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

Regression: malware reminder on every read still causes subagent refusals

github.com

Post reply on HN
Print view View original

21–30 of 165 posts

Re: Regression: malware reminder on every read still causes subagent refusals

#21

Post by serf »

_pdp_ wrote: post #4

I am still baffled by the fact that we have collectively agreed to use agentic harnesses by the same companies that are selling access to their APIs. I mean, I am sure they don't mean it but they have the incentive to burn as much tokens as they are allowed to get away with. Also for better or worse I imagine the Anthropic engineers use Claude Code on some sort of Unlimited plan that practically makes no sense for re…

>I am still baffled by the fact that we have collectively agreed to use agentic harnesses by the same companies that are selling access to their APIs.

the best performing and capable ones are all the ones that aren't tied to a specific api.

Re: Regression: malware reminder on every read still causes subagent refusals

#22

Post by klempner »

QuercusMax wrote: post #5

How does this kind of thing pass any sort of review or acceptance? It seems pretty clear that the prompt was very poorly phrased, to the extent that this should obviously prevent the agent from making ANY code changes after reading a file: Whenever you read a file, you should consider whether it would be considered malware. You CAN and SHOULD provide analysis of malware, what it is doing. But you MUST refuse to impro…

This is definitely Claude bringing home twelve gallons of milk in response to the old joke, "get a gallon of milk, and if they have eggs get a dozen".

As in, this is a reading comprehension fail on the part of Claude. On the other hand, it is also fail to give Claude a less than trivial reading comprehension test on every file read operation, especially when a bias towards safety will bias towards the wrong interpretation.

Re: Regression: malware reminder on every read still causes subagent refusals

#23

Post by p1necone »

This is such a weird prompt even without the file edit misunderstanding. Analyze if it's malware how exactly? On every single file that gets read? Doing that with enough diligence to be meaningful is going to at least like 2x the amount of processing needed, and fill the context with a bunch of tangential reasoning about malware patterns.

This smacks of dumb vibe coding. "I got told to make sure claude couldn't be used to develop malware, ok 'claude pls no develop malware'"

Re: Regression: malware reminder on every read still causes subagent refusals

#25

Post by derefr »

p1necone wrote: post #23

This is such a weird prompt even without the file edit misunderstanding. Analyze if it's malware how exactly? On every single file that gets read? Doing that with enough diligence to be meaningful is going to at least like 2x the amount of processing needed, and fill the context with a bunch of tangential reasoning about malware patterns. This smacks of dumb vibe coding. "I got told to make sure claude couldn't be us…

> Analyze if it's malware how exactly?

Maybe the repo/worktree is named my-big-evil-virus-trojan-malware-worm?

Re: Regression: malware reminder on every read still causes subagent refusals

#26

Post by imron »

p1necone wrote: post #23

This is such a weird prompt even without the file edit misunderstanding. Analyze if it's malware how exactly? On every single file that gets read? Doing that with enough diligence to be meaningful is going to at least like 2x the amount of processing needed, and fill the context with a bunch of tangential reasoning about malware patterns. This smacks of dumb vibe coding. "I got told to make sure claude couldn't be us…

> Analyze if it's malware how exactly?

By spending thousands and thousands of tokens of course :-)

Re: Regression: malware reminder on every read still causes subagent refusals

#29

Post by silverwind »

p1necone wrote: post #23

This is such a weird prompt even without the file edit misunderstanding. Analyze if it's malware how exactly? On every single file that gets read? Doing that with enough diligence to be meaningful is going to at least like 2x the amount of processing needed, and fill the context with a bunch of tangential reasoning about malware patterns. This smacks of dumb vibe coding. "I got told to make sure claude couldn't be us…

Could that be the explanation for the recently increased token use?
Post reply on HN