NVIDIA just put 2,575 simulation-ready 3D assets on Hugging Face 🤯
A 472GB collection covering everyday objects, tools and industrial equipment.
It’s called NVIDIA SimReady Assets. The models include physics settings and semantic labels for use in simulated environments.
- Train robots to pick up and move objects
- Build virtual factories and warehouses
- Generate labelled training images
- Populate 3D scenes with ready-made objects
Useful for robotics research, digital twins and anyone building detailed virtual environments.
Download the collection below 👇
holy sh*t, someone just dropped REAP, the prompt library for REA
1.7k stars in 5 days, and it turns "how do I even ask REA this" into copy-paste prompts
https://t.co/9WJMfuCf1V
the sections:
→ function tracking and binary analysis
→ software cloning and game logic reconstruction
→ porting to WASM
→ firmware, Android APK and kernel
→ Electron apps
→ vulnerability research, anti-cheat and DMA research for CTFs and security work
17 sections, swap the <placeholders> and run
every prompt forces the same rule: files, functions and addresses for every claim, confirmed facts kept apart from guesses
ask about one function at a time, the narrower the question, the more you can trust the answer
save this ⭣
Intellectual property is dead. Not dying.
Holodeck Program Guidelines #06 👇
https://t.co/URrd6Hc7dD
In one month: the PS5 was jailbroken, EA shut down BioWare, and Adobe’s stock slid as one person vibe-coded free Adobe Suite clones and AI agents can decompile and remix any game.
You can still make dope sh*t. You just can’t stop people from using it as they please.
I grew up consuming pirated art, and have unashamedly used it to paint the worlds I've created for over 30 years. Now I'm asking, should I allow my users the same privilege? To use what I've built to paint their own dreams?
i own a relatively new (2025) samsung tv.
i kept getting weirdly targeted ads.
i tried reversing the firmware but it was aes encrypted and no keys unless tear it apart and do uart/jtag magic.
zero days it is. root shell on new samsung TVs!
discovered PRIVACY HORROR. write up?
I just solved "Robotics Engineering" education with Opus 5.5 + /colorfigures skill
no need to spend $1000+ on a starter kit to learn what a resistor does, just visit my interactive site:
• 13 real boards: Arduino Uno, L298N, Raspberry Pi 5...
• build part by part: pick & place, reflow, solder, plug in
• break them on purpose: reversed caps pop, chips smoke, LEDs burn out
send the skill to your Claude:
> https://t.co/1879rauiu7
try it free, no sign-up ↓
holy sh*t, someone just mapped 12,000+ closed products you can reverse engineer today
cars, smart homes, printers, AI clients, firmware, file formats and dead online services, and most builders have no idea it's free
the drop:
→ https://t.co/NTLRt8gPW3 (28 ★ · 2 forks · 12,877 projects)
the index itself: every entry checked through the GitHub API for license, stars and status, refreshed every night, with a ready-made skill for your AI agent
→ game decompilations (3,040 projects)
decompilations, recompilations, emulators and restored engines, from shadPS4 and RPCS3 to OpenRA
→ devices, IoT & hardware (3,029 projects)
local control for smart homes, printers and wearables without the vendor's cloud
→ unofficial API clients (904 projects)
open clients for platforms that never shipped a public API
→ AI tools & LLM clients (739 projects)
reconstructed AI clients, tool definitions and OpenAI-compatible adapters for closed AI services
→ cars (520 projects)
open driver assistance, vehicle APIs and in-car systems, openpilot included
→ firmware & embedded (472 projects)
custom firmware and embedded code for devices their makers stopped updating
→ file formats (436 projects)
readers and parsers for closed formats like PSD and DWG
→ network protocols (422 projects)
open implementations of protocols that were never documented
→ online service revivals (173 projects)
replacement servers for games and services that were shut down
the formula:
search the index → clone what already exists → let REA reverse the missing piece → ship
REA + Opus 5.5 is scary f*cking dangerous...
Any closed-source device can now be reverse engineered.
Down to the binary, and you can modify it however you want.
Send this prompt to Claude and analyse how your PS5 works.
Then read the full article below on how to rea any app for Android👇
🔥 HUGE: Claude solved a 25-year-old crypto mystery in under 30 minutes, tracing the NSA-supplied seeds behind NIST's P-256 curve to a 1990s random number generator.
This is Pele, my new rendering library for diagrams and charts, compatible with the latest Mermaid syntax.
- 100-500× faster than mermaid.js
- 6× smaller bundle
- No dependencies
- Easy to theme
- Responsive to device width
In pursuit of making Obsidian insanely fast and cohesively designed, I always found the Mermaid library to be too big, too slow, and too hard to theme. Pele will be the new renderer in the next version of Obsidian, but you can start using it now.
Pele [ˈpɛlɛ] is named after the goddess of volcanoes and fire. It's open source under the MIT license. Try the playground!
You can build an entire AI content factory using just 10 free GitHub repos.
From researching viral content to generating videos, voiceovers, music, and scheduling posts, these tools cover the entire workflow.
**1. OpenCreator**\
AI workspace for scripts, images, voiceovers, translations, and editing.\
`https://t.co/CQRvjuEJ86`
**2. Social Media Research Skills**\
Find viral posts, analyze competitors, and extract winning hooks.\
`https://t.co/QYi1BGMFFX`
**3. Local Deep Research**\
Turn any topic into a detailed research report.\
`https://t.co/PpK4amG3xz`
**4. FLUX.2**\
Generate and edit images for thumbnails and social content.\
`https://t.co/nod7aHTXhC`
**5. Kokoro**\
Generate AI voiceovers locally.\
`https://t.co/iia0HrKIsD`
**6. ACE-Step 1.5**\
Create original music and instrumental tracks with AI.\
`https://t.co/zGAvtdJmef`
**7. Short Video Factory**\
Automate short-form videos with scripts, editing, voiceovers, and subtitles.\
`https://t.co/Pp2hI5uMDl`
**8. AI YouTube Shorts Generator**\
Turn long videos into Shorts with automatic highlights and cropping.\
`https://t.co/A1hTGKvW01`
**9. Podcastfy**\
Transform documents and articles into AI podcast conversations.\
`https://t.co/K8diM4XKun`
**10. Social Media Agent**\
Turn URLs into X and LinkedIn posts, review drafts, and schedule content.\
`https://t.co/hucQGDltIi`
Research → Script → Visuals → Voice → Edit → Publish.
Save this if you're building content with AI.
Make the hook more specificShorten each tool descriptionGroup the tools by workflow
This might be the most important paper on AI agents this month.
Researchers at CMU trained a small model to rewrite an AI agent’s harness, and it beat the 35B model it learned from.
The harness is the program around the model: how it gets called, which tools run, where the information goes. Here a 4B model reads a report from the last run, proposes edits to the harness, and keeps the one that scores best. The model doing the task never changes.
What they found:
> on task types it never saw, edit quality went from 0.32 to 0.62
> 5 edits in a row matched or beat the best of 8 parallel tries
> broken edits dropped from 38% to 15%
You can run the same loop today with any agent:
1. Put your harness in one file: prompts, steps, tool calls
2. Split your test cases into three sets: learn, score, final check
3. Run it on the learn set and save a short report of what failed
4. Ask a model for 5 edits based on that report
5. Run each edit on the score set and keep the winner
6. Repeat for 5 rounds, then test once on the final set
Key pages:
p.2: the full loop
p.6: results on unseen tasks
p.8: a harness the model learned