kinduff

Wrote about this last May, had this vision of a store that looks very much like this. Maybe you enjoy the read, here is the link: https://kinduff.com/2025/05/28/when-ai-sells-you-what-you-wa...

We want an oracle, not a therapist or an assistant.

Would love to add models direclty. And don't worry, we will figure it out!

Oh my, its been a minute since I played this game. I remember playing it as a teen.

I like how they plugged the MCP support into their existing tools. Pretty smart!

I found some papers [n] about this. And I think the answer is yes, the format matters asnd hence the representation.

I wonder if the author would be willing to try with another representation.

[1]: Does Prompt Formatting Have Any Impact on LLM Performance? https://arxiv.org/html/2411.10541v1

[2]: Large Language Models(LLMs) on Tabular Data: Prediction, Generation, and Understanding - A Survey https://arxiv.org/html/2402.17944v2

In Factorio's paper [1] page 3, the agent receives a semantic representation with coordinates. Have you tried this data format?

[1]: https://arxiv.org/pdf/2503.09617

Do you think the performance can be improved if the representation of the level is different?

I've seen AI struggle with ASCII, but when presented as other data structures, it performs better.

edit:

e.g. JSON with structured coordinates, graph based JSON, or a semantic representation with the coordinates

The LOC are really impressive and the implemention pretty smart. Thanks for sharing!

Because he didn't published it, but there are other similar ports that have been taken down.

For example, this C64 port was taken down: https://ko-ko74.itch.io/balatro-for-the-commodore-64-c64

I built it for macOS natively and it works like a charm!

While I understand the sentiment of Playstack and LocalThunk to take down these kind of ports, it's a shame for the community because this will only make the game grow more.

This is an interesting solution and wouldn't mind using one of my existing servers as a gateway or proxy (?).

Is there a way to be selective about what ports are exposed from the host to the target? The target could handle it but fine grained control is nice.

There is a filter for `createFilter` [1] and there is a throw if the index of the array doesn't exist for `createSelector` [2]. Maybe this is what the author refers to as hallucination-free, but falls pretty short.

[1]: https://github.com/montyanderson/incant/blob/master/mod.ts#L...

[2]: https://github.com/montyanderson/incant/blob/master/mod.ts#L...

When I scroll, the images in the README get a brown color.

Heads up that hashes cannot be read from the server.

I love this story so much.

This is funny. Is this really checking to see if a combination of two names has a website?

Thanks for the response! The combination of a seeded rng and some randomness is pretty cleaver.

Can't wait for the write up. Very interesting project and well executed.

PS: The console.log messages made me laugh

Made me giggle initially then I clicked a grey monkey and I was blown away! Great job with the project.

Are you storing everything they are typing? I'm very curious about how you are handling the data.

I've been running and using a puppeting bridge [1] for my Matrix instance for more than 4 years with 3 different numbers and no problem at all.

The controls they have in place are probably based on behavior, rather than on access.

[1]: https://github.com/mautrix/whatsapp

have you explored what you can do combining your info with what you can get with their cli?

I'm aware of these sites. But they are usually behind a paywall or it's hard to find a "download all" button. And I wouldn't use their API per request either.

You are correct, I will rename to CIDRs to make it clear.

I will add the sources. Data comes from feeds published by the orgs.

I agree. And each provider has its own way. IBM, e.g. has a markdown document, which was a pain to parse.

Thanks for sending the Bunny list, I will add it shortly.

I tried with an image that is 82.21 KB and compressed it to 252.98 KB.

I laughed out loud, great concept and use of technology!

> No hidden cost

I couldn't find the price per document in the home page.

I've been a contributor to the project and I've been using it for years. I love the community effort.

Similar projects are also the TMDB (The Movie Database) and TVDB (Television and Film Database).