Health Trend Analyzer - Automatic personal health data trend analysis and insights

Hey, you still judging health by feelings? Try this data-driven health trend analyzer

Lately, I've been obsessed with various health bands and apps, staring at step counts, heart rate, and sleep quality every day. But honestly, too much data makes it hard to see the bigger picture. Seeing that I walked 10,000 steps yesterday and 8,000 today — what does that really tell me? Is the trend improving or worsening? It wasn't until I discovered this health trend analyzer that I felt like I was truly managing my health, rather than being held hostage by numbers.

This skill is actually an open-source project on GitHub called health-trend-analyzer, hosted in the all-agent-skill repository by Anhvu1107. Its core idea is simple: aggregate health data scattered across different apps, then use statistical methods to uncover long-term trends. For example, is your heart rate variability increasing or decreasing over the past month? What about the cyclical fluctuations in sleep duration? These patterns, which are hard to spot with the naked eye, are automatically calculated by this tool.

What can it do? In a word: see through the numbers

Once installed, you can specify the data sources. Currently it supports importing files from Apple Health (HealthKit), Google Fit API, or directly uploading CSV files. Then it runs a set of analytical processes:

• Data Cleaning: Removes outliers (e.g., heart rate spikes to 200 due to sensor errors)
• Time Series Decomposition: Separates trend, seasonality, and random fluctuations
• Trend Fitting: Uses linear regression or moving average to draw a trend line
• Report Generation: Outputs a beautiful HTML page with line charts, bar charts, and key metrics summary

For example, I imported three months of sleep data. The result showed that while I average 7 hours per night, my deep sleep percentage has been declining over the last two weeks. Under normal circumstances, I wouldn't have noticed this because the total duration remained unchanged. But the trend analysis directly highlighted the downward slope and even offered suggestions: maybe it's time to adjust my pre-sleep habits.

Installation — super easy

Although the code is on GitHub, there is more than one way to install it. If you're a developer, you can take the traditional route:

git clone the repository
cd all-agent-skill/.agent/skills/health-trend-analyzer
npm install

But the more recommended method is to use npx for a one-click installation (works on all platforms):

npx skills add Anhvu1107/all-agent-skill --skill health-trend-analyzer

See? One command and you're done. After installation, fill in your data source configuration in config.js. For example, the file path for Apple Health data, or OAuth credentials for Google Fit. Then run npm start to start the analysis.

First run — I was blown away

Honestly, the first time I ran it, I was a bit nervous — afraid it wouldn't recognize my data format. But its data parser turned out to be very smart, automatically identifying timestamp and value columns. My exported Apple Health CSV had dozens of metrics, but it only extracted steps, heart rate, sleep, and calories, ignoring everything else without any manual filtering. A few minutes later, a report popped up in my browser, and seeing those charts gave me a real “aha” moment.

One of the charts was the “Weekly Step Trend.” I originally thought weekends would be higher (because I go for walks), but the chart revealed that Thursday and Friday were the peaks. After thinking about it, oh — I go to the company gym on those two days. This skill helped me look at my activity patterns more objectively.

Practical tips

• If you have a large amount of data (e.g., one year of data), it is recommended to enable aggregation mode: aggregating by week reduces noise and makes trends clearer.
• Supports custom analysis metrics: add other fields you care about in the config, such as blood oxygen, resting heart rate, etc.
• You can set up scheduled tasks: use GitHub Actions to run it every hour and automatically update the report.

Of course, it has some limitations. For example, it does not support real-time streaming data; it can only analyze exported historical data. But for a weekly health review, it's more than sufficient.

Final thoughts

Health management shouldn't just be about blindly chasing step counts. With the health trend analyzer, you can see the real changes in your data. Whether you're a fitness enthusiast or an ordinary person, you can use it to quantify your health. And it's open source and free — isn't that great? I've already added it to my daily toolkit.

Oh, and if you plan to try it out, don't forget to give Anhvu1107 a star on GitHub to show your support — after all, good projects deserve to be seen.