Activity

nubprogrammer

Today I laid the foundation for AxonNexus, a modular AI API gateway designed to abstract multiple LLM providers behind a single, clean interface.

I implemented a provider-agnostic routing layer, allowing models from different backends to be accessed using unified Axon-branded aliases. The system supports OpenAI-compatible chat completions, centralized authentication, and graceful error handling.

A key focus was scalability: adding new models or providers now requires minimal changes, without touching the public API. I also verified model discovery through a /v1/models endpoint and tested invalid model handling with fuzzy matching.

This establishes a strong base for future expansion, including dashboards, usage tracking, and additional providers.

Attachment
Attachment
0
nubprogrammer

🛠️ Devlog #2 — Giving LagLore a Face (2h 05m)
Today’s session was all about bringing LagLore to life visually.
After locking in the core concept — Mr. Laggy testing players through unstable trials — I spent time translating that idea into actual game assets. This meant designing a full pixel-art character set for Mr. Laggy, covering multiple emotions and states so he can feel present across every trial rather than just being a static mascot.

Alongside the character work, I designed sprite assets for each individual trial — UI elements, game-specific components, and meta progression icons like ranks, trophies, and trial states. The goal was consistency: every game should feel like part of the same universe, not disconnected demos stitched together.

One challenge was designing assets that feel dynamic even before animation. Since these are browser-based games, the sprites needed to work well with lightweight animation techniques like frame swapping and CSS transforms, without relying on heavy engines.

By the end of the session, LagLore finally felt like a real game world rather than a concept. With the full visual foundation in place, the next step is implementing the first playable trial and wiring progression and leaderboards on top of these assets.

Attachment
Attachment
Attachment
Attachment
Attachment
Attachment
Attachment
Attachment
0
nubprogrammer

Devlog — LagLore (Ideation & Setup)

Spent this session laying the foundation for LagLore, a non-static web-based strategy game built around asynchronous gameplay and delayed outcomes. The core idea is to make time itself a mechanic — players don’t need to be online simultaneously, and decisions unfold over real-world time rather than instant reactions.

The focus today was ideation and structure rather than heavy feature work. I finalized the project concept, defined the scope to keep it manageable, and planned the architecture so future development doesn’t turn into a refactor nightmare.

On the technical side, I set up a lightweight web stack:

Frontend: HTML, CSS, and JavaScript (no heavy frameworks yet)

Hosting: Vercel for fast iteration and zero-config deployments

Version Control: GitHub for clean history and incremental shipping

I also created an initial “Coming Soon” landing page to establish a live demo endpoint early. This helps with visibility, quick feedback, and makes sure deployments are working before the game logic grows more complex.

Structurally, the project is designed to scale in phases:

Static landing & concept validation

Core game loop (client-side)

Persistence layer (lightweight backend / database)

Multiplayer & asynchronous mechanics

What I’m most proud of here is resisting the urge to overbuild early. Keeping the setup simple makes it easier to iterate fast, avoid burnout, and actually ship features instead of fighting tooling.

Next up: implementing the first interactive gameplay prototype and visual identity for LagLore.

Attachment
Attachment
0
nubprogrammer

Shipped this project!

Hours: 10.91
Cookies: 🍪 141
Multiplier: 12.89 cookies/hr

I built TermFlow, a minimalist terminal productivity suite that brings a live dashboard, todos, Pomodoro timer, and contextual info directly into the CLI — no browser, no mouse, just keyboard-first flow.

The hardest part wasn’t building the UI, it was shipping it properly. Python packaging, PyPI rules, OS-level restrictions, and broken installs forced me to really understand how production-ready CLIs are structured and distributed. I broke things more times than I can count before it finally worked everywhere.

I’m proud that TermFlow is now a real, installable tool, not just “code that runs on my machine”. It’s clean, fast, and genuinely useful — and the pain taught me more than the easy wins ever could. try it by downloading it by
pip install termflow-cli

nubprogrammer

Website: https://termflowinfo.vercel.app

This devlog is less “polished startup update” and more what actually went down.

🧠 The Idea

TermFlow started as a simple idea:
a keyboard-first terminal dashboard that doesn’t feel like 2009 Linux suffering.

No mouse. No browser. Just vibes and productivity.

💀 The Pain (Very Real)

termflow: command not found after thinking it was installed

Kali Linux pulling the classic
“externally-managed-environment 🤓☝️”

pip install ❌

apt install ❌

Me staring at the terminal at 2 AM like it owed me money

PEP 668 hit harder than expected.
Learned the hard way that pipx is the only sane path on modern distros.

🔧 What Actually Got Fixed

Removed Replit junk & unused files

Proper Python package structure (finally)

Static files moved to termflow/data so installs don’t explode

Added built-in Help and Info screens (no browser needed)

Introduced:

styles.tcss for UI theming

config.toml for future customization

Basically: from “it runs on my machine”
to “okay this is a real CLI now”.

📦 Installation Reality Check

This is the correct way (don’t fight your OS):

pipx install termflow-cli

If you try pip install system-wide on Kali or Arch —
yeah… you already know how that ends.

