DEV Community: mabona011145629

🌸 Petal – A Daily Companion for Growth, Gratitude & Goodness

mabona011145629 — Mon, 13 Jul 2026 02:06:42 +0000

This is a submission for Weekend Challenge: Passion Edition

What I Built

Petal is a daily journal and self‑growth companion that helps people track the good they do, reflect on their challenges, and stay motivated to become better versions of themselves.

The app allows users to:

Log something good they did each day
Log something they want to improve (a lesson learned)
Make a promise to themselves and receive a gentle reminder to keep it
Receive personalized encouragement from a pink, talking doll named Petal
See their progress with stats, streaks, and a performance bar
Chat with an AI assistant that offers wisdom, comfort, and motivation

Petal is not just an app — it is a reflection of who I am. I built it because I believe the world needs more people doing more good than bad — for others, for themselves, and for the environment. I love humanity, and I love helping others. That is what makes me happy. I wish to people doing good to another more than doing bad
**

Demo login email: maxwellmabona581@gmail.com, password: Mabona

🌐 Live Demo: https://mabona.firstsuninvestment.com/devweekendchallenge

📱 Screenshots:

Video Walkthrough: Watch Petal in action

Code

🐙 GitHub Repository: https://github.com/mabona011145629/devweekendchallenge

The repository contains all source files, including:

index.php — Login and registration
dashboard.php — Main user dashboard
results.php — Response page with the talking doll
github_ai.php — AI chat integration with database fallback
petal_responses.sql — The complete database of 180+ encouraging responses
PWA manifest and service worker for offline support

How I Built It

I built Petal with a mobile-first, human-centred approach. The goal was not just to create a tool, but to create a companion — something that feels alive, warm, and genuinely caring.

Technical Stack:

Frontend: HTML, CSS (with glass‑morphism and smooth animations), JavaScript
Backend: PHP (vanilla, no frameworks)
Database: MySQL (hosted on cPanel)
AI Integration: GitHub Models API with a smart database fallback
Text‑to‑Speech: Web Speech API for the talking doll
Voice Input: Web Speech Recognition API
PWA: Service worker + manifest for offline and home‑screen installation

Key Features & Decisions:

The Talking Doll – I designed a pink, animated SVG character that speaks the user's responses using the Web Speech API. The doll's mouth moves when it talks, and it has a speech bubble that appears above its head. This makes the experience feel like a real conversation.
Smart Response System – Instead of relying purely on an AI API, I built a database of 180+ pre‑written responses (categorised by keywords, tone, and sentiment). If the AI is unavailable, the app falls back to these responses seamlessly. The system also detects user intent — whether they are greeting, thanking, sharing joy, or expressing sadness — and responds appropriately.
Promise Tracker – Users can make a promise to themselves. The app reminds them after a set time (configurable) to check in and complete their promise. This builds accountability and self‑discipline.
Progress Visualisation – The dashboard shows stats, streaks, and a performance bar that visualises the ratio of "good" entries to "challenges." This gives users a clear picture of their growth over time.
Mobile‑First & PWA – The app is fully responsive and works as a Progressive Web App. Users can install it on their phone's home screen and use it offline.
Voice Input – Users can speak their entries instead of typing, making the app more accessible and natural to use.

Prize Categories

I am not submitting to any specific prize category (Snowflake, Solana, ElevenLabs, or Google AI). My focus was on building a meaningful, working product that solves a real human need — encouraging goodness, self‑reflection, and growth.

One Final Thought

I started programming because I wanted to help my mother. I keep programming because I want to help the world.

Petal is my way of reminding people that they are worthy, they are enough, and they are loved. It is a gentle nudge to do more good than bad — not just for others, but for themselves.

I love humanity. I love helping others. That makes me happy.

And if Petal helps even one person feel a little better about themselves, then every line of code I wrote was worth it.

🌱 Be kind. Do good. Grow every day.

