Hacker Newsnew | past | comments | ask | show | jobs | submit | throwpoaster's commentslogin

Generally, across AI providers, I have come to interpret sudden degradation in existing capabilities as a signal that a new, more expensive, product tier is about to launch.

Have you tried asking Mythos for a fix?

Wait what? If I get told to come back in three hours because I'm using the product too much, I get penalized when I resume?

What's the right way to work on a huge project then? I've just been saying "Please continue" -- that pops the quota?


If you struggle with procrastination please get assessed for ADHD.

A positive diagnosis is life changing.


Only if you are okay with the idea of taking stimulant medication.

Alternatively, you must be okay to have lifelong struggle and always feel inferior to everyone around you basically doing stuff effortlessly while you suffer.

There is non-stimulant medication as well for ADHD. If you're really struggling, it might be worthwhile to suspend judgement and actually try these out for a while. In the worst case you go back to how you were without medication. For many people the potential upside is worth the experiment.

Note that not being okay with that idea is a symptom of ADHD. Talk about it with your psychiatrist.

Several people on my department got that recently. Clearly in fashon. But it seems just a license to be lazy

If you know you have dificulties you need to work on bettering yourself, small steps, not find a justification or mask it with drugs

And with the widespread addiction of social media, youtube... frequent world crumbling news, boring jobs... most people would qualify for some sort of diagnosis, which is misleading


Shuttle was awesome and the people who love to hate it can personally fight me.

A comment that takes a second to realize it’s a troll.

Repeatability (see my response to op).

Skills and MCP are useful for when you need to repeat specific processes, perpetually. Without them the task description and reasoning falls out of the context window, or is compressed, and the process fails.

An agent will eventually forget, or hallucinate, guardrails and requirements. Yes to AGENTS.md, but when you're actively managing the whole context window in a long-running task you don't want to just keep jamming stuff in there and hope for the best. Skills help budget tokens and stabilize around specific outcomes.

If your use case is not agentic, as you build a skill corpus you can begin having the model reason at higher and higher levels about the outcomes you're aiming at.

Eg: I'm super lazy now and ask Claude to launch the project instead of just running the command myself. This is probably best done as a skill.


AGENTS.md doesn't fall out of the context window so you just put whatever commands you use in there. Every time it gets too big you rewrite it.

Never had an issue doing this


How long have you had an agent performing the same task, or trying to improve performance towards the same outcome?

I’m not really talking about coding. I’m talking about autonomy.


They’re different things. You can have skills using MCP.

“A sculptor drove a truck to fund his art.”

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: