NomadMind
https://nomadmind.in

https://redd.it/1n79dbc
@r_opensource

Automating doc updates from code changes (CLI)

TL;DR: We open-sourced **Cocode**, a python CLI that turns 1-hour doc updates into a \~30s command. GitHub: [https://github.com/pipelex/cocode](https://github.com/pipelex/cocode)

You know the drill - you make a small change in your codebase, then spend your entire afternoon hunting through documentation files, updating examples, and writing changelog entries for what should have taken 10 minutes total.

So we built and open-sourced Cocode, a CLI tool that uses Pipelex AI workflows to automate the tedious parts:

* `cocode swe-from-repo-diff write_changelog v0.8.0` → Generates complete changelog from your git diff
* `cocode swe-doc-update v1.0.0` → Proposes docs rewrites based on code changes
* `cocode swe-doc-proofread --doc-dir docs` → Finds every existing mismatch between your docs and your codebase, as well as typos

**What it solves:** The 4 hours you spend manually cross-referencing code changes with existing docs, writing changelog entries from git diff, and proofreading for inconsistencies.

**Real talk:** Last week I updated a single function parameter. Coding took 10 minutes. Manual doc updates took 1 hour. With Cocode, the whole thing would be done in under 30 seconds.

**How it works**: I use Pipelex technology to scaffold LLM Pipelines. It enables me to split the documentation into smaller sections, process it in parallel, batching, and apply specific LLM prompts.

Been testing it for a few weeks and it's saved me probably 5-6 hours already. Curious what pain points others have with documentation workflows?

Its still in the early days of cocode, but feel free to help us make it better.

demo: [https://youtu.be/T56MOkoZwm8?si=z1zlampMXQaZj1rF](https://youtu.be/T56MOkoZwm8?si=z1zlampMXQaZj1rF)

repo: [https://github.com/pipelex/cocode](https://github.com/pipelex/cocode)



https://redd.it/1n7aqqt
@r_opensource

Looking for a YouTube watch together where i can force full screen and still remote control

I'm working on a vtuber setup in warduo where i wanna do watch together so me and friend are watching the video but 3rd user is what audiance sees

Warduo screen is a non interactive screen so if you use a browser website you get full page but unable to interact or scroll (is mainly for chat which eh)

I found open source watch party sites which is good but I can't just say insert full screen link to screen then i press play on my actual browser

Anyone got any ideas of making this work with Any open software

https://redd.it/1n7chy8
@r_opensource

I need to setup a family calendar/task list on a touch-screen monitor in my kitchen...

How would you suggest I approach this? From an app standpoint I can vibe-code something in no time -- that's not hard since most of the data will be pulled from Google but what are my options when it comes to getting a "blank" reasonably-priced touchscreen monitor?

I'm thinking I have two options:

1/ I can create a web app and open it up in a browser on the monitor

pros:

a/ easy & fast to develop the app
b/ easier to update the app when needed

cons:

a/ user interactions (clicking, navigation) might be clunky in the browser via touch-screen
b/ keeping the screen on all the time (which I want) is harder

2/ Create an android or iOS app

pros:

a/ user experience is much more configurable
b/ easier to manage the ecosystem (keeping the screen on, etc)

cons:

a/ harder to update app
b/ harder to develop


Am I overthinking this? Is there an easier option? I know there's a bunch of pre-paid solutions out there but they start at $600 and have a monthly fee which I want to avoid.

Thanks!

https://redd.it/1n7h8o8
@r_opensource

microfolio - Static Portfolio Generator / free & open-source
https://microfolio.net/

https://redd.it/1n7ibj4
@r_opensource

Easy way to manage/organize your code projects: archivador CLI.

