Training plans look tidy until normal life gets a vote. Work runs late, sleep slips, your legs file a complaint, and the neat week you meant to follow turns into a small archaeology project.
That is where Garmin Connect data helps. A watch does not make you wise. It keeps score when your memory starts editing. Session duration, heart rate, laps, splits, training effect, the plain fact that Tuesday hit harder than you admitted on Wednesday.
I have been building an unofficial Garmin Connect SDK for that reason. It is read-first, TypeScript, and aimed at your own data. Useful surface, clear boundaries, no pretending private Garmin endpoints are a calm foundation for wild production promises.
The interesting part starts after the import. Data can tell you what happened. It does not decide what fits tonight. Plenty of tools leave you in that gap: lots of history, little judgment.
WODIQ is my attempt to close that gap without turning training into air traffic control. The app looks at your goal, recent rhythm and available time, then tries to pick a session that fits today. High recent load should downshift the plan. Thin history should keep the proposal conservative. Missing Garmin detail should lead to a plain caveat instead of fake confidence.
That last part matters. A planning tool should not act like a sports oracle because it found a few heart-rate numbers. Good product behaviour uses the signal, admits the limits and keeps the advice useful.
So the stack stays simple in spirit. Garmin gives the training diary. WODIQ uses that diary to make a better next choice. Less scrolling for a workout, less spreadsheet guilt, and fewer evenings where choosing the session becomes the session.
Public links if you want the clean version instead of this brain dump: Garmin Connect SDK on GitHub, WODIQ site, and the WODIQ app.