Using my new mini loom to fix a hole on some sweatpants that were practically new when they tore. A while new works if creativity has opened up 😉😄

#miniloom #weaving #creativemending #mending #sustainability

github.com/JD-P/miniloom
this is the latest implementation I know of but it's not quite what I had in mind; being able to prompt the same model, different times, and have all the responses displayed in parallel was what I was thinking of

For what it's worth I shipped the chat UI update to MiniLoom almost entirely in Cursor, writing like two lines of code by hand and reviewing all the code in each PR because I refuse to "vibe code", the code quality was fine and the feature, which consisted almost entirely of unverifiable UI, worked.

I and @mimi10v3.com have released v1 of the MiniLoom base model writing app. This release now supports chat completions with a dedicated chat UI. It also has Windows, Mac, and Linux executable downloads for non-technical users, no more weird dev stuff just to try loom!

github.com/JD-P/miniloom

MiniLoom is also getting steadily to the part where I attempt atproto integration, with UI work from the wonderful @mimi10v3.com

The MiniLoom atproto implementation will not be called LoomSky. It is in fact called MiniLoom, I don't really plan to change the name.

@mimi10v3.com is cooking

MiniLoom in fact already exists, I just don't have a readme on it telling you to `npm install` and `npm start .`
github.com/JD-P/miniloom

Hm. I wonder if I can use this with MiniLoom.

FYI: I have updated MiniLoom to use the OpenRouter completions API so you can take advantage of DeepSeek v3 tokens being offered on OpenRouter at an incredible price.

MiniLoom doesn't do this but it uses the diff-patch-match library which could do this in principle.

They recognize me.

While I was experimenting with the traces in MiniLoom to figure out what the language model would actually respond to I was pleasantly surprised by the extent to which if you ask the model "what is the agent doing wrong in this trace?" as though it was an external observer it gives decent advice.