Samsung Health

Analyze Samsung Health Connect data synced to Google Drive. Use for health tracking queries like sleep analysis, step counting, heart rate monitoring, SpO2 b...

Security Scan

The skill is internally consistent with its stated purpose (analyzing Samsung Health Connect exports from Google Drive) but it instructs the agent/user to clone and install code from a third‑party GitHub repository — review that repo before installing and run it in an isolated environment.

Purpose & Capability

Name/description match the required pieces: it needs access to Google Drive exports and a local CLI to parse Health Connect data. Declared required binaries (gog, python3) align with downloading from Drive and running a Python CLI.

Instruction Scope

SKILL.md stays on-topic: it instructs how to clone the repo, create a venv, configure a per-user ~/.config/samsung-health/config.yaml, and run the shealth CLI. It does not ask the agent to read unrelated system files or export data to unexpected endpoints. The one scope-expanding step is the instruction to fetch and install external code (see install_mechanism).

Install Mechanism

No packaged install spec in the registry, but SKILL.md directs git clone from https://github.com/mudgesbot/samsung-health-skill and pip install -e . GitHub is a common host, but cloning and installing arbitrary repository code executes remote code locally (setup/install hooks). Users/agents should review the repository before running the install commands and prefer an isolated environment.

Credentials

The skill requests no environment variables and only requires gog and python3. The config file asks for a Google Drive folder_id and account (expected for Drive access). No unrelated credentials or broad environment access are requested by the skill itself.

Persistence & Privilege

always:false and the skill asks to create its own config under ~/.config/samsung-health (normal). Be aware that an autonomous agent invocation could perform the git clone/pip install steps if permitted — this is a standard platform behavior but increases risk if you do not trust the repository.

Scan Findings in Context

[no_regex_findings] expected: The scanner found nothing because the skill package contains only SKILL.md (instruction-only). SKILL.md itself includes a git clone command; that external repository was not scanned.

Assessment

This skill appears to do what it says, but it fetches and installs code from a third‑party GitHub repository. Before installing or letting an agent run it automatically: (1) manually review the repo (README, setup.py/pyproject, and any install scripts) to ensure there are no malicious install hooks; (2) run the install inside an isolated environment (container or dedicated VM) and not on a sensitive host; (3) ensure the gog CLI is configured with a Google account you are willing to grant Drive access to (use the least-privilege account possible); (4) if you don't trust the repo, decline automatic installation and ask the skill author for a signed release or a vetted package instead.

Samsung Health Connect CLI

Analyze health data from Samsung Health Connect exported to Google Drive.

Prerequisites

• Samsung Galaxy Watch or Galaxy Ring with Samsung Health
• Samsung Health Connect app with Google Drive backup enabled
• gog CLI for Google Drive access (part of Clawdbot)

Installation

cd /path/to/workspace/projects
git clone https://github.com/mudgesbot/samsung-health-skill.git
cd samsung-health-skill
python3 -m venv .venv
source .venv/bin/activate
pip install -e .

Configuration

Create ~/.config/samsung-health/config.yaml:

google_drive:
  folder_id: "YOUR_FOLDER_ID"      # From Google Drive URL
  account: "[e-mail]"  # Google account for gog CLI
  file_name: "Health Connect.zip"

goals:
  daily_steps: 10000
  sleep_hours: 8

timezone: "Europe/Copenhagen"

Commands

All commands require activating the venv first:

cd /path/to/samsung-health-skill && source .venv/bin/activate

Sync Data

shealth sync              # Download latest from Google Drive
shealth sync --force      # Force re-download

Quick Daily View

shealth today             # Today's snapshot (steps, sleep, HR, SpO2)

Status

shealth status            # Data freshness, record counts, date range

Sleep Analysis

shealth sleep             # Last 7 days
shealth sleep --days 14   # Custom period

Shows: duration, stage breakdown (Light/Deep/REM/Awake), trends.

Step Tracking

shealth steps             # Last 7 days
shealth steps --week      # Weekly view
shealth steps --month     # Monthly view

Shows: daily counts, goal progress, streaks.

Heart Rate

shealth heart             # Last 7 days
shealth heart --days 14   # Custom period

Shows: average, min/max, daily trends.

Blood Oxygen (SpO2)

shealth spo2              # Last 7 days
shealth spo2 --days 14    # Custom period

Shows: average, range, trend. Normal: 95-100%.

Workouts

shealth workout           # Last 30 days
shealth workout --days 90 # Custom period

Shows: session count, duration, types (Walking, Running, Swimming, etc.)

Health Report

shealth report            # Comprehensive 7-day summary
shealth report --days 14  # Custom period

Shows: Energy Score, sleep summary, activity, heart rate.

JSON Output

Add --json flag to any command for machine-readable output:

shealth --json today
shealth --json sleep --days 7
shealth --json report

Sleep Stage Codes

Code	Stage
1	Light Sleep
4	Deep Sleep
5	Awake
6	REM Sleep

Exercise Types

Code	Type
53	Walking
33	Running
61	Hiking
21	Cycling
58	Swimming
4	Weight Training
66	Yoga

Troubleshooting

"Database not found" — Run shealth sync first.

"Google Drive not configured" — Create config.yaml with folder_id and account.

SpO2/HRV shows 0 records — Requires Galaxy Watch 4+ or Galaxy Ring; may need enabling in Samsung Health settings.

Source

GitHub: https://github.com/mudgesbot/samsung-health-skill