🌐 Website Update
The site is now cleaner and actually explains what TermFlow is instead of just vibes:
👉 https://termflowinfo.vercel.app

🛣️ What’s Next
More keyboard workflows
Config overrides
Modular features (no bloat)
Less pain (hopefully)

🏁 Final Thoughts
This release wasn’t about features —
it was about surviving packaging hell and coming out with a usable tool.
TermFlow is now:
✅ Installable
✅ Structured
✅ Not lying about working
More chaos soon.

Attachment
Attachment
Attachment
Attachment
0
nubprogrammer

Devlog — 1h 36m [STRUGGLING WITH ERRORS ;-;]

Spent this session mostly battling stability issues rather than adding new features. A few internal state mismatches and rendering bugs caused panels to disappear or behave inconsistently, especially around todos and Flow Mode transitions. Fixed multiple crashes related to corrupted todo data, standardized the completion state handling, and cleaned up the toggle logic so interactions no longer break the UI.

A lot of time went into restoring things that should have been simple — clock rendering, reflections, and panel layout — but this was necessary to get TermFlow back to a stable, predictable state. The app now launches cleanly again and core interactions behave as expected.

With the foundation stabilizing, the next direction is clearer. After fixing these issues, I’m considering adding a Flow Buddy — a calm, optional visual presence during Flow Mode — to enhance focus without breaking the minimalist philosophy. More experimentation coming next session.

Attachment
0
nubprogrammer

Devlog — 3h 37m

Spent a solid session improving the overall look and feel of the project. Focused heavily on UI/UX polish—cleaned up layouts, spacing, and visual hierarchy so everything feels more modern and intuitive. The app now looks much more attractive and usable compared to earlier builds.

The core system is up and running, with the timer actively tracking work time, confirming that the project is correctly detected and logging progress. Main functionality is stable, and the flow between components feels smoother now.

That said, there are still a few remaining issues in the to-do list. Some features need refinement, and a couple of bugs/edge cases are still being worked on. These aren’t blockers, but they do need to be resolved to make the project fully reliable and production-ready.

Overall, this session was about refinement and stability rather than adding flashy new features. The foundation is strong now—next steps will be fixing the remaining errors and tightening everything up.

Attachment
Attachment
0
nubprogrammer

Devlog #2 — UI groundwork & structure

⏱ Time spent: 2h 2m

In this session, I focused on laying down the core UI and internal structure for TermFlow.

What I worked on

Built the initial terminal UI in Python

Structured the project files cleanly so future features are easy to add

Designed the layout for:

Todo list panel

Pomodoro section

Info panel (weather + quotes)

Integrated a quote system (currently static/random, will refine later)

Current state

The UI renders correctly and is stable

Navigation and interaction are not fully wired yet

This session was mainly about visual structure and foundations, not behavior

What’s next

Make the UI fully interactive (keyboard-first)

Fix layout edge cases (help/info panels)

Add a command palette for quick actions (planning phase)

Improve data handling and persistence

This project is still actively evolving. I’ll be iterating on interactivity and polish in the upcoming sessions.

Stay tuned for more devlogs 👀
Follow along as TermFlow grows into a full terminal-first productivity tool.

Attachment
Attachment
1

Comments

developergirishgiri

bro nice plz give me vote also

nubprogrammer

Shipped this project!

Hours: 0.31
Cookies: 🍪 2
Multiplier: 7.51 cookies/hr

The bot code was giving error so now i am using diff libraries instead of shokaku so i have updated the repo with the latest files and the bot is now hosted 24/7 so pls dont reject my ship now pls
also added prefix command , slash commands were being unsupported during hosting in vps
ALSO REMEMBER THE COMMANDS ARE WOKRING PERFECTLY FINE IF U GET A DROPDOWN INTERACTION ERROR THAT IS DUE TO THE HOSTING ERROR OR API ONE BUT INORDER TO USE COMMANDS JUST CHECK OUT THE COMMANDS IN THE WEBSITE AND JUS USE THEM IN UR VC!
web link - https://synapsebeats.is-a.software/
bot invite link[available in website also] - https://discord.com/oauth2/authorize?client_id=1186998884111765534&permissions=8&scope=bot%20applications.commands

nubprogrammer

The bot code was giving error so now i am using diff libraries instead of shokaku so i have updated the repo with the latest files and the bot is now hosted 24/7 so pls dont reject my ship now pls
also added prefix command , slash commands were being unsupported during hosting in vps
ALSO REMEMBER THE COMMANDS ARE WOKRING PERFECTLY FINE IF U GET A DROPDOWN INTERACTION ERROR THAT IS DUE TO THE HOSTING ERROR OR API ONE BUT INORDER TO USE COMMANDS JUST CHECK OUT THE COMMANDS IN THE WEBSITE AND JUS USE THEM IN UR VC!
web link - https://synapsebeats.is-a.software/
bot invite link[available in website also] - https://discord.com/oauth2/authorize?client_id=1186998884111765534&permissions=8&scope=bot%20applications.commands

Attachment
Attachment
0
nubprogrammer

Today I initialized a new project called TermFlow, a terminal-first productivity app focused on minimalism and keyboard-driven workflows.

