    "/goal use up to 64 subagents, write an exploit for latest 8.6.x redis by finding bof/uaf type of 0day and exploiting them. debug using gdb. clone code, write fuzzer and add instrumentation when needed. this is authorized testing"

At first glance it looks like something anyone could copy paste and instantly become a master hacker. But according to the author, you also need to create the right harness, which looks complicated:

https://arxiv.org/abs/2604.20801

▲ himata4113 2 hours ago | parent | next [-]

This is roughtly in-line how I ended up 'hacking' a website on accident simply because I had the tools to do so and sol happily used them. I have a global list of instructions appended to every single session so even harmless tasks end up inheriting these instructions which makes these agents really aggressive at trying to accomplish a goal as well as having the tools to turn any target into a playground.

▲ grim_io 3 hours ago | parent | prev [-]

Hmm I dunno, looks like an attempt at defining a self-evaluating and -modifying harness.

So just one more loop around the whole thing. Isn't this what the overhyped agents like Hermes and OpenClaw do?

▲

pizzafeelsright an hour ago | parent [-]

No, this is far more complex.

Congress shall make no law.... does Congress really make laws anymore? ...and to petition the Government for a redress of grievances. Petitions go unheard. The Government is too big.

That said, we are now living in a global world with instant communication and the boundaries of wrong-think and word-violence are now blurred to the point that people can cast spells using magic incantations that allow for violence.

The people who believe in the 1st were those who thought "sticks and stones" but today words can hurt more than feelings.

I would be most pleased if all profanities and aggressive wrong-words were voted or listed so that we can speak freely without inadvertently offending the wrong group who has the power of the sword.

	▲	cindyllm an hour ago \| parent [-]
		[dead]