wtf is artemisia?

Artemisia is a private, self-hosted app for tracking your menstrual cycle. All of your data is stored in a single file on your own computer, using SQLite, a lightweight database format. There are no user accounts to create, no cloud servers involved, no advertising, and no subscription fee. The whole application is one Python file that you run yourself. The app lets you log your period flow, symptoms, mood, energy level, discharge, and personal notes for each day directly on a calendar, which also marks the real phase of the moon and shows which day of your cycle you are on. It predicts your next period as a range of likely dates based on your own past cycles, rather than a single falsely precise date, and it gives similarly labeled estimates for your fertile window and ovulation. A year view lets you see twelve months at once so long term patterns become visible. An insights section shows how your cycle length has changed over time and how often certain symptoms and moods occur. A statistics section, called the Almanac, calculates which symptoms tend to cluster in which phase of your cycle, tests whether your own cycles line up with the lunar cycle, and tracks rolling variability that can flag early signs of perimenopause. If you already track your cycle in the Clue app, Artemisia can import your full data export from Clue without losing or duplicating anything. You can also export everything as a CSV file or download the raw database file as a backup at any time. The app can be installed on a phone as a standalone app if you set up HTTPS access to your own server, for example using the free tool Tailscale. Running it requires installing Python on your own computer, installing a few dependencies with pip, and starting the app with a single command, after which you access it through a web browser at a local address. The README includes a full step by step guide aimed at people who have never used Python or a terminal before. The project is released under the MIT license and it explicitly states that none of its predictions are medical advice.