Haimantika

Haimantika / plank-posture

Public

An AI-powered app that acts as a trainer and helps you correct your plank posture.

44
12
100% credibility
Found Apr 13, 2026 at 44 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A macOS desktop app that monitors your plank exercise via webcam, analyzes posture in real-time, and delivers spoken corrections and encouragement.

How It Works

1
👀 Discover Plank Coach

You hear about this handy app that watches your plank form and speaks tips to keep you straight.

2
📱 Set Up Your Spot

Place your camera to the side at chest height so it sees your body sideways like a coach beside you.

3
🗣️ Connect Voice Tips

Link a voice service account so the app can talk to you with friendly corrections and cheers.

4
🚀 Start the App

Run the program, allow camera use, and see your live mirror view with controls ready.

5
🧘 Strike a Plank

Get into position and watch colorful lines and dots show if your hips, head, and body are aligned.

6
🔊 Hear Live Coaching

Listen to voice guidance like 'Lower your hips!' for fixes or 'Great form!' when you're nailing it.

7
⏹️ Wrap Up Session

Press quit to save a quick report of your workout time, tips given, and progress.

🏆 Stronger Planks Ahead

Review your log, celebrate improvements, and feel confident in better form next time.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 44 to 44 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is plank-posture?

Plank-posture is a Python-built AI-powered app for macOS that turns your side-mounted camera into a real-time plank trainer, detecting form breaks like sagging hips, raised head, or poor body angle via on-device pose estimation. It speaks corrections through text-to-speech and logs session stats to JSON, solving the problem of solo workouts lacking instant feedback. Users get visual overlays, sensitivity presets (low/medium/high), keyboard controls for pausing or tweaking, and compliments for good form—no gym needed.

Why is it gaining traction?

This stands out in the AI-powered apps GitHub space by running pose detection locally with minimal dependencies, only needing a TTS API key for voice, unlike cloud-heavy fitness trackers. Developers dig the polished UI with debug modes, progress bars for issues, and cooldowns to avoid nagging repeats, plus easy presets that scale thresholds by body size. It's a quick win for prototyping AI-powered personal trainers without vendor lock-in.

Who should use this?

Python devs experimenting with computer vision for fitness apps, like building AI-powered posture coaches or video analyzers. Home workout coders perfecting planks during desk breaks, or indie hackers forking it into custom AI-powered apps with Claude integration. Skip if you're not on macOS or lack a side camera setup.

Verdict

Worth forking for AI-powered plank posture tweaks—solid docs and user features punch above 44 stars—but 1.0% credibility signals early maturity with no tests or broad adoption. Prototype with it, but stabilize before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.