スキル一覧に戻る

create-spatial-visualization

pjt222
更新日 2 days ago
3 閲覧
17
2
17
GitHubで表示
メタdesigndata

について

このスキルは、R(sf、leaflet)またはObservable(D3、deck.gl)を使用して、GPXまたはルートデータからインタラクティブな地図と高度プロファイルを作成します。データのインポート、座標系の処理、地図のスタイリング、HTMLや画像へのエクスポートに対応しています。ツアールートの可視化、ハイキングやサイクリングの高度プロファイルの作成、ウェブベースの旅行ダッシュボードの構築にご利用ください。

クイックインストール

Claude Code

推奨
メイン
npx skills add pjt222/agent-almanac -a claude-code
プラグインコマンド代替
/plugin add https://github.com/pjt222/agent-almanac
Git クローン代替
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/create-spatial-visualization

このコマンドをClaude Codeにコピー&ペーストしてスキルをインストールします

ドキュメント

Create Spatial Visualization

Make interactive maps, elevation profiles, spatial visualizations from GPX tracks, waypoints, route data.

When Use

  • Visualizing planned or finished tour route on interactive map
  • Making elevation profiles for hiking or cycling routes
  • Overlaying waypoints, POIs, route corridors on basemap
  • Generating static map images for print reports
  • Building web-based trip dashboard with spatial data

Inputs

  • Required: Spatial data source (GPX file, CSV with lat/lon, GeoJSON, or waypoint list)
  • Required: Visualization type (interactive map, static map, elevation profile, heatmap)
  • Optional: Basemap preference (OpenStreetMap, satellite, terrain, topo)
  • Optional: Styling params (colors, line width, marker icons)
  • Optional: Output format (HTML widget, PNG, SVG, embedded in Quarto)
  • Optional: Extra layers (POI markers, area boundaries, distance markers)

Steps

Step 1: Import Spatial Data

Load and parse spatial data into usable format.

R approach (sf package):

# GPX file
track <- sf::st_read("route.gpx", layer = "tracks")
waypoints <- sf::st_read("route.gpx", layer = "waypoints")

# CSV with coordinates
points <- readr::read_csv("stops.csv") |>
  sf::st_as_sf(coords = c("lon", "lat"), crs = 4326)

# GeoJSON
route <- sf::st_read("route.geojson")

JavaScript approach (for Observable/D3):

// GPX parsing
const gpxText = await FileAttachment("route.gpx").text();
const parser = new DOMParser();
const gpxDoc = parser.parseFromString(gpxText, "text/xml");

// Extract track points
const trkpts = gpxDoc.querySelectorAll("trkpt");
const coordinates = Array.from(trkpts).map(pt => ({
  lat: +pt.getAttribute("lat"),
  lon: +pt.getAttribute("lon"),
  ele: +pt.querySelector("ele")?.textContent || 0
}));

Verify CRS is WGS 84 (EPSG:4326) for web maps.

Got: Spatial data loaded as sf object (R) or coordinate array (JS) with valid geometries. Point counts match expected input (e.g., GPX track has hundreds to thousands of points).

If fail: GPX parsing fails? Check file is valid XML. Common issues: truncated files from GPS battery death, mixed namespaces, GPX 1.0 vs 1.1 differences. CRS missing? Assign explicitly with sf::st_set_crs(data, 4326). Coordinates look inverted (lat/lon swapped)? Check column order.

Step 2: Process and Clean

Transform raw data into analysis-ready spatial features.

Processing Pipeline:
┌─────────────────────┬──────────────────────────────────────────┐
│ Operation           │ Purpose                                  │
├─────────────────────┼──────────────────────────────────────────┤
│ Remove duplicates   │ GPS often logs identical points at stops │
│ Smooth track        │ Reduce GPS jitter in dense urban areas   │
│ Calculate distances │ Cumulative distance along track          │
│ Extract elevation   │ Build elevation profile data             │
│ Segment by day      │ Split multi-day tracks into daily legs   │
│ Buffer route        │ Create corridor for POI discovery        │
│ Simplify geometry   │ Reduce point count for web performance   │
└─────────────────────┴──────────────────────────────────────────┘

R processing example:

# Calculate cumulative distance
track_points <- sf::st_cast(track, "POINT")
distances <- sf::st_distance(track_points[-nrow(track_points), ],
                             track_points[-1, ],
                             by_element = TRUE)
cumulative_km <- cumsum(as.numeric(distances)) / 1000

# Extract elevation profile data
elevation_df <- data.frame(
  distance_km = c(0, cumulative_km),
  elevation_m = sf::st_coordinates(track_points)[, 3]
)

# Simplify for web display (keep 1% of points)
track_simple <- sf::st_simplify(track, dTolerance = 0.001)

Got: Clean spatial data with calculated distances, elevation extracted, geometry simplified for target output. No NA coordinates, no zero-length segments.

