Are people farming contributions with AI-generated PRs?
I've been contributing to Open Source for about a year now. I started out by translating docs into my native language, but over time I moved into broader contributions within the project and began climbing the membership ladder - something I'm really glad about.
Lately, though, I've noticed a strange pattern, especially when it comes to localization work:
* People request to work on issues in languages they clearly don't speak. In most cases, these accounts are brand new, often created within the last month.
* They insist on being assigned to the issue. Why? What's the deal with that assign?
* The resulting PR is usually AI-generated, from the denoscription down to the content. Guidelines are ignored, standards aren't followed, and it's pretty clear no real effort went into it.
It honestly feels like some kind of farming or grinding is going on, which makes me wonder: are people just doing this to inflate their GitHub profiles? Are some of these accounts not even real people?
https://redd.it/1ntelxq
@r_opensource
I've been contributing to Open Source for about a year now. I started out by translating docs into my native language, but over time I moved into broader contributions within the project and began climbing the membership ladder - something I'm really glad about.
Lately, though, I've noticed a strange pattern, especially when it comes to localization work:
* People request to work on issues in languages they clearly don't speak. In most cases, these accounts are brand new, often created within the last month.
* They insist on being assigned to the issue. Why? What's the deal with that assign?
* The resulting PR is usually AI-generated, from the denoscription down to the content. Guidelines are ignored, standards aren't followed, and it's pretty clear no real effort went into it.
It honestly feels like some kind of farming or grinding is going on, which makes me wonder: are people just doing this to inflate their GitHub profiles? Are some of these accounts not even real people?
https://redd.it/1ntelxq
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
Does anyone else use the outdated Lawnchair launcher?
There's just no way I'm alone on this, I must know if anyone else uses it too.
Version 1.2.0.1884 is the one I use, for example.
https://redd.it/1nths1d
@r_opensource
There's just no way I'm alone on this, I must know if anyone else uses it too.
Version 1.2.0.1884 is the one I use, for example.
https://redd.it/1nths1d
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
FFmate v2: open-source automation for FFmpeg with clustering support
We’ve been building FFmate for a while, and last week we released v2.0.
FFmate is an open-source automation layer for FFmpeg. It grew out of the need to replace fragile shell noscripts with something more reliable, cross-platform, API-first, and easier to configure.
With FFmate you get a job queue, REST API, watchfolders, presets, and webhooks. In v2 we added cluster support. Multiple instances can now share a Postgres queue, split tasks across nodes, and keep running even if one node fails.
I’m sharing this here because many of you have worked with FFmpeg in one form or another. You might find FFmate useful, or you may just have thoughts on how we approached it.
Either way, feedback and contributions are welcome.
Repo: https://github.com/welovemedia/ffmate
Docs: https://docs.ffmate.io
https://redd.it/1ntjwyo
@r_opensource
We’ve been building FFmate for a while, and last week we released v2.0.
FFmate is an open-source automation layer for FFmpeg. It grew out of the need to replace fragile shell noscripts with something more reliable, cross-platform, API-first, and easier to configure.
With FFmate you get a job queue, REST API, watchfolders, presets, and webhooks. In v2 we added cluster support. Multiple instances can now share a Postgres queue, split tasks across nodes, and keep running even if one node fails.
I’m sharing this here because many of you have worked with FFmpeg in one form or another. You might find FFmate useful, or you may just have thoughts on how we approached it.
Either way, feedback and contributions are welcome.
Repo: https://github.com/welovemedia/ffmate
Docs: https://docs.ffmate.io
https://redd.it/1ntjwyo
@r_opensource
FFmate
FFmate | FFmpeg Automation for the Modern Stack
Documentation for FFmate.io - FFmpeg Automation for the Modern Stack
Looking for AI/ML projects that need documentation help (especially beginner guides)
I've been using various open source AI tools for the past year and want to give back to the community. My coding skills are decent but not amazing, so I'm focusing on documentation and user experience contributions.
Recently helped improve docs for transformer lab and a few other projects. There's such a huge gap between "here's the github repo" and "here's how to actually use this thing" for most AI tools.
I'm particularly interested in projects that:
Need better getting started guides
Want help with example notebooks or tutorials
Could use clearer installation instructions
Need user-friendly explanations of technical concepts
My background is technical writing and I've been working with local AI setups, so I understand both the user experience side and the technical challenges.
If you maintain an AI/ML project and think documentation help would be valuable, drop a comment or send me a message. I'm looking to contribute to 2-3 projects consistently rather than doing one-off contributions.
