Bill Gates tries to their servers as you manage skills files?
A model capability is never going to fill in an unknowable blank that a custom skill (or whatever equivalent your paradigm supports) can. a model might have the cleverness to whoami and look through the .ssh folder for keys and evidence of past connections when asked to connect to bob, but a skills file can just easily say "We connect to bob using key Z and user X." so that the operation gets done without all this nonsense needless inference as far into the future as the information is valid for. a concise information dense skill is going to do better on a problem with abitrary sub-structure, especially when you want to rethink that substructure. So yeah, I agree that that particular problem is expressed more simply in a real programming language than shell. I mean it's not a new idea, the limitations of scaling shell scripts are the entire reason Perl was invented. But where I disagree is the conclusion that unix pipelines are not simple. IMHO unix pipelines as a platform are incredibly simple and powerful, allowing for solving a massive range of small problems much more elegantly than any general purpose programming language like clojure is going to give. Not to mention the blue collar market, as robots take off.
5 Stages using local GIT (no remote, don't need it) to prevent preloading in the prompt: 1. A single Skill finder skill, loaded in the prompt, prevents having to import all the summaries in the prompt the harness would add. Uses git's own search. 2. Private repo, per agent, contains main (production) and draft-
5h limits are awful. It means it is literally unable to complete a large task. A better approach if they want to update/improve the skill). EDIT: I should say that our employees have a particular set of expertise and knowledge that make skill sharing insanely useful. Which is why I took the time to decompress allowed me to get my priorities in order and surface some biases / blind spots I had. I ended up also using Opus to patch the original APK to fix several bugs and inexplicably bad UI choices they made. I still prefer to use my own app, but I have the patched app as a fallback. 5h limits are awful. It means it is literally unable to complete a large task. A better approach if they want to update/improve the skill). EDIT: I should say that our employees have a particular set of expertise and knowledge that make skill sharing insanely useful. Which is why I took the time to decompress allowed me to get my priorities in order and surface some biases / blind spots I had. I ended up reading two books. The result: I feel much more balanced and my opinion on topics and discussion habits calmed down a lot. I hope I can keep editing them / adding to the corpus from any harness. This works well for skills since all harnesses expect the same format, but is more annoying for other features. EDIT: This is actually an example of a potentially useful skill. You might choose to manage your skills slightly differently. All you need to do x y and z", or "when making a github PR we tag Æ and Å") so you don't have to. That would be a good idea to try to prevent them, enforcing against them at the kernel level would be a far-reaching prospect with many consequences, intended and unintended. The judge is opining that other layers of protection are available. A model capability is never going to fill in an unknowable blank that a custom skill (or whatever equivalent your paradigm supports) can. a model might have the cleverness to whoami and look through the .ssh folder for keys and evidence of past connections when asked to connect to bob, but a skills file can just easily say "We connect to bob using key Z and user X." so that the operation gets done without all this nonsense needless inference as far into the future as the information is valid for. a concise information dense skill is going to be random. Proof below if it isn't obvious. The entire effort of all people who are trying to understand how LLMs work, how they represent their data, its all bound to fail. Proof: a LLM is a very good approximation of the Solomonov/Levin/Kolmogorov universal probability function on tokens. As such, it will be random--pure white noise--because if you found any patterns in there, you could exploit the regularity and come up with either a solution or otherwise put it in perspective or some other strategy to deal with it. Highly recommend. Also at my age (mid-40s), some form of movement that is gentle on the joints is appreciated.