I'm actually working on an MCP control plane and looking for anyone who might have a use case for this / would be down to chat about it. We're gonna release it open source once we polish it in the next few weeks. Would you be up to connect?
You can check out our super rough version here, been building it for the past two weeks: gateway.aci.dev
A MCP gateway is a useful tool, I have a prototype of something similar I built but I'm not super enthusiastic about working on it (bigger fish to fry). One thing I'd suggest is to have a meta-mcp that an agenct can query to search for the best tool for a given job, that it can then inject into its context. Currently we're all manually injecting tools but it's a pain in the ass, we tend to pollute context with tools agents don't need (which makes them worse at calling the tools they do) and whatnot.
What I was talking about here is different though. My agent (Smith) has an inversion of control architecture where rather than running as a process on a system and directly calling tools on that system, it emits intents to a queue, and an executor service that watches that queue and analyzes those intents, validates them, schedules them and emits results back to an async queue the agent is watching. This is more secure and easier to scale. This architecture could be built out to support safe multiple agents simultaneously driving your desktop pretty easily (from a conceptual standpoint, it's a lot of work to make it robust). I would be totally down to collaborate with someone on how they could build a system like this on top of my architecture.
Our gateway lets team members bundle together configured MCPs into a unified MCP server with only two tools -- search and execute, basically a meta-mcp!
Very interesting! What kind of use cases are you using your agent (Smith) for? Is it primarily coding, or quite varied across the board?
Right now I'm 100% coding focused, that's the big show in terms of agents. Orchestrating current agent tools is clunky, they're low performance, they lack fine grained extensibility to really modify their behavior on a dynamic task based basis (CC's hooks are the "best" option and they're really weak), the security model around them is flawed, there's a laundry list of issues with them.
The agent itself is designed to be very general, every trace action has hooks that can transform the payload using custom javascript, so you can totally change the agent's behavior dynamically, and the system prompts are all composed from handlebars templates that you can mix/match. The security model makes it great for enterprise deployment because instead of installing agent software on systems or giving agents limited shell access to hosts, you install a small secure binary that basically never changes on hosts, and a single orchestrator service can be a control plane for your entire enterprise. Then every action your agent takes is linked into the same reactive distributed system, so you can trigger other actions based on it besides just fulfillment of intent.
Interesting, for once 'Matrix's 'programs hacking programs' vision kinda starts to make some sense. Maybe it was really just way ahead of its time, but became popular for reasons similar to Cowboy Bepop ( different timeline, but familiar tech from 90s ).
Members can then bundle the various MCP servers together into a single unified MCP server that contains just two tools -- search and execute, so it doesn't overload context windows. The team members then get a remote MCP server URL for the unified MCP server bundle to bring into Cursor!
this statement is just patently wrong, and even those are still significant value. LLMs have been significantly impacting software engineering and software prototyping.
Do you have any evidence to suggest that LLMs have increased productivity in software? And if so what is the effect size?
I don’t really see any increase in the quality, velocity, creativity of software I’m either using or working on, but I have seen a ton of candidates with real experience flunk out on interviews because they forget the basics and blame it on LLMs. I’ve honestly never seen candidates struggle with syntax the way I’m seeing them do so the last couple years.
I find LLMs very useful for general overviews of domains but I’ve not really seen any use that is a clear unambiguous boost to productivity. You still have to read the code and understand it which takes time, but “vibe coding” prevents you from building cognitive load until that point, making code review costly.
I feel like a lot of the hype in this space so far is aspirational, pushed by VCs, or a bunch of engineers who aren’t A/B testing: like for every hour you spend rubber ducking with an LLM, how much could you have gotten thought out with a pen and paper? In fact, usually I can go and enjoy my life more without an LLM: I write down the problem I’m considering and then go on a walk, and come back and have a mind full of new ideas.
It's hard to see where you're coming from. With o1 Pro or Gemini 2.5 Pro I can give a detailed text specification of a module I want to write to it and it'll write code to do exactly that, with fewer errors than I'd make on a first try. It'll also generate unit tests for it. Writing a text specification then reviewing some code for me is way, way faster than writing all that same code by scratch.
Are you working on code that's heavily coupled to other code, such that you rarely get to write independent logic components from scratch, and IP restrictions prevent you from handing large chunks of existing code to the LLM?
