The Open Source Ward Weekly (19 August 2025)

Kiss Translator - A simple bilingual translation extension

I'm currently planning a trip to Japan, and this trending tool has been more than handy. It's a simple, open-source translation tool. I personally use the Firefox extension, and it works perfectly! The original repository is in Chinese, but I'm pretty sure you'll find a translator to help you install it.
Tinyauth - An easy way to secure your apps with a login screen

Tinyauth is a simple authentication middleware that adds a login screen or OAuth with Google, GitHub, and other providers to all of your Docker apps. It supports popular proxies like Traefik, Nginx, and Caddy.
It's ultra-simple to use and works perfectly.
I'm currently using it for a project that will be released soon (trust), and it has saved me a lot of otherwise very boring work.
Open Lovable - An open source... Lovable by the team behind... Firecrawl
Open Lovable is a copycat of Lovable, developed and maintained in an open-source manner by Mendable AI (the company behind Firecrawl). It allows developers to clone and recreate any website as a modern React application. Open Lovable can scrape and analyze websites thanks to its Firecrawl integration.
While this project essentially acts as a proxy to use Firecrawl API credits, the final product is nonetheless interesting. If you want to quickly test an idea or product, you can use Open Lovable to copy an existing website and slightly adapt it to your own content.
LangExtract - A Python library for extracting structured information from unstructured text
Maintained by Google, LangExtract is a Python library that uses LLMs to extract structured information from unstructured text documents based on user-defined instructions.
It offers a solution to one of the most challenging problems for researchers, analysts, and developers: converting complex textual data into a structured format. LangExtract is particularly strong at maintaining alignment between the structured output and the original documents, and it is capable of efficiently processing long documents.
GPT OSS - Open Source LLMs by OpenAI
OpenAI just open-sourced two of its models, making it one of the trendiest repositories at the moment. I haven’t tried running them locally myself, but I’m sure many people have already done so and shared their experiences online.
Still, I’m sharing my thoughts on this curious move from OpenAI here.
In the news

I recently stumbled upon Dries Buytaert’s blog, and I really loved his content, so I’m happy to share it with you. In this article, he discusses the value of open source in today’s software economy and how allocating even a fraction of national budgets to open source could help support an ecosystem that serves as the true backbone of our digital society.

Wikipedia has lost its legal challenge against the UK’s Online Safety Act rules, which could require the platform to verify user identities. The Wikimedia Foundation argued the regulations were too broad and could threaten editors’ privacy and safety, but the High Court rejected these claims.

You made it to the end of this article congrats ! Here is a teaser regards to an incoming long article. We have seen impressive attacks (tease tease) against the open source supply chain leading to trust issues. This is a serious issues to tackle and big players are proactive at it.
In a previous newsletter I shared Google iniative on the topic; here’s GitHub’s take.
Member discussion