Also happy to hear from other contributors about which projects you've found most welcoming to documentation improvements. The AI space moves so fast that docs often lag behind features.
https://redd.it/1ntlw0n
@r_opensource
I've been using various open source AI tools for the past year and want to give back to the community. My coding skills are decent but not amazing, so I'm focusing on documentation and user experience contributions.
Recently helped improve docs for transformer lab and a few other projects. There's such a huge gap between "here's the github repo" and "here's how to actually use this thing" for most AI tools.
I'm particularly interested in projects that:
Need better getting started guides
Want help with example notebooks or tutorials
Could use clearer installation instructions
Need user-friendly explanations of technical concepts
My background is technical writing and I've been working with local AI setups, so I understand both the user experience side and the technical challenges.
If you maintain an AI/ML project and think documentation help would be valuable, drop a comment or send me a message. I'm looking to contribute to 2-3 projects consistently rather than doing one-off contributions.
Also happy to hear from other contributors about which projects you've found most welcoming to documentation improvements. The AI space moves so fast that docs often lag behind features.
https://redd.it/1ntlw0n
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
[Open Source] Flask Backend for a “Shop of Shops” Marketplace
https://github.com/calinvladth/shops
https://redd.it/1ntqo3u
@r_opensource
https://github.com/calinvladth/shops
https://redd.it/1ntqo3u
@r_opensource
GitHub
GitHub - calinvladth/shops
Contribute to calinvladth/shops development by creating an account on GitHub.
How do you keep momentum alive in open-source projects with friends?
I’ve been hacking on an open-source idea with a friend. The initial energy is always super high, but keeping that momentum going over the long run is where it gets tricky.
What’s worked for you when it comes to keeping open-source projects alive (especially side projects)? Weekly syncs? Clear roadmaps? Or just letting it flow naturally?
Curious to hear what’s worked for other maintainers here 🙏
https://redd.it/1ntv4g2
@r_opensource
I’ve been hacking on an open-source idea with a friend. The initial energy is always super high, but keeping that momentum going over the long run is where it gets tricky.
What’s worked for you when it comes to keeping open-source projects alive (especially side projects)? Weekly syncs? Clear roadmaps? Or just letting it flow naturally?
Curious to hear what’s worked for other maintainers here 🙏
https://redd.it/1ntv4g2
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
New TilBuci version, a free software to create interactive content
TilBuci, a free, open source tool to create interactive content reaches version 15 with a cool new feature! The new version brings a set of new tools to simplify the creation of narratives. It is now possible to register characters and use the dialogue manager to create and display conversations among them in a style similar to that found in role-playing games and visual novels.
A tutorial on using this new feature can be found at
https://youtu.be/4BJe3NEp4s8
\-------
TilBuci is an interactive content creation tool focused on development for web, mobile and desktop apps. Distributed as free software under the MPL-2.0 license, it is presented in the form of a web program, executed from a browser with functionalities for collective creation, and also as a portable desktop software for various systems. The software repository can be found at
https://github.com/lucasjunqueira-var/tilbuci
https://redd.it/1nu00y6
@r_opensource
TilBuci, a free, open source tool to create interactive content reaches version 15 with a cool new feature! The new version brings a set of new tools to simplify the creation of narratives. It is now possible to register characters and use the dialogue manager to create and display conversations among them in a style similar to that found in role-playing games and visual novels.
A tutorial on using this new feature can be found at
https://youtu.be/4BJe3NEp4s8
\-------
TilBuci is an interactive content creation tool focused on development for web, mobile and desktop apps. Distributed as free software under the MPL-2.0 license, it is presented in the form of a web program, executed from a browser with functionalities for collective creation, and also as a portable desktop software for various systems. The software repository can be found at
https://github.com/lucasjunqueira-var/tilbuci
https://redd.it/1nu00y6
@r_opensource
YouTube
Discovering the dialogue creation tool
TilBuci features a tool that simplifies the creation of dialogue among characters, using a structure similar to the one found in traditional RPGs and visual novels. In this video, we'll learn a little about how this feature works and also learn techniques…
I made a program that converts all your images in your projects to webp and updates the code
I made a program that converts all your images in your projects to webp and updates the code
basically I got sick of having to copy and convert images using websites or tools I wanted something that I can just run in the terminal and it converts all images and updates the entire code with the new images, you can also mass delete the old images if you want that option, or exclude specific images.
it made most of the projects I worked on load faster I know most ppl hate webp but its very good for websites I assume most of you know already.