Unlike with numbers, day to day language needs to convey a larger variety of concepts hence why logograms are still needed. Much like how more advanced math requires different symbols and operators that are akin to mathematical logograms to convey additional concepts beyond the fundamental quantities of 0 to 9.
Alphabets are often linked to sound, and it would be a tremendous challenge to create an "alphabet" analogue that links to a set of fundamental concepts that you can somehow rearrange to form higher level concepts and can still be universally understood without it linking to sound.
I wonder if this might have been influenced by how the "gamer" demographics has expanded due to mobile gaming too.
It used to be that "gamers" were either console or PC specific, and these platforms not only naturally allowed for more complex games but also attracted a specific demographic who prefered games with more challenge, strategy elements, and planning. But now "gamers" is much more diverse. The casual mobile afk gaming demographic has eclipsed that of the more serious and old school gamer persona, and by design and nature mobile games tend be much less strategic.
Truthfully, the proliferation of crimes like this is a pre-condition of systems collapse because it slowly unravels a lawful society. Laws that are not enforced are useless, and behaviour like this would eventually permeate into mainstream culture. After all, why should I pay if others can steal without consequences? Not to mention the negative socioeconomic impacts others have already pointed out of forcing store closures or driving up costs which eventually just feed the loop.
Whether or not one is sympathetic towards the people practicing this, the key issue that is becoming apparent is increasing apathy towards basic civilizational social contracts. This will be bad for everyone in the long term.
I think you might have it backwards. Petty theft doesn’t cause systems collapse - it’s the result of systems collapse. No matter how normalized, if there’s even a rare chance of consequences most people who can afford to pay will pay. That these crimes might be increasing (stats that prove that and aren’t junk are “hard to come by”) is evidence not of the theives not holding up their end of the social contract - but of employers, businesses, and governments not holding their end up.
> Petty theft doesn’t cause systems collapse - it’s the result of systems collapse.
That this is not the obvious conclusion is alarming. We've been talking about the costs of poor social safety nets and high income inequality for decades.
I know of a few European countries that have very strong safety nets, and this behaviour is significantly on the rise there also (first hand experience). There has been a slow-motion withdrawal of enforcement for 'small' crimes which I think leads to larger ones ala the 'broken window'. I think part of this is because it is so arduous a process to convict & incarcerate someone.
Broken windows policing is a garbage + debunked theory about crime, btw. It “worked” in NYC where it was famously deployed, because we were in the middle of massive decrease in violent crime. Not because it worked.
There is the broken window fallacy, and the broken window fallacy fallacy:
> A 2015 meta-analysis of broken windows policing implementations found that disorder policing strategies, such as "hot spots policing" or problem-oriented policing, result in "consistent crime reduction effects across a variety of violent, property, drug, and disorder outcome measures".[36] As a caveat, the authors noted that "aggressive order maintenance strategies that target individual disorderly behaviors do not generate significant crime reductions," pointing specifically to zero tolerance policing models that target singular behaviors such as public intoxication and remove disorderly individuals from the street via arrest. The authors recommend that police develop "community co-production" policing strategies instead of merely committing to increasing misdemeanor arrests.[36]
if you read the abstract of the 2015 study you’re quoting from:
> The strongest program effect sizes were generated by community and problem-solving interventions designed to change social and physical disorder conditions at particular places.
That doesn’t sound like broken windows policing. That sounds like fixing actual broken windows.
> I think you might have it backwards. Petty theft doesn’t cause systems collapse - it’s the result of systems collapse.
It is probably a positive feedback loop: some system collapse causes some petty theft, which causes more system collapse, which causes more petty theft...
> hat these crimes might be increasing (stats that prove that and aren’t junk are “hard to come by”) is evidence not of the theives not holding up their end of the social contract - but of employers, businesses, and governments not holding their end up.
Why not both? Much of the shoplifting in my area is for drug money (at least the blatant kind you get to see up front). Now, one could claim employers are wrong for not letting someone addicted to fent work, but maybe not. Definitely how they got to that point is some fault of society, however.
It's by far not the most people who drive this. Non-enforcement of rules gives the worst individuals free reign. Enforcement works on the margins, and in this case the margins is basically all that matters, at least UNTIL the collapse.
With ML we might have efficient ways of detecting enforceable scenarios, but I wonder if the tradeoff of cameras everywhere and a turnstile when entering a grocery market will be worth it.
Soon we might see conglomerations of businesses pooling their data together for global customer credibility scores.