— Mabona Maxwell Sihlongonyane

📧 maxwellmabona581@gmail.com

🐙 GitHub

🌐 Portfolio

About the Author

I am a software developer, network engineer, and robotics teacher from Eswatini volunteering. I am passionate about building technology that empowers people and communities. This project is a reflection of my deepest values: goodness, kindness, and growth.

GitHub Finish-Up-A-Thon Challenge

mabona011145629 — Sat, 06 Jun 2026 21:25:40 +0000

This is a submission for the GitHub Finish-Up-A-Thon Challenge

What I Built

Soulmate is not a dating app. It is a digital love platform.

I built Soulmate because I believe love is physical and emotional — and now we have found a digital place for love too. The app connects people through normal chats for lovers and friends, and therapy chats with Love Doctors — licensed relationship helpers who can only be accessed with a PIN.

Key features:

Normal chat (text, voice notes, pictures, music, videos — no documents)
Love Doctor chat (PIN secured, screenshot prevention in Java)
Request Partner feature (one partner asks Love Doctor to invite the other into joint therapy)
Search and love requests
Post feeds with gender-based filtering (females see more male posts, males see more female posts, plus LGBTQ+ support)
Profile editing and hiding
"Keep me logged in"
Notifications and post management
Admob integration (test mode)
All gender identities supported — love is for everyone

My saying: Love is Physical. Love is Emotional. Now we have found a Digital Platform for Love.

I chose pink because love, in my mind, is pink or red. I believe people will love it.

Tech stack:Java (screenshot prevention), Firebase, Admob ads for monetization, PHP backend.

GitHub Repository: [github.com/mabona011145629/Soulmate-Love-App](https://github.com/mabona011145629/Soulmate-Love-App)

Demo

Screenshots:

Video Walkthrough: [https://mabona.firstsuninvestment.com/soulmate/screenshots/soulmate-walkthrough-video.mp4](https://mabona.firstsuninvestment.com/soulmate/screenshots/soulmate-walkthrough-video.mp4)

Live Download (APK): mabona.firstsuninvestment.com/soulmate/app-release

*Landing Page with more details: *

## The Comeback Story

Soulmate started as an idea in my head: a love platform for everyone, regardless of gender. I saw a great need to bring love doctors in a mobile device where everyone can reach without transport costs and other general added costs. People want talk and get advices but the issue is the normal methods to get love doctors all are not accessed so easily here just a type by your hand love doctor sees you get help.

Before this challenge:

The app was functional but missing privacy features
Love Doctor chats existed but without proper security
No screenshot prevention
No "Request Partner" feature

What I changed, fixed, or added:

Added PIN security for Love Doctor chats
Implemented screenshot prevention using Java's FLAG_SECURE
Built the Request Partner feature (Love Doctor can invite both partners into a joint chat)
Added voice notes to all chats
Created post filtering by gender preference
Added "Keep me logged in" for better UX
Integrated Admob (test ads currently)

The biggest challenge was the screenshot prevention. I am a Flutter developer primarily, but I had to write lower-level Java code to block screenshots in the Love Doctor chat. I learned it, broke it, fixed it, and finally got it working.

My biggest dream? That someone in Eswatini — or anywhere in Africa — finds their soulmate or saves their relationship through this app.

## My Experience with GitHub Copilot

GitHub Copilot helped me write the Java code for screenshot prevention. I am comfortable with Flutter, but Android's WindowManager and FLAG_SECURE were new to me. Copilot suggested the correct implementation and helped me debug the window manager calls.

Copilot also helped with:

Firebase notification setup
Post filtering logic
Voice notes recording and playback
PHP backend endpoints for the Request Partner feature

Copilot didn't build the app for me — but it made me faster, especially when jumping between Java, and PHP. It felt like having a senior developer looking over my shoulder, suggesting the next line.

Love is a gift from above. And love is for everyone.

— Mabona Maxwell Sihlongonyane, Eswatini

GitHub Username: mabona011145629
This is a submission for the GitHub Finish-Up-A-Thon Challenge