LLM stuff
Qwen 3.8 27b is scary good! At least at more routine stuff where I tell it what to do.
LLM stuff
The scary part here is just how long it can remain coherent as its contexts fills up. Even 3.6 would get confused quite quickly and need to be compacted after a few hours. This thing managed to ingest files from like three codebases and still have space for generating a weird setup for testing a react-native component.
LLM stuff
I think part of this is that I've gotten better at working within it's limitations and strengths as I've been updating my "agent" to be more effective.
LLM stuff
I haven't had much of a need for skills so far because the models are pretty capable even with minimal prompting and I found that skills often fill up the context more than is strictly necessary. I was thinking for ergonomics I'd have users define skills in their "documents" folder to make it more intuitive for graphical interface users.
re: LLM stuff
I created a skill for this too. It works like a charm. Has all the standard engineering rigor, best-practices, security, tests, accessibility, etc, applied as a step by step workflow which must be validated before marking any task complete. It's a total game changer, improving the output by an order of magnitude. Highly recommended.
re: LLM stuff
@toiletpaper Is your skill published somewhere I could read? :o
I found I haven't needed to spell out how rigorous it needs to be because it's pretty good at that out of the box and past that I'm guiding it.
re: LLM stuff
No. I can send it to you privately if you want. I'd publish it, but I've been doing my best to compartmentalise my personal life from my fedi presence, for hopefully obvious reasons. Happy to share though, just not directly on fedi.
LLM stuff
I've recently added support for AGENTS.md so I don't need to tell it how I want to do testing and code style checking every time and generally save some extra file reads whenever it starts a new feature. Next I'm gonna play with "skills" so I can load up commonly used instructions instead of repeating myself.