Shamelessly promoting in here, I created an architecture that allows an AI agent to have those so called "tools" available locally (under the user control), and works with any kind of LLMs, and with any kind of LLM server (in theory). I've been showing demos about it for months now. Works as a middle-ware, in stream, between the LLM server and the chat client, and works very well. The project is open source, even the repo is outdated, but simply because no one is expressing interest in looking into the code. But here is the repo: https://github.com/khromalabs/Ainara. There's a link to a video in there. Yesterday just recorded a video showcasing DeepSeek V3 as the LLM backend (but could be any from OpenAI as well, or Anthropic, whatever).

The lack of interest may be from the crypto aspect:

> While the project will always remain open-source and aims to be a universal AI assistant tool, the officially developed 'skills' and 'recipes' (allowing AI to interact with the external world through Ainara's Orakle server) will primarily focus on cryptocurrency integrations. The project's official token will serve as the payment method for all related services.

Thank you for the feedback... actually I need to update that, the crypto part of my project will be closed source (an specific remote server) but the idea behind the project itself is universal and open since the very beginning, I already developed dozens of skills including a meta-search engine (searches in several engines at once and combines results dynamically, all balanced by the AI) which are open source as well. Crypto just kind of showed itself as way of funding project, with no strings attached, and till this very day no one else showed up.

What kind of sorcery did you use to create Claude? Honest question :)

Reticulating...

The Vim configuration is something deeply personal, but I'd recommend as a wise choice always explore first the default settings because assuming those in your workflow gives an huge advantage using any new unconfigured vim environment eg to get out of any of the edit modes <C-c> works by default and is a great choice. To use CUA alike shortcuts there's already: ``` source $VIMRUNTIME/mswin.vim ``` And finally, is also a good idea to get used to use <Leader> as a prefix for your own shortcuts, in my case I use the space bar as <Leader>.

Strong agree. Failure to grok what comes with Vim often results in a permenent Nerdtree pane.

I've been using Neovim for about 6 months now but as a former VS Code user I was mostly investing into the various plugins. Fairly recently I started digging deeper into vim's built-in features such as vimgrep and quickfix and they are incredibly powerful. It will take me probably another year to learn to use all these tools effectively.

In its contrib directory, git comes with jump. `git jump grep` will load grep results into quickfix.

Eh. I tried to buy into this for years but I think my poor working memory just pushes me towards having something like it open.

Maybe a little less now that I’ve become a heavy user of tabs. When I start working on a unique task I create a new tab with a few splits with the files I’m interested in. In a way tab views are how I externalize my working memory. But a file tree is still useful to me because file names don’t stick so using a command or picker to swap is often slower

note that exiting a mode with <c-c> prevents autocmd events from firing. (so I have inoremap <c-c> <esc>)

This video demonstrates the real time capabilities of the Ainara AI Framework, an open source project which tries to bring to real life, with a strong focus in local setups, the AI assistants that we have seen previously in Sci-Fi. In this case analyzing Bitcoin market value, news and possible relationships. Github repo: https://github.com/khromalabs/Ainara

Coming from the "old ages" when a PC didn't even feature a (3D capable) GPU and all was about getting all the possible juice of the CPU... shaders now allow incredible things in a very small size, as long as you stick to procedural content. shadertoy.com features an insane amount of great samples.

Last year I've been working in a Golang open source tool with a more modest approach by now (just command line) but with a similar goal (to keep personal info), in my tool formats are described using simple YAML templates and stored in a sqlite db file (https://github.com/khromalabs/keeper), glad to know about more open source tools exploring similar ideas.

"Everything we listen is an opinion, not a fact. Everything we see is a perspective, not the truth" - Marcus Aurelius

I bought an enclosure and related cards from Geekworm [1] some time ago, it's not using a NVMe, instead a 1TB SSD HD. Works like a charm.

[1] https://geekworm.com/collections/raspberry-pi/products/raspb...

I tried Gentoo and Artix, both being a very good experience. Some minor issues with laptop suspension and lightdm session handling in Artix tho. Flawless behavior with Gentoo, pity that an update broke my system a while ago.

An alternative way of having offline documentation directly in the browser: https://devdocs.io/