Y
Hacker News
new
|
ask
|
show
|
jobs
by
nperez
241 days ago
Seems like a more organized way to do the equivalent of a folder full of md files + instructing the LLM to ls that folder and read the ones it needs
1 comments
j45
241 days ago
If so it would be most welcome since LLMs doesn't always consistently follow the folder full of MD files to the same depth and consistency.
link
RamtinJ95
241 days ago
what makes it more likely that claude would read these .md files then?
link
j45
241 days ago
Skills is hopefully put through a deterministic process that is guaranteed to occur, instead of a non-deterministic one that can only ever be guaranteed to happen most of the time (the way it is now).
link
adastra22
240 days ago
It is literally just injecting context into the prompt.
link
adastra22
240 days ago
It includes both the file names and a configurable description string. That’s where you put the TLDR of when to use each skill.
link
j45
240 days ago
This improves it a great deal but at a certain point, maybe 60-80% of the way it can start fading.
link
phildougherty
241 days ago
trained to
link