I will keep working on it the source code is here https://github.com/SeifEldin15/images-to-webp or you can install it and run it as an npm package
npm install -g webp-image-converter
webp-converter
if you think more features can be added or something can be changed lmk, I have no idea if there's an option out there that does this already, hope you guys find it useful
https://redd.it/1nu4v0g
@r_opensource
I made a program that converts all your images in your projects to webp and updates the code
basically I got sick of having to copy and convert images using websites or tools I wanted something that I can just run in the terminal and it converts all images and updates the entire code with the new images, you can also mass delete the old images if you want that option, or exclude specific images.
it made most of the projects I worked on load faster I know most ppl hate webp but its very good for websites I assume most of you know already.
I will keep working on it the source code is here https://github.com/SeifEldin15/images-to-webp or you can install it and run it as an npm package
npm install -g webp-image-converter
webp-converter
if you think more features can be added or something can be changed lmk, I have no idea if there's an option out there that does this already, hope you guys find it useful
https://redd.it/1nu4v0g
@r_opensource
GitHub
GitHub - SeifEldin15/images-to-webp: cli tool webp images converter javanoscript nodejs npm opensource
cli tool webp images converter javanoscript nodejs npm opensource - SeifEldin15/images-to-webp
Any Stable Linux Smartphone OS?
I just watched some reviews of Mobian and Ubuntu touch. As a user who has strong dislike for android, should I invest in having a "Linux" smartphone? I saw Mobian and Ubuntu touch are still unstable and lack features. Should I just install a full desktop Linux on a tab, and forget al about these? (Note: suggest only fully Open Source Linux smartphone OS, which has Open Source app development kit)
https://redd.it/1nu62p3
@r_opensource
I just watched some reviews of Mobian and Ubuntu touch. As a user who has strong dislike for android, should I invest in having a "Linux" smartphone? I saw Mobian and Ubuntu touch are still unstable and lack features. Should I just install a full desktop Linux on a tab, and forget al about these? (Note: suggest only fully Open Source Linux smartphone OS, which has Open Source app development kit)
https://redd.it/1nu62p3
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
I want to start contributing to open source to learn, where should I start? any recommendations?
I want to do it for fun and to help mostly but also I need to get something out of it which will be learning and maybe a better github account I just don't know where to start.
https://redd.it/1nu6db7
@r_opensource
I want to do it for fun and to help mostly but also I need to get something out of it which will be learning and maybe a better github account I just don't know where to start.
https://redd.it/1nu6db7
@r_opensource
Reddit
From the opensource community on Reddit
Explore this post and more from the opensource community
Just launched a new landing page for my open-source canvas tool 🚀
https://github.com/A-ryan-Kalra/react-scribble-pad
https://redd.it/1nubndm
@r_opensource
https://github.com/A-ryan-Kalra/react-scribble-pad
https://redd.it/1nubndm
@r_opensource
GitHub
GitHub - A-ryan-Kalra/react-scribble-pad: (npm package & chrome extension)- Transparent Overlay for Sketch, Text, and Stickers.
(npm package & chrome extension)- Transparent Overlay for Sketch, Text, and Stickers. - A-ryan-Kalra/react-scribble-pad
Access Claude Code features for free using multiple free models + centralized MCP hub — Chatspeed
Hi everyone 👋 I’m the creator of Chatspeed, an open-source AI proxy + desktop assistant.
# Why Chatspeed exists
For developers, it’s often hard to know whether a model supports tool calls — even the same open-source model may behave differently on different platforms. CCProxy solves this by enabling tool calls for any model: models can invoke tools seamlessly, regardless of their native support, lowering the mental overhead for developers.
Many AI models are either paid or limited in functionality. Claude Code is powerful but expensive. With CCProxy’s protocol conversion, tool compatibility mode, and prompt enhancement, developers can integrate free models from various platforms (e.g., Nvidia’s
Another common pain point is fragmented MCP tool management. Developers often use multiple AI IDEs or plugins, each with its own MCP installation, which is cumbersome to manage. With CCProxy, users can install MCP tools directly within the module, centralizing management and exposing a unified set of tools externally via SSE or Streamable HTTP. Built-in WebSearch and WebFetch tools further enhance other clients’ ability to perform tool calls and fetch/process information efficiently.