If fail: Elevation data missing (common with some GPS devices)? Use DEM lookup service or note elevation profile unavailable. Track simplification removes critical shape detail? Lower tolerance value. Distance calculations produce NA? Check for empty geometries with sf::st_is_empty().

Step 3: Select Visualization Type

Pick and configure right visualization for data and audience.

Visualization Decision Matrix:
┌─────────────────────┬──────────────────────┬───────────────────┐
│ Type                │ Best for             │ Tool              │
├─────────────────────┼──────────────────────┼───────────────────┤
│ Interactive map     │ Web, exploration     │ leaflet (R),      │
│                     │                      │ deck.gl (JS)      │
├─────────────────────┼──────────────────────┼───────────────────┤
│ Static map          │ Print, reports       │ tmap (R),         │
│                     │                      │ ggplot2 + ggspatial│
├─────────────────────┼──────────────────────┼───────────────────┤
│ Elevation profile   │ Hiking/cycling       │ ggplot2, D3       │
│                     │ analysis             │                   │
├─────────────────────┼──────────────────────┼───────────────────┤
│ Heatmap             │ Visit density,       │ leaflet.extras,   │
│                     │ coverage             │ deck.gl HeatmapLayer│
├─────────────────────┼──────────────────────┼───────────────────┤
│ 3D terrain          │ Mountain routes      │ rayshader (R),    │
│                     │                      │ deck.gl TerrainLayer│
└─────────────────────┴──────────────────────┴───────────────────┘

Configure basemap tiles fitting content:

  • OpenStreetMap: General purpose, good labels
  • Stamen Terrain: Hiking and outdoor routes
  • ESRI World Imagery: Satellite context
  • OpenTopoMap: Topographic contours for elevation context

Got: Clear decision on visualization type and toolchain. Basemap picked to complement route data.

If fail: Picked tool cannot handle data volume (e.g., 100,000+ track points in leaflet)? Simplify geometry first or switch to canvas-based renderer (deck.gl). Basemap tiles unavailable (rare)? Fall back to OpenStreetMap as most reliable free option.

Step 4: Render Map or Chart

Build visualization with all layers and styling.

Interactive map (R/leaflet):

leaflet::leaflet() |>
  leaflet::addProviderTiles("OpenTopoMap") |>
  leaflet::addPolylines(
    data = track,
    color = "#2563eb",
    weight = 4,
    opacity = 0.8
  ) |>
  leaflet::addCircleMarkers(
    data = waypoints,
    radius = 8,
    color = "#dc2626",
    fillOpacity = 0.9,
    popup = ~name
  ) |>
  leaflet::addScaleBar(position = "bottomleft") |>
  leaflet::addMiniMap(position = "bottomright")

Elevation profile (R/ggplot2):

ggplot2::ggplot(elevation_df, ggplot2::aes(x = distance_km, y = elevation_m)) +
  ggplot2::geom_area(fill = "#93c5fd", alpha = 0.4) +
  ggplot2::geom_line(color = "#2563eb", linewidth = 0.8) +
  ggplot2::labs(
    x = "Distance (km)",
    y = "Elevation (m)",
    title = "Elevation Profile"
  ) +
  ggplot2::theme_minimal()

Add supplementary layers as needed: distance markers every N km, day-break indicators, difficulty-colored segments, POI icons.

Got: Rendered visualization clearly shows route, waypoints, any extra info. Interactive maps responsive with working popups and zoom. Elevation profiles have right axis scales.

If fail: Map renders but shows no data? Check coordinates in right CRS (EPSG:4326 for leaflet). Popups empty? Verify column names in popup formula. Elevation profile has extreme spikes? Filter GPS elevation errors (values off more than 100 m from neighbors).

Step 5: Export and Embed

Save visualization in target format.

