Forwarded from OrangeFox Recovery Support (Agent A)
We're excited to share that as of today, our infrastructure is now fully open source!
You can explore all of our in-house projects over at: https://gitlab.com/OrangeFox/infrastructure
We're open to contributions, whether it’s to the recovery or infrastructure, and we’ve set up a dedicated email for any questions or collaboration inquiries: volunteers@orangefox.tech.
Come take a look, get involved, and help us build something awesome.
You can explore all of our in-house projects over at: https://gitlab.com/OrangeFox/infrastructure
We're open to contributions, whether it’s to the recovery or infrastructure, and we’ve set up a dedicated email for any questions or collaboration inquiries: volunteers@orangefox.tech.
Come take a look, get involved, and help us build something awesome.
GitLab
Infrastructure · GitLab
Small weekend projects of mine.
I've drafted the FoxWiki (in-house made wiki engine that pipelines markdown to react elements)
I've also drafted the FoxPaste, the stupidly simply pastebin-like service.
Both of them are using the sets of frameworks and tools I've made for the new downloads website and FoxBox which I've shown here already.
I expect to launch the FoxPaste one at the end of the month as the backend side should be pretty straightforward to do.
I've drafted the FoxWiki (in-house made wiki engine that pipelines markdown to react elements)
I've also drafted the FoxPaste, the stupidly simply pastebin-like service.
Both of them are using the sets of frameworks and tools I've made for the new downloads website and FoxBox which I've shown here already.
I expect to launch the FoxPaste one at the end of the month as the backend side should be pretty straightforward to do.
❤2
Visually, I also want to share, that my vision of the design, is that it has to be stupidly plain, yet playful.
The main question I've always asked myself is how can I make my thing as less cognitive complex as it could even be?
The answer is not to overcomplicate the product, but to highlight the most important elements wisely.
The main question I've always asked myself is how can I make my thing as less cognitive complex as it could even be?
The answer is not to overcomplicate the product, but to highlight the most important elements wisely.
This media is not supported in your browser
VIEW IN TELEGRAM
Today - animations, optimizations and folders support
❤3
Today - variants, that could be Languages, Versions or whatever wiki variants you can think of.
Internally those are just simple first-level folder of the content :D
Internally those are just simple first-level folder of the content :D
❤4
Forwarded from Sophie NEWS 📢
Long time no see!
Today, it's rather an up-keeping [#Beta] update for you all.
First, the default model was changed to Gemini 2.5 Flash, which is exactly on 0.5 times smarter than previously used Gemini 2.0 Flash. Apparently, it's now able to count r's in 'strawberry' correctly! 🎉
The model's instruction was improved to provide more clarity and encourage AI to save information in the memory more often.
Lastly, the DuckDuckGo search tool was replaced with Tavily, which will provide more search capacity for Sophie.
🗓 As always, this update is rolling out first to users with Beta mode enabled and will gradually reach more users in the coming weeks through an A/B release process, ensuring the seamless and hassle-free update.
To use AI Features you need to activate them using
The Privacy Policy (https://sophie-wiki.orangefox.tech/docs/Privacy%20policy) has been updated to reflect the addition of Tavily.
Today, it's rather an up-keeping [#Beta] update for you all.
First, the default model was changed to Gemini 2.5 Flash, which is exactly on 0.5 times smarter than previously used Gemini 2.0 Flash. Apparently, it's now able to count r's in 'strawberry' correctly! 🎉
The model's instruction was improved to provide more clarity and encourage AI to save information in the memory more often.
Lastly, the DuckDuckGo search tool was replaced with Tavily, which will provide more search capacity for Sophie.
To use AI Features you need to activate them using
/aienable yes first.The Privacy Policy (https://sophie-wiki.orangefox.tech/docs/Privacy%20policy) has been updated to reflect the addition of Tavily.
Please open Telegram to view this post
VIEW IN TELEGRAM
Hey Nerds!
I won a battle with Ansible and GitLab CI, so the foxwiki is finally deployed!
Take a look - https://foxwiki.orangefox.tech/
I won a battle with Ansible and GitLab CI, so the foxwiki is finally deployed!
Take a look - https://foxwiki.orangefox.tech/
❤5👍3
I have wonderful news, Vercel.com agreed to sponsor OrangeFox, this is amazing as it will lower the load of maintaining our own CI/CD and infrastructure.
To celebrate this I've just hosted a very small project made using parts of the foxpaste and foxwiki, a playground, to see how foxwiki's markdown engine renders markdown!
Try it here - http://foxwiki-playground.orangefox.tech/
To celebrate this I've just hosted a very small project made using parts of the foxpaste and foxwiki, a playground, to see how foxwiki's markdown engine renders markdown!
Try it here - http://foxwiki-playground.orangefox.tech/
yachu
Hey Nerds! I won a battle with Ansible and GitLab CI, so the foxwiki is finally deployed! Take a look - https://foxwiki.orangefox.tech/
Today night:
- Backend changes to prefetching wiki data on server before sending to client, so the client side wouldn't piggy-back the server for the data right after being loaded.
- Added default re-fetching every 30 minutes, if the wiki page was changed after the user loaded the page, it will still be updated for user, eventually.
Surely, there's a lot to do about performance, and here's the major points I want to implement:
- prefetch wiki page data on hover.
- on first load due to custom fonts, wiki would use fallback fonts for a short time, resulting in website blinking
- due to material design icons support, we inbuilt the whole icons library to wiki, is there any better way?
- Backend changes to prefetching wiki data on server before sending to client, so the client side wouldn't piggy-back the server for the data right after being loaded.
- Added default re-fetching every 30 minutes, if the wiki page was changed after the user loaded the page, it will still be updated for user, eventually.
Surely, there's a lot to do about performance, and here's the major points I want to implement:
- prefetch wiki page data on hover.
- on first load due to custom fonts, wiki would use fallback fonts for a short time, resulting in website blinking
- due to material design icons support, we inbuilt the whole icons library to wiki, is there any better way?
Today in FoxWiki:
- Support for PDF files, natively shows up in a sidebar, just drop a file to your docs directory
- Support for HTML / iframe pages, would require creating a custom Page name.json with an url or html-content, icon, ... properties.
- Support for PDF files, natively shows up in a sidebar, just drop a file to your docs directory
- Support for HTML / iframe pages, would require creating a custom Page name.json with an url or html-content, icon, ... properties.
Forwarded from Sophie NEWS 📢
- Added a new command
/aiprovider to change the AI provider (see attached screenshot)- We now support Mistral AI — a European provider using open-weight models, offering top-tier data privacy!
- The default model is now ChatGPT GPT-5-mini
- AI Chat can now access Sophie’s documentation, in addition to Inbuilt Memory and Internet Searches — just ask her directly how to use specific features (see attached screenshot)
- Added a shortcut in /help to quickly start an AI chat
- Significant upgrades were made across the entire AI stack
- Added English (United Kingdom) 🇬🇧 language
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
👍1