# Core module: CCProxy (Chat Completion Proxy)
CCProxy is more than API forwarding — it’s a fully-featured AI middleware:
Protocol conversion: Converts client requests (e.g., OpenAI-compatible) into the target model’s native protocol (Claude, Gemini, Ollama, etc.) and converts the model’s output back, enabling seamless communication across protocols.
Tool compatibility mode: Even models that don’t natively support tool calls can invoke tools through CCProxy.
Proxy groups + prompt management: Scenario-based configuration for different clients or workflows, with dynamic prompt replacement/enhancement.
Global load balancing: Multi-key, multi-model proxying reduces
Secure key isolation: Clients only see proxy keys, keeping real AI keys private.
MCP aggregation: Centralizes all MCP tools installed in CCProxy and exposes them via SSE or Streamable HTTP. Built-in tools include:
WebSearch: Query multiple search engines (Google, Bing, DuckDuckGo, Brave, Tavily, Serper)
WebFetch: JS-rendered page support, precise content extraction, outputs text or Markdown, saving token costs
Desktop assistant features: Translation, mind maps, flowcharts, search, and more
# Tech stack
🛠️ Rust + Vue3 + Tauri
💻 GitHub: [https://github.com/aidyou/chatspeed](https://github.com/aidyou/chatspeed)
📖 Docs: https://docs.chatspeed.aidyou.ai
# Development story
Chatspeed is my first AI-related open-source project and first cross-platform desktop app. In building it, I’ve encountered many challenges — from Rust’s lifetimes to workflow and agent system design — but these experiences shaped CCProxy into a robust and flexible module.
Some challenges I faced:
Spent over a month attempting a text selection tool, ultimately abandoned
Developed DAG Workflow and ReAct Agent in Rust, but ReAct didn’t meet expectations and wasn’t released
Built plugin systems (Deno, pyo3), but shifted focus to MCP support as it matured
Many other small challenges, especially Rust lifetimes 😅
Recently, I used
Looking forward to your questions and feedback! 🚀
https://redd.it/1nue7o6
@r_opensource
Hi everyone 👋 I’m the creator of Chatspeed, an open-source AI proxy + desktop assistant.
# Why Chatspeed exists
For developers, it’s often hard to know whether a model supports tool calls — even the same open-source model may behave differently on different platforms. CCProxy solves this by enabling tool calls for any model: models can invoke tools seamlessly, regardless of their native support, lowering the mental overhead for developers.
Many AI models are either paid or limited in functionality. Claude Code is powerful but expensive. With CCProxy’s protocol conversion, tool compatibility mode, and prompt enhancement, developers can integrate free models from various platforms (e.g., Nvidia’s
qwen3-coder, deepseek-v3.1) into Claude Code workflows, effectively enabling zero-cost access to Claude Code features. Global load balancing allows aggregation of multiple free models to maximize throughput and reliability.Another common pain point is fragmented MCP tool management. Developers often use multiple AI IDEs or plugins, each with its own MCP installation, which is cumbersome to manage. With CCProxy, users can install MCP tools directly within the module, centralizing management and exposing a unified set of tools externally via SSE or Streamable HTTP. Built-in WebSearch and WebFetch tools further enhance other clients’ ability to perform tool calls and fetch/process information efficiently.
# Core module: CCProxy (Chat Completion Proxy)
CCProxy is more than API forwarding — it’s a fully-featured AI middleware:
Protocol conversion: Converts client requests (e.g., OpenAI-compatible) into the target model’s native protocol (Claude, Gemini, Ollama, etc.) and converts the model’s output back, enabling seamless communication across protocols.
Tool compatibility mode: Even models that don’t natively support tool calls can invoke tools through CCProxy.
Proxy groups + prompt management: Scenario-based configuration for different clients or workflows, with dynamic prompt replacement/enhancement.
Global load balancing: Multi-key, multi-model proxying reduces
429 errors by intelligently distributing requests.Secure key isolation: Clients only see proxy keys, keeping real AI keys private.
MCP aggregation: Centralizes all MCP tools installed in CCProxy and exposes them via SSE or Streamable HTTP. Built-in tools include:
WebSearch: Query multiple search engines (Google, Bing, DuckDuckGo, Brave, Tavily, Serper)
WebFetch: JS-rendered page support, precise content extraction, outputs text or Markdown, saving token costs
Desktop assistant features: Translation, mind maps, flowcharts, search, and more
# Tech stack
🛠️ Rust + Vue3 + Tauri
💻 GitHub: [https://github.com/aidyou/chatspeed](https://github.com/aidyou/chatspeed)
📖 Docs: https://docs.chatspeed.aidyou.ai
# Development story
Chatspeed is my first AI-related open-source project and first cross-platform desktop app. In building it, I’ve encountered many challenges — from Rust’s lifetimes to workflow and agent system design — but these experiences shaped CCProxy into a robust and flexible module.