Export Options:
┌───────────────────┬────────────────────────────────────────────┐
│ Format            │ Method                                     │
├───────────────────┼────────────────────────────────────────────┤
│ HTML widget       │ htmlwidgets::saveWidget(map, "map.html")   │
│ PNG (static)      │ mapview::mapshot() or ggplot2::ggsave()    │
│ SVG (vector)      │ ggplot2::ggsave("plot.svg")                │
│ Quarto embed      │ Place leaflet/ggplot code in .qmd chunk    │
│ GeoJSON export    │ sf::st_write(data, "output.geojson")       │
│ KML (Google Earth)│ sf::st_write(data, "output.kml")           │
└───────────────────┴────────────────────────────────────────────┘

For Quarto embedding:

  1. Place visualization code in code chunk with right labels
  2. Use #| fig-cap: for static plots or #| label: fig-map for cross-referencing
  3. Set self-contained: true in YAML to bundle tile images (grows file size)

Got: Exported file viewable in target context (browser for HTML, report for embedded, print for PNG/SVG). File size reasonable (under 5 MB for HTML widgets, under 1 MB for images).

If fail: HTML widget too large? Cut tile caching or simplify geometries. Quarto rendering fails with leaflet? Confirm htmlwidgets package installed and output format is HTML (leaflet does not render to PDF). For PDF output, use static map alternative (tmap with tmap_mode("plot")).

Checks

  • Spatial data imports without errors, has right CRS
  • All track points and waypoints render in expected geographic area
  • Elevation profile (if included) shows plausible values without extreme spikes
  • Interactive map has working zoom, pan, popups
  • Distance and elevation scales labeled right
  • Export file viewable in target format
  • File size fits delivery method

Pitfalls

  • CRS mismatch: Mixing EPSG:4326 (degrees) with projected CRS (meters) → data renders in wrong location or wrong scale. Always transform to EPSG:4326 for web maps.
  • GPS elevation noise: GPS-derived elevation far less accurate than horizontal position. Smooth elevation data or use DEM-based elevation for profiles.
  • Tile server rate limits: Fetching many tiles fast can trigger rate limits on free tile servers. Cache tiles locally for repeat rendering. Respect usage policies.
  • Over-detailed tracks: Raw GPS tracks with 1-second logging → huge files. Simplify before web display.
  • Leaflet in PDF: Leaflet maps cannot render in PDF output. Use tmap or ggplot2 with ggspatial for print formats.
  • Missing popups: Forgetting to add popup = ~column_name → markers with no info on click.

See Also

  • plan-tour-route — generate route data this skill visualizes
  • generate-tour-report — embed visualizations into formatted tour report
  • plan-hiking-tour — source of GPX and elevation data for hiking visualizations
  • create-quarto-report — Quarto rendering for embedding spatial visualizations

GitHub リポジトリ

pjt222/agent-almanac
パス: i18n/caveman/skills/create-spatial-visualization
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

関連スキル

content-collections

メタ

このスキルは、Content Collections(Markdown/MDXファイルを型安全なデータコレクションに変換するTypeScriptファーストのツール)の本番環境でテストされた設定を提供します。Zodバリデーションによる型安全性を実現し、ブログ、ドキュメントサイト、コンテンツ重視のVite + Reactアプリケーション構築時にご利用ください。Viteプラグインの設定、MDXコンパイルから、デプロイ最適化、スキーマバリデーションまで、すべてを網羅しています。

スキルを見る

polymarket

メタ

このスキルは、開発者がPolymarket予測市場プラットフォームを活用したアプリケーション構築を可能にします。API統合による取引や市場データの取得に加え、WebSocketを介したリアルタイムデータストリーミングにより、ライブ取引や市場活動を監視できます。取引戦略の実装や、ライブ市場更新を処理するツールの作成にご利用ください。

スキルを見る

creating-opencode-plugins

メタ

このスキルは、開発者がコマンド、ファイル、LSP操作など25種類以上のイベントタイプにフックするOpenCodeプラグインを作成することを支援します。JavaScript/TypeScriptモジュール向けに、プラグイン構造、イベントAPI仕様、および実装パターンを提供します。カスタムイベント駆動ロジックでOpenCode AIアシスタントのライフサイクルをインターセプト、監視、または拡張する必要がある場合にご利用ください。

スキルを見る

sglang

メタ

SGLangは、高性能なLLMサービングフレームワークであり、RadixAttentionプレフィックスキャッシュを活用したJSON、正規表現、エージェントワークフロー向けの高速で構造化された生成を特長とします。特にプレフィックスが繰り返されるタスクにおいて、大幅に高速な推論を実現し、複雑な構造化出力やマルチターン対話に最適です。制約付きデコードが必要な場合や、広範なプレフィックス共有を伴うアプリケーションを構築する場合は、vLLMなどの代替案ではなくSGLangを選択してください。

スキルを見る