The fixed fee plan is because the agent and the tools have internal choices/planning about cost. If you simply pay for API the only feedback to them that they are being too costly is for you to stop.

If you look at tool calls like MCP and what not you can see it gets ridiculous. Even though it's small for example calling pal MCP from the prompt is still burning tokens afaik. This is "nobody's" fault in this case really but you can see how the incentives are and we all need to think how to make this entire space more usable.

Anthropic should do a full inquiry into this and fire people for it. No questions asked. Fire them. And then reply with automated bits.

How about both worlds?

Instead of asking the agent to execute it for you, you ask the agent to write an install.sh based on the install.md?

Then you can both audit whatever you want before running or not.

So... What you are saying is that we don't need 'install.md'. Because a developer can just use a LLM to generate a 'install.sh', validate that, and put it into the repo?

Good idea. That seems sensible.

Bonus: LLM is only used once, not every time anyone wants to install some software. With some risks of having to regenerate, because the output was nonsensical.

> What you are saying is that we don't need 'install.md'

I think the point was that install.md is a good way to generate an install.sh.

> validate that, and put it into the repo

The problem being discussed is that the user of the script needs to validate it. It's great if it's validated by the author, but that's already the situation we're in.

> The problem being discussed is that the user of the script needs to validate it. It's great if it's validated by the author, but that's already the situation we're in.

The user is free to use a LLM to 'validate' the `install.sh` file. Just asking it if the script does anything 'bad'. That should be similarly successful as the LLM generating the script based on a description. Maybe even more successful.

I still dont understand why we need any of them. If I am installing something, It would take me more time to write this install.md or install.sh than if I just went to the correct website and copied the command, see the contents, run it and opening help.

And since LLM tokens are expensive and generation is slow, how about we cache that generated code on the server side, so people can just download the pre-generated install.sh? And since not everyone can be bothered to audit LLM code, the publisher can audit and correct it before publishing, so we're effectively caching and deduplicating the auditing work too.

This is much better. Plus you get reproducibility and can leverage the AI for more repeat performances without expending more tokens.

then how about you cut out the llm middleman and just audit the bash scripts already provided?

AI (chat) companies now have enough data to recommend h2h (human to human) but they are not building this.

They could literally find people who are working on he same things and recommend them for networking etc.

But that would take you off platform. Off attention.

Who wants to build this. Others must be thinking the same thing?

What exactly was the bad stuff? He was insensitive about empirical reality or he was literally wrong about something in the sense of being very confident about something despite having little data? Or something else? I only remember the cartons really but was aware some people seemed to be irked about him recently.

Some random internet poll said many people of race A agreed it was "not OK" to be a person of race B. Adams said if that were true, then people of race B should probably not hang out with people of race A that thought it was not OK to be race B. The internet did its thing and quoted him out of context, and tried to cancel him. He dug in his heels and doubled down. He also liked a certain president that many dislike. And here we are.

> The internet did its thing and quoted him out of context

Let's not act like this is some case of out of context quotes. Here's the actual quote for people to decide for themselves:

"I'm going to back off from being helpful to Black America because it doesn't seem like it pays off. I get called a racist. That's the only outcome. It makes no sense to help Black Americans if you're white. It's over. Don't even think it's worth trying. I'm not saying start a war or do anything bad. Nothing like that. I'm just saying get away. Just get away."

>Let's not act like this is some case of out of context quotes. Here's the actual quote for people to decide for themselves:

you quote, but you did not include the context, so your attempt not to be out of context is a fail.

Can you not use PAL MCP for this? Have one top agent as controller etc? It's not ideal but it feels like the space of multi agent stuff is evolving ... I notice that there are a lot of posts on hn about these things so we are trying to do the same thing really.

It's nice to see someone else going mad, even deeper down the well.

I don't known the details but I was wondering why people aren't "just" writing chat venues any commns protocols for the chats? So the fundamental unit is a chat that humans and agents can be a member of.

You can also have DMs etc to avoid chattiness.

But fundmantally if you start with this kind of madness you don't have a strict hierarchy and it might also be fun to see how it goes.

I briefly started building this but just spun out and am stuck using PAL MCP for now and some dumb scripts. Not super content with any of it yet.

Thank you for saying this. I often find this "glib" explains of ML stuff very frustrating as a human coming from an Applied Math background. It just makes me feel a bit crazy and alone to see what appears to be a certain kind of person saying "gosh" at various "explanations" when I just don't get it.

Obviously this is beautiful as art but it would also be useful to understand how exactly these visualizations are useful to people who think they are. Useful to me means you gain a new ability to extrapolate in task space (aka "understanding").

Everyone should be simply posting algorithmic content to Facebook. Screenshots, etc not giving them your own life stuff imo. We need to push back on personalized feeds. Share a high percentage of what you see so that there is a digital commons and not just some island for each person.

Social media platform used to be less about passive consumption.

I would be curious to know if the treatment of statues in terms of "making them ugly and ridiculous to the point of being insulting" is roughly uniform across the different historical cultures being treated to this "reconstruction" procedure.

i.e. is there evidence that there is comfort in trolling using Roman or Greek vs Assyrian, Nubian etc. Or do they just like to make everything bright and blocky.