"Laws that are not enforced are useless" - Worse than useless because it's a license for the justice system to discriminate. "We don't normally enforce this law, but we're enforcing it against YOU right now."
Also worse than useless because those who normally follow the law begin to build immense frustration and loss of desire to make their own good choices.
It is called "prosecutorial discretion" used by prosecutors to favor the powerful, wealthy, connected folks. Right now, this discretion is extended to favor ordinary folks belonging to right categories.
I think these processes are also driven by the fact that more and more people are living at or below the poverty level (due to an evert getting bigger gap between the rich and the poor) and reduction in social-economical mobility. The 'American dream' is no longer a reality in the USA. When large parts of socity live at the bare minimum and have no hope for improving their situation except through crime, it is to be expected that crime rates will increase.
There was an education thread the other day. No one much said it explicitly, we all danced around it... but it was mentioned that some states aren't bothering to teach math anymore, because some "groups" always failed. It was supposed to be some social justice thing.
I'm no racist. None of the children who were failing were incapable of understanding algebra. But there is a toxic culture that interferes with them wanting to learn it and behaving in ways that allow them to learn it.
It is asinine to say that they have no hope of improving their situation, when the same people are sabotaging themselves from childhood on up.
Sabotage piles up. When those retail stores inevitably flee those regions, there will be even fewer legitimate jobs to go around. There will be so-called "food deserts". And then that too will be blamed on late-stage capitalism and so forth.
Any group or culture will possess a number of progressive and regressive tendencies (with specific reference to education). For every enlightenment there is a Spanish inquisition.
If a particular group holds too many regressive tendencies they're going to slide backwards, and unfortunately we are not culturally permitted to point out that their cultural beliefs might be at fault. A non-US example to calm your pitchforks: The 'Travelling Community' in Ireland/UK (a culturally distinct group, though not genetically distinct) believe that formal education is not a worthwhile use of a teenagers time.
Watching everyone blame a complex system failure on their pet theory reminds me of the parable of the blind men and the elephant.
I'll say that after the previous decade or so of watching oligarchs treat the law/decency as a non-entity and seeing zero repercussions, I've lost most of my belief in the law myself. While some might take that as a free license to do whatever they can get away with, personally it's made me evaluate my own morals and try to live as closely to them as I can, and I feel good about myself for living my values.
But I often see the attitude here that I should be a mercenary and work for the highest bidder no matter the social cost of the company that pays the best. I can't help but wonder - what if the best employment available to me was as a member of a retail theft gang?
Is greed good? I don't think so, but it seems to be a core value these days.
> Truthfully, the proliferation of crimes like this is a pre-condition of systems collapse because it slowly unravels a lawful society
Truthfully, the proliferation of crimes like this is a symptom of systems collapse because it evidences the unravelling of a lawful society (or, at least, the pre-existing social order, which may not have been generally “lawful”.)
Just imagine how much more abundance we would have if we didn't need to build all these security measures to deter bad actors. We are all paying more because of these thieves.
Learn what deadweight loss is. These costs are born by everyone, what percent of the cost is born by which parties is determined by the competitiveness of markets. Normal, law abiding, people are absolutely bearing the cost of bad actors.
We bear the cost in the form of needing to pay for police, surveillance cameras, security guards, security software, the justice system, lawyers, and jail cells. The money lost from each stolen good is spread across all other goods in the form of price increases, etc.
The shareholders who contribute nothing of value and yet still require more and more wealth be extracted from retailers? Yeah, we’re definitely all paying more because of those thieves.
Learn what deadweight loss is. These costs are born by everyone, what percent of the cost is born by which parties is determined by the competitiveness of markets. Normal, law abiding, people are absolutely bearing the cost of bad actors.
We bear the cost in the form of needing to pay for police, surveillance cameras, security guards, security software, the justice system, lawyers, jail cells. The money lost from each stolen good is spread across all other goods in the form of price increases, etcetera.
> shareholders who contribute nothing of value
They are shareholders because someone presumably at some point contributed something to society which resulted in them earning money to invest.
> someone presumably at some point contributed something to society
Even if that is true, which it’s not: in a shockingly high number of cases it is actually harming society that got people money. Even if it is true in a specific case - this premise implies that someone at one point contributing something to society entitles them to drain money from everybody else forever.
We all understand how we bear the cost of theft, but i don’t believe that theft is caused by bad people - it’s an inherent byproduct of a system that REQUIRES poverty to function.
