Health.md Visualizations

approved

by codybontecou

This plugin has not been manually reviewed by Obsidian staff. Render Apple Health data visualizations from JSON, CSV, or Markdown files, with custom date and time range filters.

9 stars593 downloadsUpdated 25d agoMIT
CleanShot 2026-04-21 at 09 31 33@2x

Health.md Visualizations

An Obsidian plugin that renders rich Apple Health visualizations from data files in your vault. Drop a fenced code block into any note (including a daily note) and the plugin renders an interactive canvas chart pulled from your local health data.

Supported data formats: JSON, CSV, Markdown frontmatter, and Obsidian Bases (YAML frontmatter).

Download Health.md on the app store to easily export and get access to your Apple Health data.

Installation

Install Health.md from the Obsidian Community Plugins directory:

https://community.obsidian.md/plugins/health-md

Manual

If you prefer to install manually:

  1. Download main.js, manifest.json, and styles.css from the latest release.
  2. Copy them into <your vault>/.obsidian/plugins/health-md/.
  3. Reload Obsidian and enable Health.md Visualizations in Settings → Community plugins.

From source

git clone https://github.com/CodyBontecou/health-md-visualizations.git
cd health-md-visualizations
npm install
npm run build

Then copy main.js, manifest.json, and styles.css into your vault's plugin folder.

Quick start

  1. Put your Apple Health export files in a folder inside your vault — by default the plugin looks at Health/.

  2. In any note, add a fenced code block:

    ```health-viz
    type: heart-terrain
    ```
    
  3. Switch to reading view (or live preview) and the chart renders.

You can also run the Insert health visualization command from the command palette to open an insertion wizard. Pick a visualization category and type, then fill in the date range, renderer-specific options, and optional size before the plugin inserts the health-viz block at the cursor.

Settings

Open Settings → Health.md Visualizations:

SettingDescription
Data folderPath inside the vault where the plugin looks for health files. Default Health. Includes folder autocomplete in settings to reduce path typos.
File patternGlob to filter which files in that folder are loaded. Examples: * (all supported), *.json, 2026-*.md, health-*.csv.
Data formatauto (detect by file extension), json, csv, markdown, or bases. Markdown support requires YAML frontmatter (Bases-style).
Themeauto matches Obsidian, or force dark / light.
Default widthDefault canvas width in pixels (charts shrink to container width).
Default heightDefault canvas height in pixels.

The plugin watches your data folder and automatically refreshes its cache when files are added, modified, or deleted.

Visualization types

Specify one of these as the type: field in your code block. The gallery below shows each renderer with a short description; see examples/visualization-reference.md for complete argument tables, defaults, and copy/paste examples.

intro-stats visualization

intro-stats

HTML summary card — totals, averages, and highlights for the selected dataset.

Extra arguments: none.

summary-card visualization

summary-card

Apple-style headline card with large KPI, sparkline, range, and comparison delta.

Extra arguments: metric, compareWindow.

trend-tile visualization

trend-tile

Apple Health Trends-style HTML card with direction arrow, percent delta, narrative, and two-period sparkline.

Extra arguments: metric, currentWindow, priorWindow.

activity-rings visualization

activity-rings

Apple's Move / Exercise / Stand rings; single-day large ring set or multi-day small multiples.

Extra arguments: moveGoal, exerciseGoal, standGoal.

vitals-rings visualization

vitals-rings

Health.md radial activity/vitals rings: steps, calories, and heart-rate context per day.

Extra arguments: none.

bar-chart visualization

bar-chart

Apple-style vertical bars with latest day highlight, optional goal line, and optional average line.

Extra arguments: metric, goal, showAverage.

activity-heatmap visualization

activity-heatmap

GitHub-style activity calendar shaded by daily steps, calories, or distance.

Extra arguments: metric.

step-spiral visualization

step-spiral

Daily step counts arranged on a spiral, with older days near the center and newer days spiraling outward.

Extra arguments: none.

weekday-average visualization

weekday-average

Seven bars showing a metric's average by weekday with an overall-mean line.

Extra arguments: metric, weekStart.

heart-terrain visualization

heart-terrain

Heart-rate samples plotted as daily terrain / ridgeline rows over time.

Extra arguments: none.

heart-range visualization

heart-range

Per-day min-to-max heart-rate capsule with an average dot and optional resting-HR reference line.

Extra arguments: metric.

hrv-trend visualization

hrv-trend

HRV trend line from daily HRV or HRV samples.

Extra arguments: none.

oxygen-river visualization

oxygen-river

Blood oxygen samples as a flowing band with summary stats.