I notice that every day I repeat the same commands to change projects, set up services for work, and launch the code editor (obviously nvim, haha). So I created a simple tool to have an easy way to switch between projects and start coding, and maybe it can help you too. I’m sharing the repo here; it’s written in Rust. As I said, it’s a simple tool, but it helps me organize my code projects and prevents me from repeating many commands (it also remembers project paths).

https://codeberg.org/a-chacon/archivador

https://redd.it/1n7k2km
@r_opensource

This is Vercel for backend
https://github.com/MotiaDev/motia

https://redd.it/1n7kziz
@r_opensource

Aralez: An OpenSource reverse proxy on Rust and Cloudflare's Pingora

Some time ago I have created a project [**Aralez**](https://github.com/sadoyan/aralez) **.** It's a complete reverse proxy, ingress controller implementation on top of Cloudflare's [**Pingora**](https://github.com/cloudflare/pingora)

Now I'm happy to announce about the completion of another major milestone, **Aralez** is also an ingress controller for **Kubernetes** now..

What we have:

* Dynamic load of upstreams file without reload.
* Dynamic load of SSL certificates, without reload.
* Api for pushing config files, applies immediately.
* Integration with API of Hashicorp's Consul API.
* Kubernetes ingress controller.
* Static files deliver.
* Optional Authentication.
* Pingora at heart, with crazy performance .
* and more .....

Here in [**GitHUB**](https://sadoyan.github.io/aralez-docs/) pages is the full documentation .

Please use it carelessly and let me know your thoughts :-)

https://redd.it/1n7k32t
@r_opensource

Software for taking study notes

Hello!

Lately, I've only been using a physical notebook to take notes in classes, document ideas, and etc. And I really wanted some free and open-source software that could help me with this. Sort of "simulating" (?) this type of physical note-taking, if such a thing exists.

Thank you!

https://redd.it/1n7qw0c
@r_opensource

QueryWeaver - Text2SQL using graph-powered schema understanding.
https://github.com/FalkorDB/QueryWeaver/

https://redd.it/1n7s8bu
@r_opensource

(Android) CuteMusic went Expressive!

Hey folks! a year and a half ago, I released CuteMusic, a feature rich, beautiful and open-source offline music player app for Android. Today I released v3.0.0, with a fresh new design based off Material 3 Expressive, if you were looking for a M3E music player, then CuteMusic may be your new love :)

You can check it here: https://github.com/sosauce/CuteMusic

Thank you so much to everyone who contributed in a way or an another to CuteMusic's growth, y'all make me enjoy my passion even more ❤️!!!

Until next time we connect 😉

https://redd.it/1n7u74s
@r_opensource

I over-engineered a URL shortener for insane speed. Here’s the code.
https://github.com/killcod3/shorty

https://redd.it/1n819uj
@r_opensource

GitHub - Burnsedia/waywind: A CLI tool that generates Waybar themes from your TailwindCSS and DaisyUI config. Perfect for Hyprland users who want a consistent, riced desktop.
https://github.com/Burnsedia/waywind

https://redd.it/1n81vkm
@r_opensource

colorrs — a faster, cross-platform, feature-rich Rust alternative to shell-color-noscripts
https://github.com/emmalexandria/colorrs

https://redd.it/1n81v2h
@r_opensource

Has anyone worked on detecting fake job postings? Looking for references

I’m exploring an idea to tackle fake job ads by cross-verifying postings with official company sites (extract company → check careers page → confirm if the job exists).

Before I dive in, I’d like to know:

Has anyone seen similar research, startups, or tools?
Any references, datasets, or prior work I should look into?

Thanks for any pointers 🙏

https://redd.it/1n82gya
@r_opensource

Love Bruno API client but hate Electron apps? Trayce is a Bruno-interoperable client.

Hi all, I would like to share an app I have been developing called Trayce. It is born out of my frustration with existing API clients which seem to all be based on Electron or some kind of browser-rendered GUI.

I really like the way Bruno lets you save requests in git-friendly files, so I decided not to re-invent the wheel and made Trayce use exactly the same file format as Bruno. This means you can open Bruno collections with Trayce, modify them, and they will still work with the Bruno client.

On top of that it lets you monitor Docker network traffic, including TLS-encrypted traffic, without the need for a proxy or custom CA certificates.

