Hacker News new | ask | show | jobs
by porphyra 6 days ago
A lot of prompt engineering goes out of date quickly. Nobody nowadays goes "you are an expert software engineer. make no mistakes" lol.

As a personal anecdote, I find that a lot of big prompts and skills use up context window budget and in many cases agents will eagerly try to use a skill even if it isn't super relevant or necessary for the current task. So when I have too many skills I have to spend a bunch of time toggling the checkboxes to figure out which ones are needed for the task at hand before starting...
2 comments
Royce-CMR 5 days ago
I can't find the link now, but Anthropic has a post about using either a light model call or other logic (regex etc) to dynamically decide what tools to expose per incoming request.

I've run into the same issue and I still end up manually curtailing what's exposed to the model, limiting to the task at hand, but I like the idea of another (smaller I hope) model doing 70% of the clipping instead, automagically.

link
bmitc 4 days ago
> Anthropic has a post about using either a light model call or other logic (regex etc) to dynamically decide what tools to expose per incoming request.

How? Using the agent SDK or Claude Code? If the latter, it'd be nice if they figured that out. There's a huge amount of quality of life things missing from Claude Code. It's a pretty raw frontend to the backend models. And either Claude Code or the backend models get convinced they don't need skills they've been asked to read or even built-in capabilities like reading PDFs.

link
oefrha 5 days ago
> Nobody nowadays goes "you are an expert software engineer. make no mistakes"

You know what, I checked Opus 4.8's instructions to a review subagent the other day and it literally opened with

> You are a senior infrastructure/security engineer doing a thorough, adversarial code review...

I didn't say anything like that myself.

link
mathgeek 5 days ago
Much like agents, I can tell myself I'm a senior infrastructure/security engineer doing a thorough, adversarial code review, but that doesn't change the results much.
link
tclancy 5 days ago
You have to be looking in a mirror and slap your face a couple of times to make it work.
link