Extra arguments: none.

oxygen-range visualization

oxygen-range

Daily SpO₂ or respiratory min/max capsule with warning-zone shading.

Extra arguments: metric.

breathing-wave visualization

breathing-wave

Respiratory-rate samples as a wave.

Extra arguments: none.

sleep-schedule visualization

sleep-schedule

Horizontal bedtime-to-wake bars against a sunset→night→sunrise backdrop.

Extra arguments: sleepGoal, windowStart, windowEnd.

sleep-quality-bars visualization

sleep-quality-bars

Stacked nightly bars for deep, core, REM, and awake time.

Extra arguments: none.

sleep-architecture visualization

sleep-architecture

Linear timeline of sleep stages with depth bands.

Extra arguments: none.

sleep-polar visualization

sleep-polar

Polar clock view of sleep stages per night.

Extra arguments: none.

walking-symmetry visualization

walking-symmetry

Walking speed and asymmetry / gait metrics.

Extra arguments: none.

workout-log visualization

workout-log

Workout timeline with duration bars and workout-type colors.

Extra arguments: none.

workout-heart-rate visualization

workout-heart-rate

Heart-rate time series and optional zone bands for one workout.

Extra arguments: date, workout, maxHeartRate.

workout-map visualization

workout-map

GPS route map for one outdoor workout, colored by speed or heart rate.

Extra arguments: date, workout, colorBy.

All canvas chart types support hover tooltips and click-to-pin. The intro-stats, summary-card, trend-tile, and workout-map types are HTML/SVG/Leaflet renderers (no canvas tooltip layer) for sharper typography and interactive map rendering.

Bundled examples

Starter dashboards live in the examples/ folder — copy any of them into your vault to see the code blocks render:

  • examples/visualization-reference.md — comprehensive reference for every visualization, including supported arguments, defaults, metric values, and copy/paste blocks.
  • examples/apple-dashboard.md — full Apple Health-style summary using the Apple-inspired visualizations (summary cards, activity rings, heart range, bar chart, sleep schedule, weekday average, oxygen range, trend tiles).
  • examples/daily-dashboard.md — single-day overview for daily notes.
  • examples/weekly-overview.md — rolling week-at-a-glance across activity, heart, respiratory, sleep, mobility, and workouts.
  • examples/sleep-analysis.md — sleep-focused drill-down.

This repo also ships deterministic mock data in examples/Health/ (one JSON file per day from 2025-11-19 through 2026-12-31). When the default Health/ folder is empty or missing, the plugin falls back to this bundled dataset so cloned examples render immediately. You can also set Settings → Health.md Visualizations → Data folder to examples/Health explicitly.

Embedding charts in notes

A code block requires a type and accepts any of the optional config keys below. Each entry is a key: value line. Lines starting with # are comments.

```health-viz
type: vitals-rings
width: 600
height: 400
```

Common config keys

KeyTypeDefaultDescription
typestring(required)Visualization type — see the gallery above.
widthnumberfrom settingsCanvas width in pixels (chart shrinks to container width).
heightnumberfrom settingsCanvas height in pixels.
fromdate or datetimeStart of the data window (inclusive).
todate or datetimeEnd of the data window (inclusive).
lastnumberNumber of calendar days back to include.

Individual visualization types may accept additional keys — see examples/visualization-reference.md for every supported renderer-specific argument, default, and accepted value.

Filtering by date or date+time

Every visualization can be scoped to a custom window using from, to, and/or last. The filter is applied uniformly across all chart types — no need to learn per-chart syntax.

Just a date

from and to accept ISO calendar dates:

```health-viz
type: step-spiral
from: 2026-01-01
to: 2026-03-31
```

Open-ended ranges are fine too:

```health-viz
type: oxygen-river
from: 2026-04-01
```

Last N days

last: N is a rolling window of N calendar days ending today. last: 1 is just today; last: 30 is today plus the previous 29 days.

```health-viz
type: heart-terrain
last: 30
```

Combine last with to to anchor the window on a specific day instead of today:

```health-viz
type: vitals-rings
to: 2026-03-31
last: 7
```

This shows the 7-day window ending March 31, 2026.

Sub-day windows with datetimes

from and to also accept ISO datetimes — YYYY-MM-DDTHH:MM or YYYY-MM-DDTHH:MM:SS, with an optional Z or ±HH:MM timezone suffix. When you provide a time component, the plugin slices sub-day samples on the boundary days so the chart only shows data inside the requested window.

```health-viz
type: heart-terrain
from: 2026-04-09T06:00:00
to: 2026-04-09T12:00:00
```

The chart above renders only morning heart rate samples for April 9, 2026.