Some challenges I faced:
Spent over a month attempting a text selection tool, ultimately abandoned
Developed DAG Workflow and ReAct Agent in Rust, but ReAct didn’t meet expectations and wasn’t released
Built plugin systems (Deno, pyo3), but shifted focus to MCP support as it matured
Many other small challenges, especially Rust lifetimes 😅
Recently, I used
CCProxy logs to analyze prompt behavior in systems like Claude Code, Cline, Zed, and Crush. Learning from Claude Code’s prompts was particularly insightful, and I’m planning to relaunch the ReAct module soon.Looking forward to your questions and feedback! 🚀
https://redd.it/1nue7o6
@r_opensource
GitHub
GitHub - aidyou/chatspeed: An AI chat proxy with universal tool access, protocol conversion, load balancing, key isolation, prompt…
An AI chat proxy with universal tool access, protocol conversion, load balancing, key isolation, prompt enhancement, centralized MCP hub, and built-in WebSearch & WebFetch — more than an AI...
I'm building an open source .env file replacement and secrets manager
https://envie.cloud/
https://redd.it/1nufadb
@r_opensource
https://envie.cloud/
https://redd.it/1nufadb
@r_opensource
Envie
Envie - manage your environment variables
Speed up development: One command away from prod, staging, or dev. Keep your and your team's environments secure and organized.
So my open source special needs devices won editor choice at bay area maker faire last weekend.
Got a lot of interest in my AAC communication aids, especially the lanyard sized one for “situationally nonspeaking” people.
I am exhausted after 3 days of constant people'ing. But it was well worth it to spend time with my daughter while explaining my communication devices to everyone and encouraging them to "PUSH THE BUTTON!!". Talked to a lot of fellow autistic people and a lot of special education teachers.
Background - built 2 full language system in college, and decided to start building similar devices again in my spare time. I have 2 systems in the hands of special needs individuals that I provided free of cost and am looking for more people with unmet needs to create solutions for.
Project website https://tssfaa.com/
https://redd.it/1nugi8e
@r_opensource
Got a lot of interest in my AAC communication aids, especially the lanyard sized one for “situationally nonspeaking” people.
I am exhausted after 3 days of constant people'ing. But it was well worth it to spend time with my daughter while explaining my communication devices to everyone and encouraging them to "PUSH THE BUTTON!!". Talked to a lot of fellow autistic people and a lot of special education teachers.
Background - built 2 full language system in college, and decided to start building similar devices again in my spare time. I have 2 systems in the hands of special needs individuals that I provided free of cost and am looking for more people with unmet needs to create solutions for.
Project website https://tssfaa.com/
https://redd.it/1nugi8e
@r_opensource
Why pgEdge ‘Ripped the Band-Aid Off’ To Go Totally Open Source
https://thenewstack.io/why-pgedge-ripped-the-band-aid-off-to-go-totally-open-source/
https://redd.it/1nuib09
@r_opensource
https://thenewstack.io/why-pgedge-ripped-the-band-aid-off-to-go-totally-open-source/
https://redd.it/1nuib09
@r_opensource
The New Stack
Why pgEdge ‘Ripped the Band-Aid Off’ To Go Totally Open Source
After a number of notable moves to make software proprietary, some companies are insisting on Open Source Initiative-approved software licenses.
Meet ReadIt - Customise Reddit your way.
https://github.com/readit-mod/readit
https://redd.it/1nulxil
@r_opensource
https://github.com/readit-mod/readit
https://redd.it/1nulxil
@r_opensource
GitHub
GitHub - readit-mod/readit: A usernoscript for Reddit Web with plugin support.
A usernoscript for Reddit Web with plugin support. Contribute to readit-mod/readit development by creating an account on GitHub.
EU citizens, please answer this survey on the governance and sustainability of critical Open Source Software by the European Commission (EU)
https://ec.europa.eu/eusurvey/runner/FOSSEPS_Governance_and_Sustainability_Survey
https://redd.it/1numyxl
@r_opensource
https://ec.europa.eu/eusurvey/runner/FOSSEPS_Governance_and_Sustainability_Survey
https://redd.it/1numyxl
@r_opensource
ec.europa.eu
Study of the European Commission: Survey on the Governance and Sustainability of Critical Open Source Software
EUSurvey is an online survey-management system built for the creation and publishing of globally accessible forms, such as user satisfaction surveys and public consultations.