Any feedback would be much appreciated, I would especially like to know if there are any features you would like to see added. Thanks!


https://trayce.dev/

https://redd.it/1n84ek8
@r_opensource

IndiaExams Database - Practice contributing to open source

Hello everyone :>

Remember how sometimes it feels like there are a million entrance exams in India, and you’re never quite sure which ones actually fit your background or interests? Well, the IndiaExams Database is aiming to solve that by crowdsourcing detailed info on all the exams you’ve probably never heard of (but might want to know about).

Right now, there are over 200 exams waiting to be filled in with details like eligibility, dates, and official links. And here’s the exciting part: since contributions doesn't require coding, we can practise making pr's and open source contribution through this and we’ll fill up this resource really fast too— making life easier for thousands of students to come.

How to contribute (it’s really simple!):

1. Go to the Issues tab in the project GitHub repo and choose an exam you want to work on.
2. Each issue corresponds to a text-only .yml file — open it to see what info is needed and do some research online. Official websites, notifications, and PDFs are the best sources.
3. Fill out the blanks in the file with verified info.
4. Create a Pull Request (PR) on GitHub and mention the corresponding issue number.

That’s it! No complicated tasks — just helping add info exam by exam. Plus, if you prefer later, you can help by verifying info others added, too.

Whether you’re looking to boost your GitHub profile, contribute to an impactful project, or just learn about all the exams out there, this is a friendly, collaborative way to help your fellow students out.

Shoot me a message or comment below if you have any doubts or suggestions
Let’s make this project really shine together!

https://redd.it/1n8499s
@r_opensource

Looking for realtime speech to text that works offline in a smartphone

I was looking at realtime speech to text solutions that work offline & can run on a smartphone.

I stumbled upon Google's Live transcibe which worked flawlessly even when I turned off my internet on my phone. Way better than Samsung galaxy voice trannoscription which I guess needs internet as well.

Google claims to have open sourced the tech but they don't mention how their offline models work.
https://github.com/google/live-transcribe-speech-engine

Does anyone has any idea how can I access the live transribe API offline? I wanted to build a audio note taking app based on it.

https://redd.it/1n86wyj
@r_opensource

YAMLResume updates: section customization and dev mode
https://www.youtube.com/watch?v=xytlzedEQ_w&t=6s

https://redd.it/1n82oy3
@r_opensource

How do you keep a private version of your repo that is sync with the public one ?

Hello, I want to make an open source project but I'd like to have only the source code in the public repo, and have a private repo that would contains workflow, secrets and other things to deploy it on my own server
In github or even in general how would you achieve this ? Since you add workflow files to the private repo wouldn't you have conflict between the public and private repo ?

https://redd.it/1n8c66z
@r_opensource

My first solo OpenSource project! OutputBuddy, a CLI tool for redirecting command output intelligently

https://github.com/zmunro/outputbuddy

Often times I find myself wanting to run commands in the terminal that are going to be running for a long time, and I want to look at the logs later. One way of solving this is to pipe stdout and stderr to a file (if you remember how to do that) and then tail -f that file while the command is running to see what it says. My grievances with this process was that it took up time, I had to either choose to see loading bars and progress meters and have them clutter up the logs or to forego seeing progress bars entirely by disabling them.

OutputBuddy allows you to easily redirect stdout and stderr to files and/or the terminal while stripping the the ANSI characters for terminal colors and loading bars/spinners when writing to the files.

The minimum you have to do to use it is:
ob -- <your-command-here>

The above command will redirect stdout and stderr to a file called buddy.log in the current directory, and also write stdout and stderr to a file. You can override this default behavior using by doing things like the following:

# Custom logging: redirect both to a specific file AND show on terminal
ob stderr+stdout=output.log stderr+stdout -- python noscript.py

# Or use the shorthand of 2 and 1 for stderr and stdout respectively
ob 2+1=output.log 2+1 -- python noscript.py

# Separate stdout and stderr to different files
ob 1=out.log 2=err.log -- make

# Only log errors, but still show them on screen
ob 2=errors.log 2 -- ./my-program

https://redd.it/1n8eerh
@r_opensource