in case anyone is interested in having their Agent manage the MCP servers of the services they are using, they can also check out the companion repo here:
https://github.com/Wpnx330/pharos-discovery
14 karma · joined August 12, 2026
I think only way we really differ is that I don't believe in preventing others from seeing AI writing because I saw it and decided it doesn't have value for me. Instead, I leave it to them to decide. I let the "great filter" of quality do its job.
My point from the start was. People flag content generated by AI as bad, when they don't even read the content themselves to determine if it is bad. It's a bias reaction. they are trying to curb the tide of AI use. The reality is, they can't. What they can do, is decide for themselves if what they see is valuable or not, and react accordingly. the truly bad stuff will fail. the good stuff will succeed.
Though, a few hours ago my original belief was it may take a little while before people who insta-reject AI writing are fooled into thinking the writing was done by a human... but it would seem based on other comments in this thread, people already can't tell what is written by people or llm.
That seems far more concerning. based solely on the other comment, I bet somewhere out there, people who use no AI at all and are getting labeled as AI writers and auto blocked by those individuals with their anti-AI guard up.
to be completely clear, i posted an article i thought was very relevant to things i care about and someone flagged it as ai drivel. but it was in fact valuable information worthy of being shared.
so i did some searches to see if people are actually against ai being used in writing and found this article. posted here to see peoples opinions. So, far id say its not so easy. some are okay with it, some are against.
People once thought cars we dumb toys for rich people and completely impractical forms of transportation given the state of the roads and economy at the time. Look at society today.
People also once thought the internet was a fad. And, steel-manning your point. we had a dot com bubble. never-the-less the whole world runs on the internet today. Arguably for the better. Many innovations around the world have been due to digitally shared information.
Just because some people try to use it to get rich, doesn't mean the world wont be the better for it. I'm just looking at the big picture. sure, some people will use AI poorly... their businesses, and ideas, and blogs, etc.. they wont last. some people will use AI well, and they will. Either way. people are going to use AI, for everything.
I think the answer is: people have to help improve their own lives, and when AI can help them do that legitimately, then I think they will.
AI is used in literally everything in society at this point. You drive to work, AI enhanced navigation, or even self-driving cars. You take a photo on your mobile phone, AI enhanced images. You screw up the text in your group chat with your friends, AI enhanced text correction. You browse for products in your favorite online store, AI cross-product suggestions. The Ads you see on said favorite online store, AI selected. The songs and the order they are played on your favorite music app... likely AI selected.
every bit of text, video, audio, and still imagery has, at some point in the chain, likely been touched by AI. And, if it hasn't today... in time it will be.
Every facet of society is being optimized via AI. Maybe to many people's chagrin. Never-the-less, the AI turns. So, my take... rather than resist the inevitable, carry on doing what you've always done. filter and choose the better product, service, content. The genuinely bad will fail, AI generated or not.
The trust crisis is more than just who gets to run what models. It about what those models can and actually do accomplish. The average person isn't actually doing anything wowing with models. let alone concerned they might hack a business or conspire with other models. They are still trying to figure out how to make those models simplify their job roles and make their businesses more money. The promises are high, the delivery on those promises not so much. Instead the focus should be on shipping things that actually work and provide business value. Model intelligence and capabilities have arguably already passed the threshold required to do it. they just needs to be utilized.
We need lockfiles, audit commands, and signed provenance, just like we got with npm.
local cleaned models would become even more important. No longer just to cut the token bill, or allow enterprising individuals to host adult content sites as a business.
When I am in hitting the gym, Spotify, pick rock song from the 2ks and hit radio mode.
With an AI agent, I can literally brain dump to a prompt and say turn this into a great prompt that includes a goal, implementation details, and an acceptable quality target. Then just say, okay now execute. Try that with a person and you'll quickly find, you've got no leadership skills.
I remember the 4-hour work week through line. outsource your work to someone else in a more economical region. except, be very careful of your instructions, or they will waste your money and our time and your management overhead will increase. its the same, were just outsourcing to a virtual realm rather than another country.
Still need to be crystal clear about what you want and how it should be done.
1. Impeccable (github.com/pbakaus/impeccable) — a CLI that scans HTML/CSS/JSX for 59 deterministic anti-pattern rules. It catches the actual visual tells: zero-offset dark glows on text, 1px borders paired with 40px shadow blur, Inter/Roboto/Geist as the default "tasteful" font.... etc.. I wont list all 59 here.
2.Taste Skill (github.com/Leonxlnx/taste-skill) — preventive rather than detective. It's a set of SKILL.md files you load into your coding agent (Cursor, Claude Code, etc.) before generation so the slop never gets created in the first place.
---
everyone knows about Em-dash overuse, but there are other tells like the dark background offset glow, thick boarders, flat display style, over use of techy color schemes like navy blue (especially with gradients) and emojis in headers as large style iconography. there are like 5 common fonts ai's use etc.. the key is, your llm is going to reason that these are all good safe choices and even when you tell them to be different they will give you great sounding logic as to why to use them... which is exactly the problem. every llm 'reasons' themselves into using the same choices so the sites all look the same.
I used to have: $200 Claude-max $10 github copilot $20 chatgpt $20 Cursor
---
then, I spent roughly $5000 in tokens for a project which turned out to be a dud. Figured, inference wasn't going anywhere as its now 80% of my day. So, I built a 4x DGX Spark cluster and put GLM5.2 on it. Then put a few qwen models for vision and embedding purposes on my original DGX Spark and completely went local. Tron, my Hermes agent, uses agent storming to orchestrate openCode agents and hermes subagents to handle most of my daily tasks.
For fun, I reworked a public Github vLLM Dashboard to track my usage and cloud cost avoided. original dashboard was by github.com/niklasfrick but I extended it. github.com/Wpnx330/spark-dashboard
---
For my day job I spend about $1000 + $20 a month in cursor costs.
We don't need large collections of assorted tools in one MCP server. Not trying to detract from the effort OP made... but we need a mechanism for finding, installing and managing our MCP servers effectively. Then monolithic collections aren't needed.
I believe collection servers are a band aide for the real problem. Useful for the individuals who built them, and often times not for too many others..... which is why I am building management tools for MCP servers. Hopefully to minimize the need for doing this. Because OP has filled a gap in the path available to them.