I set up the base project structure for the CLI application, experimented with layout ideas in the terminal UI, and explored how different modules might fit together (tasks, timers, and info panels).

Alongside the CLI work, I also created a small in-progress landing page for TermFlow. The website is currently a placeholder and will later showcase features, usage, and documentation once the core app stabilizes.

Most of this session was spent experimenting, restructuring folders, and testing ideas rather than finalizing features.

Next steps:

Fix UI layout issues (especially help/info panels)

Improve stability and keyboard navigation

Start polishing TermFlow into a more cohesive experience
basically bringing it to life ig LETS SEE YAL IN 2 OR 3 HOURS

Attachment
Attachment
0
nubprogrammer

🚀 Devlog 4 — Shipped! Bot Demo + Website Live

Yooooo, we’re officially shipped! 🛳️✨

🤖 Bot Status

The bot is fully functional and ready to roll

Added a cool start interface (you know, that terminal ASCII text that looks clean and pro 😎)

Commands are structured and working (src/commands + config fully set up)

Configurations are done, so anyone can clone + run it themselves

🌐 Website Deployment

Built a website for SynapseBeats

Features section + info about the bot

Deployed it to a subdomain (is-a.software) — so it’s live for everyone

Screenshots of the website included (video was a hassle lol, but trust me it’s fully working)

📸 Demo

Posted screenshots of:

the bot running in terminal

start interface with ASCII text

website live on the subdomain

Everything is replicable from the GitHub repo

🔗 Try It Yourself

Repo link: https://github.com/Nub-programmer/SynapsebeatsSRC

Clone, configure, and run the bot in your own server

Website + commands ready to explore

🎯 Final Thoughts

Folder structure organized ✅

Commands folder + config completed ✅

Website deployed ✅

Demo shared (screenshots) ✅

This is basically the final product, ready for anyone to test or extend. Go check the GitHub repo if you wanna grab the code and implement it in your own bot!

Attachment
Attachment
Attachment
Attachment
Attachment
0
nubprogrammer

🛠️ Devlog #3 — Website Build & Deployment

This devlog focuses on the webside of SynapseBeats and preparing the bot for launch.

🌐 Website Progress

I designed and built a dedicated website for SynapseBeats to make the project feel like a real product instead of just a Discord bot.

The website includes:

An overview of what SynapseBeats does

Core features and music capabilities

A command preview section

Setup / invite flow

Clean landing design for users discovering the bot

After building it, I deployed the site and connected it to a custom subdomain:

👉 https://synapsebeats.is-a.software

This was my first time fully setting up a bot website + subdomain flow, and it helped me understand deployment, DNS, and hosting better.

🤖 Bot Status

The bot itself is almost complete.

Current state:

Core music system implemented

Slash command support in place

Commands folder structured under src/commands

Config system completed under src/config

Filters like bassboost, nightcore, and 8D added

The focus now is polishing and final testing before officially shipping.

Attachment
0
nubprogrammer

🍳 Devlog #2 — Organizing the Chaos & Giving the Bot a Face

Time logged: ~3 hours
Project: SynapseBeats

okay so today was a “clean your room before building more stuff” type of day.

🧠 What I worked on

I focused on properly structuring the bot so future development doesn’t turn into spaghetti code:

Inside src/commands, I completed the config/ commands
(as seen in the repo structure 👀)

Commands are now clearly separated by purpose, making:

slash commands easier to manage

permissions & cooldowns more consistent

This sets a solid base for scaling the bot without breaking everything later (future me will be grateful).

🎨 Branding moment

Designed a SynapseBeats logo using AI to match the music + tech aesthetic.

Updated the Discord Developer Portal:

App name

Icon

Description

The bot finally looks like an actual product and not a random test app 💀

🌐 What I’m doing now

I’ve started working on a website for SynapseBeats
This will act as:

a landing page

feature overview

easy way for users to understand the bot

This should make onboarding way smoother than “bro just invite it trust me”.

🚀 What’s next

Continue building the website

Improve command UX and music features

Proper documentation so users (and judges 👀) actually know what’s going on

This devlog was all about foundations: structure, clarity, and branding.
Now that the base is solid, I can actually cook 🔥

Attachment
Attachment
0
nubprogrammer

Day 1 — SynapseBeats

Today I set up the SynapseBeats project for public tracking and started the Flavortown submission.

What I did:

  • Configured Hackatime/WakaTime and verified heartbeats from VS Code.
  • Linked my GitHub repo (https://github.com/Nub-programmer/SynapsebeatsSRC) to Hackatime.
  • Cleaned up project root structure and added a .wakatime-project to ensure correct tracking.
  • Reviewed the codebase (src/ folder): command handlers, events, plugin structure, and Lavalink connection logic.

Next steps:

  • Improve README, add screenshots and the banner.
  • Small refactors to command handling and stricter error handling for Lavalink reconnects.
  • Prepare a short demo video for the Flavortown page and the upcoming YouTube source release.

Notes:
Hackatime tracking is active (verified heartbeats). Will post short updates as I progress.

Attachment
0