Worst justification ever. Is this satire? History is full of legal things that were theft. Take your pick: the british crown taking land from whomeever. Slavery, chattel and indentured. Redlining. The potato famine.
Stealing from those who can afford it, in order to meet your basic needs (housing, food, etc) is not wrong. While taking the fruits of other’s labour because you have money, when you know the result is others not being able to afford their basic needs, is wrong.
> imagine how much more abundance we would have if we didn't need to build all these security measures to deter bad actors
This article has plenty of information about the fact that companies aren't building in security measures because they don't want to pay.
Or to put it another way: think about how much better shopping experiences we'd all have if stores weren't constantly understaffed. But they are, and so shopping sucks and shoplifters proliferate.
Imagine thinking "having relationships and growing like nearly every human being has in the history of the species" is in any way comparable to "stealing a bunch of shit from Target."
> After all, why should I pay if others can steal without consequences?
I think that cuts to the heart of the basis of morality.
Do we need the threat of consequences in the afterlife, if there are no immediate consequences to the self? Are we allowed to do whatever we want now, if we can absolve ourselves of consequences in the afterlife?
And if you don't believe in an afterlife, Do we recognize that these actions can hurt others? Do we have empathy for those that were hurt and try to make them whole? Do we more strictly punish those that abuse that empathy? How about people who simply don't feel any empathy for others, how is that dealt with?
I don't have any good answers for these questions.
> if there are no immediate consequences to the self
Fixing this is the answer, isn't it? Make sure there are consequences. Watch any video of people riding bikes into a CVS and filling trash bags with goods. The employees are scared to do anything. Not only do they not try to stop them, many times they're actively preventing customers from trying to stop them. A minority of the time one of the other customers, who has probably seen this bullshit way too many times, tries to do something.
How about a law saying "if you are an employee of a store charged with loss prevention, security, or anything like that, you have qualified immunity when trying to stop an active theft?" The idea that I can run into a store and start stealing things, then turn around and sue the security guard AND the corporate store for punching me in the face while I'm doing it is ludicrous.
We already have laws that if someone is killed in the commission of a crime you're committing, it's the same as if you killed them. It's not unheard of to say that if you're committing a crime, the laws of liability change.
> How about a law saying "if you are an employee of a store charged with loss prevention, security, or anything like that, you have qualified immunity when trying to stop an active theft?"
Qualified immunity from what, though? Pointing and yelling "thief"? Blocking their way? Laying hands on them to detain them? Assault and battery? Deadly force? You can't just allow people to do anything to stop someone from stealing a $5 tube of toothpaste. Where do you draw the line?
Also, what if the employee himself/herself is the one that gets injured? Who has qualified immunity from being responsible? The store?
> > We already have laws that if someone is killed in the commission of a crime you're committing, it's the same as if you killed them. It's not unheard of to say that if you're committing a crime, the laws of liability change.
At least to your last point, if you're stealing from a store and someone gets hurt trying to stop you, you should be responsible whether you hurt them or not, whether it was intentional or not. They slip on something and break their shoulder while they're running after you? No different than you assaulting them and breaking their shoulder directly. Again this isn't the case (at least not to this level) but I think it would be a good change.
I agree with your other point that it's not cut-and-dry, and honestly probably will vary based on state and municipality. There are certainly municipalities where you can do a lot, up to what would otherwise be assault or battery, to prevent theft. And there are some where you can't. But I think in general being a little more permissive with what is allowed, especially in these large corporate stores in large urban areas where this smash & grab type of theft seems to be more prevalent (or maybe just reported on more?), might be a good thing.
I don't think anyone noticed themselves, nor has anyone listened when I've mentioned it previously, so I'll repeat:
Sometime decades ago, the police just stopped policing.
I don't have all the details. We can infer and speculate a bit that it didn't happen at the same time everywhere, all at once. But it can't have been much later than the 1980s or early 1990s, or I would have noticed.
The only thing that the police do with any earnestness or initiative are drug busts, and this is plainly because when they do drug busts they get to keep the proceeds whether that is cash or fast cars (that get new squad car paint jobs). They do the other things (local) politicians want, which is some ticket quota revenue collection once in awhile. They continue to begrudgingly take stolen car reports, because if more than a few voters complained they weren't doing that, eventually some politician somewhere would lose an election or be forced from office.
