It's always worked in Firefox for me? I also like this Firefox extension (https://github.com/WorldThirteen/youtube-watch-later-shortcu...) which allows mapping a keyboard shortcut for quickly adding YT videos to the "Watch Later" playlist.

Thanks for sharing your state. I just tried again with troubleshoot mode, it works ...

Then it seems to be in conflict with some addon I'm using.

I found that it is caused by a userscript running below code.

    // version 1
    let body = document.createElement('body')
    body.innerHTML = document.body.innerHTML

    // version 2
    let body = document.body.cloneNode(true)

Either version can trigger the bug of the pop up menu now showing up. The menu is able to show up if doing early return to skip that code.

This is wired, that code should not be mutating existing DOM.

Update: Below version of sandboxed clone doesn't break the youtube popup menu.

    // version 3
    let parser = new DOMParser()
    let doc = parser.parseFromString(document.body.outerHTML, 'text/html')
    let body = doc.body

$900 for this is laughable. Shame to see OSS tainted in this way.

Paid tier for a widget? The web is truly dead.

The epidemic of clearly LLM-generated READMEs has to stop. The emoji bullets are killing me.

Observation: If your blog post starts with a clearly AI-generated abstract, why would I be motivated to read the rest of it? It's unfortunately indistinguishable from AI slop.

> delves

Dead giveaway :D

if that was the only one … :)

These are extremely common these days. Here are a few I've collected over the past few months:

- [files-to-prompt](https://github.com/simonw/files-to-prompt) (from the GOAT simonw)

- [code2prompt](https://github.com/mufeedvh/code2prompt)

- https://gh-repo-dl.cottonash.com/

- [1filellm](https://github.com/jimmc414/1filellm)

- [repopack](https://github.com/yamadashy/repopack)

- [ingest](https://github.com/sammcj/ingest)

What makes yours better?

I found this tool (repo2file) helpful for my workflows - quickly giving context for questions to my local LLM about my working (small) repo in the terminal. Until I saw this post, I wasn't aware of any of those.

What makes his better? Since you're asking, I tried these and here's my verdict:

- [files-to-prompt](https://github.com/simonw/files-to-prompt) (from the GOAT simonw) --> There's no option to specify files to include, must work backwards with ignore option

- [code2prompt](https://github.com/mufeedvh/code2prompt) --> It always puts the output to the paste buffer even if you specify output file

- https://gh-repo-dl.cottonash.com/ --> There's no CLI

- [1filellm](https://github.com/jimmc414/1filellm) --> Many dependencies and complicated setup(have to setup GitHub access token which I've never done)

- [repopack](https://github.com/yamadashy/repopack) - [ingest](https://github.com/sammcj/ingest) --> haven't tried these yet, but they actually look promising...

I tried [repopack] and it did a very good job (y) Simple installation too.

I tried code2prompt but the interface has a quirk—even if you specified an output file, it will always put the output to the paste buffer.

Which one of these you find the best? It’s quite tempting to write one myself for something as simple as this.

I think most of these projects are bit overkill, OG poster's repo is about what I'd do myself.

How did you compare the above? I'd love to see a "clear winner".

sammcj of sammcj/ingest reporting in! Funny to see my little tool pop up in comment threads.

He misinterpreted my post as saying "Use LangChain in the Browser".

And so now your interpretation of things is that I misinterpreted his misinterpretation. Great work and thanks for your helpful insights.

This only happens because you’re using Cloudflare DNS. Any other DNS will load the site just fine.

what the hell? you're right, apparently. I'd set my router to use 1.1.1.1 a long time ago and forgotten about it.

why does using cloudflare mess it up? how does that even work?

The people running archive refuse service to queries served by cloudflare dns.

It’s very strange. I went down this rabbit hole a few months ago and this HN comment (https://news.ycombinator.com/item?id=19828702) was the best summary I could find of the situation.

TLDR: archive.is doesn’t play nice with its DNS results and Cloudflare refuses to fix it on their end.

This is the fault of archive.is — they’re using an outdated DNS load balancing approach that is inherently not robust. They’re mad at Cloudflare specifically but the reality is that there are many other similar failure scenarios.

The mature thing to do would be to switch to something like anycast IP routing, which is robust against issues like this.

I suspect the “problem” isn’t actually that big a deal to begin with, and the archive folks are making a mountain out of a molehill for philosophical reasons.

In other words they hate Cloudflare and will find any excuse to start a fight.

Why the problem with their DNS?

You shouldn't. If Kagi doesn't offer enough value for you to be worth that extra friction, don't use it. That's the free market at work.

I would say the value is definitely there, when you consider the top links on Hooli, I mean Google, are frequently malicious. Though the biggest thing to me for Kagi is the ability to uprank and downrank various domains, so I can ensure the results I am actually looking for tend to be on the top, and sites which don't cater content I need personally filter downwards.

It’s open source mate. Just dig around in the source code for long enough. It’s not pretty, but you can usually figure out how they’re abstracting.