Reddeet

49 readers
0 users here now

Welcome !

This instance is open to ideas as to where it should go. Contact the admin at [email protected] if you have any suggestions/issues.

Like the old Reddit style ?

Cool links !

Technical

This instance is hosted on an ARM based server (Hetzner CAX Server) :

Analytics

You can check out the data we collect when you visit this instance right there : analytics.kawa.zip/reddeet.com

None of this data is sold to anyone, it is used for educational purposes only.

founded 2 years ago
ADMINS
1
 
 

Anthropic, the flagship AI company, has inadvertently exposed the source code for its major CLI tool Claude Code. It has already been extracted with mirrors published on GitHub.

Chaofan Shou announced the discovery on X.

“Claude Code source code has been leaked via a map file in their npm registry!” the user posted.

The post immediately stirred the AI community, attracting nearly 10 million views and 1.5 thousand comments.

Claude Code is a massively popular agentic AI coding assistant that runs in terminal. It can edit files and manage entire projects locally. Anthropic’s tool is closed-sourced and distributed as an obfuscated npm package.

However, Anthropic's published npm package containing Claude Code v2.1.88 allegedly exposed the source map file cli.js.map, which contained the full and unobfuscated TypeScript source code.

It appears that Anthropic scrambled to remove the npm package. However, it was too late. Multiple GitHub users already exposed copies of the project.

One of the GitHub repositories has already amassed nearly 30,000 stars and 40,200 forks.

Another one has 425 stars and 520 forks, with many developers dissecting the inner workings of the tool.

The author claims that the leaked Anthropic’s .map file’s size was 57MB, mapping 1,900 files and 512,000 lines of code.

The leaked code includes the core engine for LLM API calls, handling streaming responses, tool-call loops, thinking mode, retry logic, token counting, permission models, tools, etc. Some Hacker News users noted the extensive regex filter containing many swear words for detecting negative sentiment in users' prompts. This leak doesn’t expose the AI models themselves or user data.

Exposed internal logic makes it very easy to reverse-engineer the tool, identify security risks, or steal intellectual property. Users already noted that Claude Code is using axios as its dependency, a tool that was just hacked.

Many GitHub users now advertise their own build of Claude Code. However, they’re risking legal action.

“Just because the source is now 'available' *DOES NOT MEAN IT IS OPEN SOURCE*. You are violating a license if you copy or redistribute the source code, or use their prompts in your next project! Don’t do that,” posted full-stack developer Justin Schroeder on X.

The .map files are typically used by developers fixing programs in software. These plain text files act like a mini map of the original code, helping developers trace where the errors or problems occur. However, .map files usually don’t include the full source code.

Anthropic hasn’t yet released an official statement. Cybernews reached out to the company and will include its response.

2
3
4
 
 
5
 
 

I suggest watching the video, https://m.youtube.com/watch?v=QkC1aK7jfLo but the article has an OK summary.

Also a Mastodon shout-out in the video.

6
7
14
ah shit (reddthat.com)
submitted 3 hours ago by [email protected] to c/[email protected]
 
 
8
81
submitted 5 hours ago* (last edited 5 hours ago) by [email protected] to c/[email protected]
9
166
Boooooo (crazypeople.online)
submitted 6 hours ago by [email protected] to c/[email protected]
 
 
10
 
 

Catodon is a new Sharkey/Misskey fork.

11
236
Anon dips (sh.itjust.works)
submitted 7 hours ago by [email protected] to c/[email protected]
 
 
12
13
28
The Button 🔘 (thebutton.lemmy.zip)
submitted 6 hours ago* (last edited 6 hours ago) by [email protected] to c/[email protected]
14
15
16
 
 

cross-posted from: https://lemmy.zip/post/61746701

The Button 🔘

An experiment has begun.

There’s a button.

That’s it. Just a button… and a timer counting down.

Anyone can press it. Everyone sees what happens. And every press changes things forever.

  • One shared button
  • No undo, no reset
  • A collective test of curiosity, restraint… or chaos

Some of you will hold the line. Some of you won’t last five seconds.

Either way, once it’s over… it’s over.

The Button

So the only question is:

Do you press it?


The Button is open to anyone on any federated platform, be that Lemmy, Piefed, Mastodon... even Stegodon!

Any user from any instance can sign in and press the button, but you only get one chance.

To add another layer to this, users can choose which faction to join, and the "winner" is the faction that holds the button the longest amount of time before it eventually runs out.

Initially a press of the button resets the timer to 15 minutes - this may reduce if the game carries on for a while. At the time of this post, there is about 60 minutes until the button expires. The first button press will be available when there's 15 minutes left, i.e. in 45 minutes, but you can sign up now to be ready!

There is also a permanent record of the first pusher :)

17
 
 

The vast data centers that power artificial intelligence are so energy hungry that they’re heating up their surroundings, according to new research. It’s an alarming finding given the number of data centers is predicted to explode over the next few years.

18
19
 
 
20
 
 

An internal memo dispatched by senior execs at Red Hat suggests the software biz is starting to push AI tooling within its Global Engineering department. RHEL may be about to get some Windows 11-style "improvements."

It carries the heading "Engineering that's evolved and amplified for the AI era," and for any AI skeptics in the developer teams at Red Hat, the tone of the email may raise alarm bells. The times are changing, it states.

21
 
 

Most of the time the companies say the layoffs are because AI is doing people jobs, but the reality is that making things that use AI is just a lot more expensive. And enabling employees to use AI more than the free tier is really expensive too. So, they need to cut cost elsewhere to balance it out.

22
 
 

the FTC alleged that OkCupid gave an unauthorized third party access to the personal data of millions of OkCupid users in violation of its privacy policies.

23
 
 

Is it possible to do this? It may be silly but the automatic profile picture with the initials bother me.

24
12
Rate my stack: (sh.itjust.works)
submitted 8 hours ago by [email protected] to c/[email protected]
 
 

this is my current plan, but I've yet to selfhost for longer than a month or two previously. what do y'all think of my choices?

Proxmox HV running TrueNAS+Debian Stable Server

Prowlarr: Indexer manager Sonarr: TV show management automation Radarr: Movie management automation LazyLibrarian: Book management automation Lidarr: Music management automation Homarr: Dashboard for managing applications Seerr: Media request management system Jellyfin: Media server qBittorrent: Torrent client NZBGet: Usenet downloader WireGuard: VPN software Surfshark: VPN service Portainer: Docker container management UI Watchtower: Automated Docker container updates Immich: Photo gallery & backup Mealie: Meal planner Moonlight: Low latency remote gaming (retro game emulator focused) Kavita: Ereader for books, manga, audiobooks, most formats Funkwhale: Music streaming

open to suggestions, but wanted to see if the community would perceive this as a reasonably interlocked software system or if i need to be using other software.

incredibly new and lowkey uninformed by trying my best to learn. plz be nice lol

25
view more: next ›