PickopenGym: Self-hosted health tracker on your server, own your data without subscription fees

DuarteSantos8/openGym

·2026.10.05 06:03

This is a self-hosted health tracker that keeps your workout records entirely in your hands, with no subscription fees or ads. Instead of entrusting server data to third-party clouds, you run it directly with Docker and store all information in local folders. It supports Face ID and fingerprint recognition via Passkey-based authentication and works fully offline.

1 / 6

It offers a library of 1,324 exercises with animated demos, allowing you to build routines based on your available equipment. It supports basic plans like Push/Pull/Legs as well as complex training techniques such as Supersets and Drop sets. It automatically suggests weights based on your last record and guides your workouts with guided sessions that include rest timers and PR notifications.

It applies various progressive overload rules, such as Greyskull LP and double progression, and automatically suggests a Deload when progress stalls. It analyzes the structural balance ratios from Poliquin and Thibaudeau to visualize volume and recovery status by muscle group. You can import data from major apps like FitNotes, Strong, and Hevy, and export all records as JSON for permanent storage.

If desired, you can connect your own API keys from providers like Anthropic or OpenAI to enable AI coaching features. It analyzes your training history to draft weekly routines and suggest changes, but final application requires user approval. Through an MCP server, external assistants like Claude Desktop can also reference your local data to answer questions.

GitHub
GitHub repository

DuarteSantos8/openGym

Self-hosted gym & body-weight tracker — plan routines, log workouts (supersets, warm-ups, cardio), see which muscles are trained, fatigued or detrained, import from FitNotes/Strong/Hevy, passkey login. Your data, your server.

JavaScript

This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.