A new open-source platform for intentional human connections
A few of us in the open-source community have just launched **Compass** — a free, open-source platform designed to help people form deep, intentional connections (platonic, romantic, or collaborative).
We’re in the community seeding phase right now and we’re looking for both early adopters and open source contributors to help shape its direction.
Compass was created because most platforms in this space follow the same pattern: they start promising, but they’re closed-source, investor-driven, and eventually get swallowed by Match Group or similar companies, shifting their priorities from user well-being to monetization.
Compass is different by design:
Fully open source – anyone can inspect, fork, or contribute to the [code](https://github.com/CompassConnections/Compass).
Community-governed – decisions follow a democratic constitution, preventing platform drift.
No ads, no subnoscriptions (just a gift) – funded by [donations](https://www.compassmeet.com/support), not attention mining.
Transparent database and keyword search – no opaque algorithms; you can search profiles directly (e.g., “neuroscience”, “meditation”, “Rust”).
Notifications instead of endless scrolling – you’re alerted when new profiles match your criteria.
We’re trying to prove that something built for the community and by the community can remain aligned with its mission — and never be turned into a product designed to extract value from users.
If you care about open source, human connection, and building alternatives to extractive platforms, we’d love your help. Whether that’s testing it, giving feedback, or contributing code — everything helps.
Try it here: https://www.compassmeet.com/register
Join the community on Discord (see in [About](https://www.compassmeet.com/about))
View the source: https://github.com/CompassMeet/Compass
To know more about me and my other open-source projects, you'll find my contact and socials here.
Would love any thoughts, critique, or suggestions from this community — and if you’re interested in contributing, please reach out!
I really hope we can build something that does a lot of good.
https://redd.it/1nuolgc
@r_opensource
A few of us in the open-source community have just launched **Compass** — a free, open-source platform designed to help people form deep, intentional connections (platonic, romantic, or collaborative).
We’re in the community seeding phase right now and we’re looking for both early adopters and open source contributors to help shape its direction.
Compass was created because most platforms in this space follow the same pattern: they start promising, but they’re closed-source, investor-driven, and eventually get swallowed by Match Group or similar companies, shifting their priorities from user well-being to monetization.
Compass is different by design:
Fully open source – anyone can inspect, fork, or contribute to the [code](https://github.com/CompassConnections/Compass).
Community-governed – decisions follow a democratic constitution, preventing platform drift.
No ads, no subnoscriptions (just a gift) – funded by [donations](https://www.compassmeet.com/support), not attention mining.
Transparent database and keyword search – no opaque algorithms; you can search profiles directly (e.g., “neuroscience”, “meditation”, “Rust”).
Notifications instead of endless scrolling – you’re alerted when new profiles match your criteria.
We’re trying to prove that something built for the community and by the community can remain aligned with its mission — and never be turned into a product designed to extract value from users.
If you care about open source, human connection, and building alternatives to extractive platforms, we’d love your help. Whether that’s testing it, giving feedback, or contributing code — everything helps.
Try it here: https://www.compassmeet.com/register
Join the community on Discord (see in [About](https://www.compassmeet.com/about))
View the source: https://github.com/CompassMeet/Compass
To know more about me and my other open-source projects, you'll find my contact and socials here.
Would love any thoughts, critique, or suggestions from this community — and if you’re interested in contributing, please reach out!
I really hope we can build something that does a lot of good.
https://redd.it/1nuolgc
@r_opensource
Compass
The platform for intentional connections
Local LLM web agents in an open-source browser
https://www.browseros.com/
https://redd.it/1nun2vc
@r_opensource
https://www.browseros.com/
https://redd.it/1nun2vc
@r_opensource
Browseros
BrowserOS - Open-Source AI Browser | Privacy-First Alternative to Perplexity Comet
BrowserOS is the open-source agentic browser that runs AI agents locally on your computer. Privacy-first Chrome alternative with AI superpowers.
Debian GNU Hurd needs only the boot loader for UEFI to work
Not bad but am not knowledgeable enough to know how to fix it.
https://lists.debian.org/debian-hurd/2025/08/msg00139.html
https://redd.it/1nuruj2
@r_opensource
Not bad but am not knowledgeable enough to know how to fix it.
https://lists.debian.org/debian-hurd/2025/08/msg00139.html
https://redd.it/1nuruj2
@r_opensource