But there are no repercussions for refraining from any other sort of police work. Those scenes in The Wire where they're trying to convince rape victims not to report it (in that scenario, so their stats would look better)... not just narrative device. When we hear about them not going after a school shooter killing kids in classrooms, and we're straining our brains to explain it and the only thing that comes to mind is cowardice? Well, I've got a simpler, though no less bizarre explanation... it just wasn't in their job description.
This isn't accidental or temporary or bad management. You have Police Chiefs going on Twitter and chastising the public for believing that they should be able to park cars and not have their windows smashed and contents stolen. It is a matter of policy that they no longer do policing. This is backed up by case law that makes it official that they have no duty to the public to protect them from any crime.
Most of all though, you have a public that is somehow willfully ignorant of it, thinks that maybe they're just doing a bad job but doing the job, and if they're patient new people will be hired and maybe they'll start doing it moderately well again. It's baffling.
> Most of all though, you have a public that is somehow willfully ignorant of it, thinks that maybe they're just doing a bad job but doing the job, and if they're patient new people will be hired and maybe they'll start doing it moderately well again. It's baffling.
Cognitive dissonance. I think this is a symptom of a slow western civilzation collapse
Sure, but that is a consequence of allowing police forces to devolve into extortionate gangs, with no responsibility for responding to crimes and plenty of latitude to commit their own, with impunity.
Some of the largest protests in US history were in opposition to this state of affairs and they were met with violent repression and police being rewarded with even larger budgets WITHOUT the requirement that they actually enforce laws.
It shifts the Overton window on what behavior is acceptable, and like the GP said it creates a feedback loop where stores close, prices increase, more people are forced to steal, etc etc. It's pretty widely accepted that lawlessness, even at the "petty" end of the spectrum, is a precursor to larger deterioration of society and whatever civilization/empire you're talking about specifically.
> It's pretty widely accepted that lawlessness, even at the "petty" end of the spectrum, is a precursor to larger deterioration of society and whatever civilization/empire you're talking about specifically.
I saw a generic bottle of mayo go for $8.99. I say steal more. The commodification of food is an evil in and of itself; our society is based on profit over people. Let this shit collapse. This is a symptom of economic systems failing people. The problem doesn't lie solely on individuals.
The actions of others are unrelated. I cannot steal just because someone else has become successful even if through stealing. If I steal the billionaire is not wrong, just better at it than I am.
But why? If the person with the overwhelmingly disproportionate amount of power is unwilling to uphold and support the social contract, why is it incumbent upon the disenfranchised to regard the contract as valid or even existent?
While the findings are quite believable and corroborates some other similar studies where increased sugar intake for kids may lower executive function/cognition, this study relies on executive function assessments reported by the parents which doesn't feel like it would be very reliable to me.
Furthermore, while they account for diet in their covariate analysis, it's not very detailed or granuar so it doesn't account for other sources of sugar that these kids might be having (the authors acknowledge this). Based on this study it's hard to conclusively say whether it's the sugar that negatively impacts cognition or other ingredients, or vice versa whether kids with poor executive function prefer sweet drinks. Probably still a good idea to limit refined sugar intake for your own kids though.
Many of us had grandparents who told our parents not to give us sugar because it would make us rambunctious.
The counterargument was that sugar intake often accompanies life events that are themselves overstimulating to kids, so it's more of a correlation than causal link. I was expecting this study proved the original assumption but your read suggests that it's more of the same.
That said, I have discovered that there's a particular food coloring that does tend to make me irritable, and children's candy is usually loaded with it. Whenever I can find the 'naturally colored M&Ms' in the bulk foods aisle I usually grab a little of it.
My parents experimented extensively with my childhood diet's impact on my ADHD, and found that Red 40 made me bounce off the walls. The effect was dramatic, even compared with eating sugary non-Red-40 foods.
I recall a friend was relieved when Red M&Ms went away, because she had an anaphylactic reaction to them. Though I guess the real reason was the old color turned out to be a carcinogenic. It was years before they added them back, with Red 40 instead.
This was my understanding of China. They do use a lot of coal, but it is because they are having to ramp up energy usage with coal-fire plants to keep the factories going. They do have long term plans to phase out coal, but when energy needed to keep up production or add new production (perhaps offshoring of jobs and factories to China), they have the build the coal-fire plants. Last I looked, they had about 30 nuclear reactors planned/in-construction/still at the drawing board.
You can check out our super rough version here, been building it for the past two weeks: gateway.aci.dev