AI of course
87 karma · joined April 7, 2024
It's the SQL client I've always wished I had. It's a desktop app, but I made it work in the browser too thanks to DuckDB Wasm.
- https://sprites.dev - https://mise.jdx.dev - https://fnox.jdx.dev
Develop a small API that accepts a prompt, starts a sprite sandbox, and then you wait for the AI agent to open a pull request.
I developed that in less than a week and now my client owns the code, can customize it to their needs, and if they want to use a different compute environment, they can with minimal code changes needed.
Month 2 of building the SQL client I've always wished I had.
One feature I'm especially proud of is the visual query builder. Drag & drop to build SQL queries.
There's also an entire SQL tutorial section for anyone who wants to learn or refresh SQL knowledge.
A day of work to get the prototype working and a few hours the next day to allow multiple users to authenticate.
It's surprisingly simple.
Web app submits the prompt, a sandbox starts on sprites.dev and any Claude output in the sandbox gets piped to the web app for display.
Not sure I can open source it as it's something I built for a client, but ask if you have any questions.
Awesome, thanks for being an early adopter!
I got some great feedback already, so I'll continue building it out.
Roadmap: - Release binaries for Intel Mac, Linux, Windows - Add / test support for more database engines - Wrap up the LLM integration
Holidays are coming up, it may be a productive time haha
However, from personal experience I'm a lot more efficient when I use LLMs to help with tedious, boilerplate-like code writing but I remain in control over structuring the project so it's maintainable by more than machines only.
I use LLMs every day to write tests for example, it's a massive time saving and I wouldn't want to write tests manually ever again.
ZenQuery looks nice, I like the focus on supporting regular files.
Communication between the frontend and backend is via an `invoke` method, easy as pie :).
It's not open source, but once I get the Apple developer stuff all done, I'll add an auto-update feature and open source it.
"But there are many already!" I hear the crowd exclaim.
I respond, "Yes, but..."
It's really something I want for myself. Lightweight, as fast as humanly possible, extensible via plugins (in fact the entire app is mostly plugins, with a small core to glue it together), and a tiny bit of LLM (call it AI if you wish) integration to ask questions about the database or generate/review queries.
It's something I hear from many others too, hence my interest in trying to build an incredibly fast and lightweight product.
The insomnia creator also has a Tauri http client now called Yaak(.app). A huge inspiration to work on my database client.
Also, TIL about PopOS, thank you!
Initially, I thought Tauri with Svelte. Then I learned about gpui (https://www.gpui.rs/) and I'm intrigued.
TIL about iced, which seems to have been around for a while.
Before I embark on building a prototype UI for my app with all three libraries, does anyone have hands-on experience they can share?
It's what Zed(.dev) is based on. While not quite ready for prime time from what I understand, if Zed is the reference implementation, I'm sold!
Think n8n, but you bring your own code and optionally even your hardware to execute pipelines.
That way, there's no lock-in and people can easily use their existing todo.txt to try the TUI.
I'm looking for a TUI idea to explore Ratatui, so maybe this is it right here!
Is there anything you'd like to do in a TUI but can't find a TUI that does it?
Lots of updates this year to reach parity with React Flow.
The use case for a single html file for me is to provide a local dev experience for my customers that mirrors production.
`my-cli dev` spins up the services needed for my platform, including the SvelteKit web app.
hi (at) pipestack.dev should work
Think Zapier or n8n, but you either use existing processing nodes or upload your own code, written of course in any language that compiles to Wasm Components.
It's week 2 but it works and it's live at https://pipestack.dev
Also, scaling across clouds, edge, on-prem with minimal effort.
Another benefit I enjoy now is changing links between components at runtime. I can for example deploy v2.0.0 of a component and send traffic to it by simply updating a link - and reverse it should things break.
All that comes at a much lighter cost in terms of actual $, but also size (kilobytes/megabytes vs megabytes/gigabytes with Docker containers)
That led me to wasmCloud[1]. As the article says, Wasm on the server isn't a winner yet, but it has a chance.