Hacker News Reader

Comment by simonw | Hacker News Reader

Agents as staff replacements that can tackle tasks you would normally assign to a human employee didn't happen in 2025.

Agents as LLMs calling tools in a loop to perform tasks that can be handled by typing commands into a computer absolutely did.

Claude Code turns out to be misnamed: it's useful for way more than just writing code, once you figure out how to give it access to tools for other purposes.

I think the browser agents (like the horribly named "ChatGPT Agent" - way to burn a key namespace on a tech demo!) have acted as a distraction from this. Clicking links is still pretty hard. Running Bash commands on the other hand is practically a solved problem.

View on HN

We still sandbox, quarantine and restrict them though, because they can't really behave as agents, but they're effective in limited contexts. Like the way waymo cars kind of drive on a track I guess? Still very useful, but not the agents that were being sold, really.

Edit: should we call them "special agents"? ;-)

Have you been in a Waymo recently or used Tesla FSD 14.2? I live in Austin and my Model 3 is basically autonomous - regularly going for hours from parking space to destination parking space without my touching the steering wheel, navigating really complex situations including construction workers using hand motions to signal the car.

Thanks for collecting training data for the rest of us, 'coz I don't trust Musk with my life.

Have you been in a Waymo recently or used Tesla FSD 14.2?

One of these things is not like the other...

"special" agents from the CIA (Clanker Intelligence Agency)

> Agents as LLMs calling tools in a loop to perform tasks that can be handled by typing commands into a computer absolutely did.

I think that this still isn't true for even very mundane tasks like "read CSV file and translate column B in column C" for files with more than ~200 lines. The LLM will simply refuse to do the work and you'll have to stitch the badly formatted answer excerpts together yourself.

Try it. It will work fine, because the coding agent will write a little Python script (or sed or similar) and run that against the file - it won't attempt to rewrite the file by reading it and then outputting the transformed version via the LLM itself.

Reply on news.ycombinator.com