"Building a skill" is the agentic equivalent of
"writing a script." It is done all the time by almost everyone who uses agents. A skill can be trivial or massive.
Most of the time, I'd advise against writing a skill by hand directly.
A common pattern is walking through how to do something step-by-step and then after it does what you want successfully, ask the agent "Create a new skill to capture this workflow."
The Catholic answer is relatively straightforward in terms of decisions at various councils (or similar structures) about the trinity, iconoclasm, clerical celibacy etc.
With some mix of apostolic succession providing authority and the Holy Spirit guiding the big picture.
It is the "they are terror group kind of like Gestapo" part. It is the "murder and then celebrate murder" part. The "kidnap people, beat them, throw them out" part. It is the "run out of the car, kidnap citizen" part. It is the "throw flashbang and teargas on law obeying people" part. It is the "blind a legal protester". With honorable mention of beating a female woman for going to doctor.
It is also the "their violence is staggering" and "they operations have no respect to law" and "they actually intentionally terrorize citizens too" parts.
All mentioned in article and well documented at this point.
I think that the fact that both Biden and Obama were finding and deporting illegal immigrants in record numbers and there wasn't this kind of outcry indicates that it's not the law itself that people are angry about. It's the cruel, brutal, poorly-targeted, and lawless tactics that are currently being used that are the problem.
It’s misleading to treat record deportations as evidence of restriction. Both Obama and Biden presided over record illegal inflows, and deportations rose largely because the pool of removable people grew faster than enforcement capacity. Inflow dominated outflow, and policy choices further reduced future removability, so net illegal presence increased despite high deportation counts.
That's unrelated to the point that I was making. My point is that they were engaging in record deportations without causing a great deal of upset, and that indicates that it's not the enforcement of the law that people are objecting to now, it's the manner of the enforcement.
This seems like an attempt to avoid the argument rather than engage with it, because creating an imaginary world which we currently don't live in has nothing to do with the reality we're currently dealing with.
The reality being that one informs the other. I know people that would agree in regular situations that we need some enforcement of immigration law. But this is not a regular situation, and when you have an agency tasked with 'enforcing' immigration law who is not enforcing the law at all and in fact violating the law, people question why said immigration laws and agency exists in the first place.
> If ICE were doing exactly what its mandate says on paper (enforcing immigration law and deporting people who are in the country illegally) would you still say the same thing?
They are not doing what the mandate says on the paper. So, the question is moot. If the "paper" said, "ICE is allowed to use arbitrary force and abuse" I would still call them Gestapo, because Gestapo was also legal.
> It seems that critics oppose the law itself, but then frame the argument as “ICE is bad because it breaks the law,” rather than acknowledging that the real disagreement is over whether those laws should exist at all.
This is bad faith claim. It is possible to disagree about multiple things at the same time. It is possible to disagree about the law. And simultaneously find ICE completely morally depraved and violent, just as their defenders.
In fact, one can agree about the law and still think that government force killing people and being violent against opposition is a bad thing.
So very openly: I think that stealing from shops should be illegal. I think that police beating people for stealing would be wrong and murdering people who oppose the "beat people who steal" law also wrong.
There's nothing bad faith here, it was a simple question.
Sounds like you aren't even sure whether the law is good or bad "[I don't care what the paper says]". You hate ice because they seem like they're Gestapo. Then maybe next time, don't appeal to the law to argue "ICE is illegal"?
I didn't see anything in there about working with the judicial system to ensure due process. Identifying and apprehending people is a bad thing when there is no due process.
> “It is in nobody’s interests for there to be worries and instability in the U.S.,” said Jonathan Haskel, an economics professor at Imperial College Business School and former member of the rate-setting committee at the Bank of England. “Other countries hold lots of American assets. Savers in Europe will implicitly be invested in the American stock market. America is in many ways a flagship engine, with the A.I. revolution going on. Nobody in the world wants to see that at risk.”
DOGE was mostly about government efficiency and restructuring rather than actually proving vast fraud. What DOGE did accomplish is cutting a lot of politicized / ideologically driven spending.
Fraud is a serious crime and it is not just "political actions that I don't like." The Somalian fraud rings are actually fraud on a vast scale. So please stop defending actual fraud, which is supposed to not be a partisan issue
Musk was still talking about Trillion dollar government fraud on Twitter a few days ago.
He repeatedly claimed fraud as part of DOGE. The 150 year olds claiming Social Security for example.
Just like now, he was bullshitting and didn't care that it wasn't true if it was useful to him.
His allies are pretty much on record saying that too, e.g. that firing staff increased government costs since the same expertise was contracted out at a higher rate, but that is considered a win for them. Which feels like fraud to me.
> DOGE did not find $2T in fraud, but that doesn’t matter, Musk allies say
This is a strange test to care about. As if people choose linux vs. windows based off of performance on one specific benchmark on one specific hardware platform.
Would you rather they cut out whatever information they're trying to "hide" from the underlying HTML so that it never hits your browser and you have no chance of seeing it?
I wouldn’t mind if some information were omitted entirely, or even hidden by default, as long as the approach is transparent and users are given the option to reveal it if they want to.
What feels concerning here is that model identifiers (like GPT-5.2) are included in the DOM but hidden through CSS properties like clip-path, opacity: 0, and user-select: none. This doesn’t feel like typical UX simplification—it looks more like deliberate obfuscation.
If the goal were simplicity, a toggle or clearly labeled section would work just as well, without undermining trust. I think users generally appreciate being informed and offered choices.
From a regulatory standpoint, this kind of design could also raise questions under frameworks like the GDPR and the EU AI Act, which emphasize transparency, informed consent, and the right for users to understand how automated systems operate. Intentionally hiding relevant model information in the DOM without clear disclosure could be seen as inconsistent with those principles.
Most of the time, I'd advise against writing a skill by hand directly.
A common pattern is walking through how to do something step-by-step and then after it does what you want successfully, ask the agent "Create a new skill to capture this workflow."