuleMark(x: .value("Selected", selectedDay, unit: .day))
        .foregroundStyle(.gray.opacity(0.3))
        .zIndex(-1)
        .annotation(
            position: .top, spacing: 0,
            overflowResolution: .init(x: .fit(to: .chart), y: .disabled)
        ) {
            SelectionDetailCard(day: selectedDay)
        }
}

For fully custom hit-testing, drop to ChartProxy inside .chartOverlay with a GeometryReader: proxy.value(atX:) converts gesture locations to data values, proxy.position(forX:) converts back.

SectorMark: Pies and Donuts (iOS 17+)

Chart(data) { item in
    SectorMark(
        angle: .value("Sales", item.sales),
        innerRadius: .ratio(0.62),     // donut hole
        angularInset: 1.5              // 1.5 per side = 3pt gaps between sectors
    )
    .cornerRadius(4)
    .foregroundStyle(by: .value("Name", item.name))
}
.chartBackground { proxy in
    GeometryReader { geo in            // headline metric in the donut hole
        if let anchor = proxy.plotFrame {
            let frame = geo[anchor]
            Text("Best: \(topSellerName)")
                .position(x: frame.midX, y: frame.midY)
        }
    }
}

Large Datasets: Scroll a Window

Don't cram a year into one screen — show a window and scroll (WWDC23):

Chart(yearOfData) { ... }
    .chartScrollableAxes(.horizontal)
    .chartXVisibleDomain(length: 3600 * 24 * 30)   // 30-day window
    .chartScrollPosition(x: $scrollDate)           // read/write the scroll offset
    .chartScrollTargetBehavior(
        .valueAligned(
            matching: DateComponents(hour: 0),                    // land on day boundaries
            majorAlignment: .matching(DateComponents(day: 1))     // snap paging to month starts
        )
    )

Accessibility Per Mark

Auto-generated VoiceOver descriptions read raw values; per-mark labels beat them:

BarMark(x: .value("Day", sale.day, unit: .day), y: .value("Sales", sale.count))
    .accessibilityLabel(sale.day.formatted(date: .abbreviated, time: .omitted))
    .accessibilityValue("\(sale.count) pancakes sold")

With hundreds or thousands of points, bucket into reasonable intervals and expose one

accessibility element per interval rather than per point (WWDC21 10122).

Audio Graphs: AXChartDescriptor (WWDC21 10122)

Audio Graphs let VoiceOver play a data series as a continuous tone — pitch = Y value,

time = X position — with an explorer view (rotor → "Audio Graph" → Chart Details) that

plays the sonification, scrubs it (double-tap-and-hold; pausing speaks the value at the

current position), and shows automatically computed summary statistics. Swift Charts

generates a default descriptor; custom-drawn charts conform to AXChart (UIKit) or use

the .accessibilityChartDescriptor(_:) modifier with an AXChartDescriptorRepresentable

(SwiftUI, iOS 15+):

var accessibilityChartDescriptor: AXChartDescriptor? {
    let xAxis = AXNumericDataAxisDescriptor(
        title: "Cups of coffee",
        range: 0...10,
        gridlinePositions: [],                              // gridlines render as haptics during playback
        valueDescriptionProvider: { "\(Int($0)) cups" })    // "5 cups", never a bare "5"

    let yAxis = AXNumericDataAxisDescriptor(
        title: "Lines of code",
        range: 0...100,
        gridlinePositions: [],
        valueDescriptionProvider: { "\(Int($0)) lines of code" })

    let series = AXDataSeriesDescriptor(
        name: "Productivity",
        isContinuous: true,      // line → one continuous tone; false for bars/points → discrete tones
        dataPoints: model.points.map { AXDataPoint(x: $0.x, y: $0.y) })

    return AXChartDescriptor(
        title: model.title,
        summary: model.summary,  // 1–2 sentence alt text; spoken in the explorer view
        xAxis: xAxis,
        yAxis: yAxis,
        additionalAxes: [],
        series: [series])
}

Categorical axes use AXCategoricalDataAxisDescriptor; localize and pluralize the

valueDescriptionProvider output in production.

Visual floor for any chart (WWDC21 10122): foreground/background contrast ≥ 4.5:1

(verify with Accessibility Inspector's Color Contrast Calculator); **never pair

red + green (the most common color blindness) and avoid blue + yellow** (the second

most common); use symbols in addition to color so series stay distinguishable with

no color perception at all.

Top Mistakes

| # | Mistake | Fix |

|---|---------|-----|

| 1 | Forgetting to import Charts | Both SwiftUI and Charts imports are required |

| 2 | Using .chart3DPose(.default) and expecting drag-to-rotate | Use a @State binding: .chart3DPose($pose) for interactive rotation |

| 3 | Setting yRange that does not cover actual function output | Match the yRange in .heightBased() to the actual min/max of your function output |

| 4 | Applying .roughness() without .foregroundStyle() | Roughness modifies existing surface appearance; set a foreground style first |

| 5 | Using orthographic projection for presentation/demo contexts | Prefer .perspective for visual appeal; use .orthographic for precise data reading |

Review Checklist

Imports and Setup

[ ] Both import SwiftUI and import Charts are present
[ ] Deployment target is iOS 26 / macOS 26 or later
[ ] Chart3D wraps all SurfacePlot content

Surface Configuration

[ ] Axis labels (x:, y:, z:) are descriptive and meaningful
[ ] foregroundStyle applied to each SurfacePlot for clear visual distinction
[ ] yRange in .heightBased() matches the actual output range of the function
[ ] roughness value makes sense for the use case (0 = reflective, 1 = matte)

Interactivity

[ ] Pose is a @State binding if drag-to-rotate is intended
[ ] Pose is a constant (non-binding) if rotation should be locked
[ ] Initial pose angle provides a good default view of the data

Camera

[ ] Camera projection set appropriately (.perspective for visual, .orthographic for precision)
[ ] If using .automatic, verified the system choice looks acceptable

Multiple Surfaces

[ ] Each surface has a distinct color or opacity to differentiate
[ ] Axis labels are unique per surface or shared where appropriate

Accessibility

[ ] Contrast ≥ 4.5:1; no red+green or blue+yellow series pairs; symbols supplement color (WWDC21 10122)
[ ] Custom-drawn charts expose an AXChartDescriptor (Audio Graphs) — or keep Swift Charts' default intact
[ ] Large datasets expose one accessibility element per interval, not per point

References

Other skills for the same job

different authors, same section of the catalogue

XLSX

by anthropics

vendor ×15

Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify existing spreadsheets while preserving formulas, (4) Data analysis and visualization in spreadsheets, or (5) Recalculating formulas

Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

Picks random winners from lists, spreadsheets, or Google Sheets for giveaways, raffles, and contests. Ensures fair, unbiased selection with transparency.

Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.

MATLAB and GNU Octave numerical computing for matrix operations, data analysis, visualization, and scientific computing. Use when writing MATLAB/Octave scripts for linear algebra, signal processing, image processing, differential equations, optimization, statistics, or creating scientific visualizations. Also use when the user needs help with MATLAB syntax, functions, or wants to convert between MATLAB and Python code. Scripts can be executed with MATLAB or the open-source GNU Octave interpreter.

UMAP dimensionality reduction. Fast nonlinear manifold learning for 2D/3D visualization, clustering preprocessing (HDBSCAN), supervised/parametric UMAP, for high-dimensional data.

Creating interactive data visualisations using d3.js. This skill should be used when creating custom charts, graphs, network diagrams, geographic visualisations, or any complex SVG-based data visualisation that requires fine-grained control over visual elements, transitions, or interactions. Use this for bespoke visualisations beyond standard charting libraries, whether in React, Vue, Svelte, vanilla JavaScript, or any other environment.

Access AlphaFold 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.

7k tokens

How to use it

Copy the folder

Take rshankras/charts-3d from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.