I should mention I made that one for my research/stats workflow, so there's some specific stuff in there for that, but you can prompt chat gpt to generalize it.
I mean, damn. Are terms like “executable oracles” and “hermetic boots” related to your domain, or are you using these as terms of art for an agent? Oracle being a source of truth, hermetic meaning no external dependencies or side effects - definitions in furtherance of your request for concise language. Would love to understand more.
This prompt is for scientific research. In general my goal is to instruct the agent to build as much validation scaffolding as possible, so rather than holding its hand I can just give it a series of concrete hurdles and tell it not to come back until they're met. I don't want it finishing the basic tasks and coming back to me saying the app is "production ready," I want to come back after a few hours to the agent having "proven a spec" with a demo or a paper that I can iterate on.
The errors make sense. You can't put a comma-separated initializer into a macro... What even is the symbol `entity`? It's not even clear to me what is meant by that, he doesn't define it anywhere.
edit: Looks like he updated his header since I first tried to compile. Now it works fine but the header looks significantly more complicated.
here's a script that will highlight previously visited squares as red. as a nice bonus it replaces the instructions with more accurate ones
// Get all square elements
squareElements = [...document.querySelectorAll("[data-testid='white-square']"), ...document.querySelectorAll("[data-testid='black-square']")];
// create array of elements that have been landed on
// these need to be set to red every time because the board resets every move
let modifiedElements = [];
observer = new MutationObserver(function (mutations) {
mutations.forEach(function (mutation) {
if (mutation.addedNodes.length > 0) {
mutation.addedNodes.forEach(function(addedNode) {
key = addedNode.getAttribute('data-testid').split('-')[1];
if (!modifiedElements.includes(key)) {
modifiedElements.push(key);
}
for (let i = 0; i < modifiedElements.length; i++) {
el = squareElements.find(e => {
return e.getAttribute('data-squareid') === modifiedElements[i]
})
el.style.backgroundColor = 'red';
}
});
}
});
});
// observe all square elements
for (let i = 0; i < squareElements.length; i++) {
observer.observe(squareElements[i], { childList: true, subtree: true });
}
// replace instructions with more accurate instructions
instructions = document.querySelector('section p')
instructions.innerHTML = 'Move to the square indicated below the board without moving to a square the queen can capture and without capturing the queen. Once accomplished a new square will be indicated. Repeat until all possible squares are done.';
I was thinking of developing a website that helps you analyze your games vs naroditsky's games. let me know if you have any ideas on what would be useful.
I would love that - been thinking about how to automate Youtube's generated closed captions from positions I'm in based on positions he's in. I typically play Caro-Kann and Danish Gambit so it seems likely I probably see common positions in the first half-dozen moves or so.
It looks like they're solving a different problem.
Lots of analog credentials do exist that should be digitized (the SCUBA one is a great example that I've personally encountered), but I'm talking about a replacement for traditional, outrageously expensive 4-year Universities where a fraction of what you pay for is directed toward what you'll do for work - but more importantly, folks that wish to career switch from whatever they'd doing to a more technical field but lack that 4-year "degree".
