salil999

Hi HN,

I created a small app that acts as an AI tool organizer across a few popular providers: Claude Code, Cursor, VS Code and OpenCode. It has support for: - MCP Servers - Skills - Rules - Hooks - Subagents - API Credentials

All of the data used and stored by the app is stored on your $HOME directory under ~/.ai_tools_manager. I'm new to open source so I'm not sure if I "did" it correctly but open to feedback!

Reminds me of WarpStream: https://www.warpstream.com

Similar idea but you have the power of S3 scale (if you really need it). For context, I do not work at WS. My company switched to it recently and we've seen great improvements over traditional Kafka.

Can't say it's a perfect solution but one way I've tried to prevent this is by wrapping secrets in a class (Java backend) where we override the toString() method to just print "***".

I hear this on almost every bad post about Meta. No they are not cooked. They still generate tons of profit and their user base is one of the biggest in the world. They're not going anywhere any time soon.

I got around this by just editing the HTML. Worked like a charm

I think you can already do some of this in iOS with the Shortcuts app

I'm confused as to why this needs to be standardized? OTel makes sense because multiple applications can consume metrics and logs. Feature flags are typically contained within a company and they typically don't need to be transferred across to other places.

I would also take a look at exa: https://the.exa.website/

It's been my ls replacement for a while now and it's very customizable!

The claim I've heard is that you're essentially feeding your own knowledge for free into a proprietary system that can be used to generate cash for whatever corporation owns that system (i.e. ChatGPT from OpenAI). I think it's pointless to redact content for this purpose as well but clearly some people have strong takes against AI training.

> If Meta starts being all open and generous about their core assets

I think they are pretty open about it?

- https://www.meta.ai/ requires no log in (for now at least)

- PyTorch is open source

- Various open models like LLama, Detectron, ELF, etc

- Various public datasets like FACET, MMCSG, etc

- A lot of research papers describing their findings

In the example they have

> const urlstring = "this is not a URL";

> const not_a_url = URL.canParse(urlstring) && new URL(urlstring);

Why would you need the latter part of the condition? Couldn't it just be this?

> const isValidURL = URL.canParse(urlstring);

I do see on the bottom left:

Log in to save your conversation history, sync with Messenger, generate images and more.

It's specifically advised NOT to use ObjectId for distribution. This data type has a notion of time embedded in it which results in monotonically increasing ObjectIds. Use UUID or a hashed version of the ObjectId

Why is this a reason to stop using Google?

Yeah but what about something that won't make my wallet cry?

Most browsers use a SQL database for storing data like this.

If you're in the Apple ecosystem then Safari has had this for a while.

Why?

I see this Jepsen link posted all the time. People: PLEASE don't use outdated software. MongoDB has made mistakes and they are public about their data issues on https://www.mongodb.com/alerts. MongoDB 4.2.6 is old and I believe it's approaching EoL based on https://www.mongodb.com/support-policy/lifecycles

I'm not going to push for you to use MongoDB but am merely trying to provide some context around that Jepsen analysis.

For my own knowledge, if the options were between using SMS for 2FA or not having 2FA at all then what is better? I've heard mixed things about this.

It's pretty easy to start with. MQL is also pretty easy to understand + MongoDB kinda makes it fun.

Note: I work at MongoDB

But WhatsApp has had it for a while, no? Or at least I think it's a solved problem looking at both WhatsApp and iMessage

I am curious why it was very difficult for messenger to implement E2EE by default. iMessage has been doing it* (with a giant asterisk because there are flaws in the way Apple markets its E2E on iMessage but in general it has something there). The implementation they're describing on the blog seems very similar to iMessage (keys stored on a server so that syncing is possible

Asides from iMessage, they pretty much have most of this working for WhatsApp from the perspective of the user. The challenges they've mentioned seem like they've mostly been solved in WhatsApp? I could be totally naive here though.

> If you want me to trust your system, show me the complete source code

I think this is a bit extreme and not really plausible for something like messenger at its scale.

Whatever opinions I have against Apple I just cannot deny their ability to not only innovate but also deliver every year. I wonder just how far into the future they're working on stuff internally at the company. Well done to this company.