Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: What am I doing wrong Re Agentic coding

news.ycombinator.com

Post reply on HN
Print view View original

11–17 of 17 posts

Re: Ask HN: What am I doing wrong Re Agentic coding

#11

Post by crashprone »

A while ago, a HN comment linked to this blog by Mario Zechner https://mariozechner.at/posts/2025-06-02-prompts-are-code/ which was exactly what I needed back then. The workflow helped me analyse about 300 files. It definitely made up stuff along the way and missed valuable context that I would have seen had I done the analysis manually, but it's a good starting point.

Without a structured workflow, as you said, after 5 files, it's starts going berserk.

That said, Claude has been really bad the last few weeks, especially in VSCode Copilot. If you ask it repeatedly, it admits that's its Sonnet 3.5 and not Sonnet 4. Not sure if it's true, but Sonnets workflow has degraded the last few weeks.

You would get better results probably by using search and replace or a python script to make those changes.

The comments that insist you'd get better results by talking to it like you would to a human, I hope they're trolling. The idea that the tone of your prompt has any influence at all, is laughable.

Re: Ask HN: What am I doing wrong Re Agentic coding

#13

Post by 8note »

the task is pretty unclear. youve got one real ask, for it to remove a log statement, but you've buried it in irrelevant context.

its better if you can let it read the before code, then read the after code, then you give it the ask - remove the log, switch callers to the new interface.

then, have it write up a work/prompt ppan, and keep progress in an markdown file

Re: Ask HN: What am I doing wrong Re Agentic coding

#14

Post by yurifury »

While you aren't getting consistent, usable results from your setup you shouldn't allow it to many any changes that you disagree with. For every change it proposes, review thoroughly and tell it what you'd prefer instead. It will get better during the session, and you can ask it to codify the rules you've communicated implicitly in the CLAUDE.md file.

Re: Ask HN: What am I doing wrong Re Agentic coding

#15

Post by asdfbank »

spott wrote: post #4

The mode is getting lost because the task you gave it, is way up the context chain from what it is currently working on. It loses track of its task and starts working on other things that it notices. The way to get around this is to never have the model just “do the thing”. Have it create a plan and create a todo list from the plan (it will do this on its own typically in Claude Code), the. You “approve” the plan, th…

I agree with this, break it down further.

When instructing it to execute the plan, i would even have it only execute step 1, then you review. then step 2, review and so on. this prevents it from going down a rabbit hole and you can adjust the plan's steps or change course more easily.

Re: Ask HN: What am I doing wrong Re Agentic coding

#16

Post by sky2224 »

Are you providing the relevant files in your prompt as context? Additionally, even with files added as context, your prompt isn't very well focused. You essentially said, "do the thing", with a couple of details.

Outline through examples what you want the model to do. You mentioned there's a new signature to follow. Add that to the prompt.

Re: Ask HN: What am I doing wrong Re Agentic coding

#17

Post by gangtao »

My feeling, work with LLM is like teaching kids

Dont say not to do somthing, always say do something instead

This positive instruction style works because it gives the LLM a clear target to aim for rather than a boundary to avoid. It's like telling a child "Walk carefully on the sidewalk" instead of "Don't run into the street"

Post reply on HN