A multi-day window with precise endpoints:

```health-viz
type: oxygen-river
from: 2026-04-01T22:00:00
to: 2026-04-08T07:00:00
```

Includes April 1 from 10 PM onward, the full days April 2 through 7, and April 8 up to 7 AM.

You can mix datetimes with last:

```health-viz
type: breathing-wave
to: 2026-04-09T12:00:00
last: 7
```

A 7-day calendar window ending April 9, with samples after noon on April 9 trimmed.

Explicit timezones work too:

```health-viz
type: sleep-architecture
from: 2026-04-09T22:00:00-07:00
to: 2026-04-10T08:00:00-07:00
```

If you omit the timezone, the time is interpreted in your local timezone (matching JavaScript's Date.parse semantics).

Day-level aggregates are recomputed

When a sub-day window slices a boundary day's samples, day-level fields like averageHeartRate, bloodOxygenAvg, totalDuration, deepSleep, bedtime, etc. are automatically recomputed from the sliced samples. This means the stats shown alongside your charts (in intro-stats, sleep tooltips, vitals rings, and other panels) reflect the requested time window — not the full day.

The fields that are recomputed:

  • Heart: averageHeartRate, heartRateMin, heartRateMax, hrv
  • Vitals: bloodOxygenAvg/Min/Max (and the legacy bloodOxygenPercent), respiratoryRateAvg/Min/Max (and legacy respiratoryRate)
  • Sleep: totalDuration (deep + REM + core), deepSleep, remSleep, coreSleep, awakeTime, bedtime, wakeTime, plus all formatted-string variants
  • Workouts: filtered by startTime

A guard ensures aggregates aren't clobbered for days that were parsed from daily summaries without per-sample data — those days pass through unchanged.

Limitation: activity totals

Apple Health exports activity.steps, activity.activeCalories, activity.exerciseMinutes, activity.flightsClimbed, activity.standHours, activity.basalEnergyBurned, and mobility.* as daily totals only, with no underlying sub-day samples. There is no truthful way to slice those numbers for a partial day, so they pass through unchanged on boundary days. This affects the step ring in vitals-rings, the totals in step-spiral, and any walking/mobility metrics on a boundary day. Heart-rate–derived fields inside the same charts are recomputed correctly.

Validation

The plugin validates the date range up front and renders an inline error if something is off:

  • Invalid "from" value: ... Use YYYY-MM-DD or YYYY-MM-DDTHH:MM[:SS].
  • Invalid "last": ... Use a positive number of days.
  • "from" (...) is after "to" (...).
  • No health data in range (...). — when the window is valid but produces an empty result.

Daily-note tip

Add a health-viz block to your daily-note template (Templates or Templater plugin) and have a moving "last N days" view automatically appear in every new daily note:

```health-viz
type: heart-terrain
last: 7
```

Because last is anchored on today by default, each new daily note shows the most recent 7 days at the moment you open it. The plugin's data cache invalidates whenever files in your data folder change, so the chart always reflects the latest export.

Data format reference

The plugin auto-detects the data format from the file extension. Each file should represent one day of health data and live inside your configured data folder.

  • .json — A HealthDay object (see src/types.ts for the full shape).
  • .csv — Section headers (Heart, Sleep, Vitals, Activity, Mobility, …) followed by Metric,Value rows. See src/parsers/csv-parser.ts.
  • .md — A markdown file with YAML frontmatter that uses fields like heart_rate_avg, sleep_deep, steps, etc. Plain markdown without frontmatter is not parsed. See src/parsers/markdown-parser.ts. This format is compatible with Obsidian Bases.

The top-level date field on each day must be a YYYY-MM-DD ISO date — the date filter does fast lexicographic comparisons against this field.

Development

npm install
npm run dev      # esbuild watch mode
npm run build    # production build

Source layout:

  • src/main.ts — plugin entry point and settings tab
  • src/renderer.ts — code-block processor, config parsing, date range filtering, and aggregate recomputation
  • src/data-loader.ts — vault-aware data loader with cache invalidation
  • src/parsers/ — JSON, CSV, and Markdown parsers
  • src/visualizations/ — one file per chart type, plus intro-stats.ts (HTML)
  • src/canvas-utils.ts — shared canvas helpers and color palettes
  • src/types.tsHealthDay, VizConfig, HitRegion, render-fn signatures

License

MIT — see package.json.

For plugin developers

Search results and similarity scores are powered by semantic analysis of your plugin's README. If your plugin isn't appearing for searches you'd expect, try updating your README to clearly describe your plugin's purpose, features, and use cases.