frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

OpenAPI and HTTP Tool Call is usually enough, no MCP is needed

https://blog.terrydjony.com/openapi-and-tool-is-enough/
2•terryds•42s ago•0 comments

Is my data used for model training?

https://privacy.claude.com/en/articles/10023580-is-my-data-used-for-model-training
1•leakycap•1m ago•1 comments

SSA: Learning to Reason Across Parallel Samples for LLM Reasoning

https://user074.github.io/ssa-parallel-reasoning/
1•lukejagg•2m ago•0 comments

Radio Starlight – Your Generative Radio Station

https://apps.apple.com/us/app/radio-starlight/id6749476669
1•craftkit•3m ago•0 comments

Not By AI Badge

https://notbyai.fyi/
1•modinfo•4m ago•0 comments

Making Unicode things fast in Go

https://clipperhouse.com/go-unicode/
2•mwsherman•6m ago•0 comments

Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to Stop Prompt Injections

https://www.archestra.ai/blog/dual-llm
5•ildari•6m ago•1 comments

We open-sourced XAI's Macrohard, an autonomous computer-using agent

https://github.com/LLmHub-dev/open-computer-use
1•PrateekJ17•7m ago•0 comments

Indian-Americans Are Holding Their Home Country at Arm's Length

https://www.bloomberg.com/opinion/newsletters/2025-10-11/indian-americans-are-staying-out-of-us-v...
1•thelastgallon•8m ago•0 comments

Fujitsu pumps £280M into UK arm to keep lights on after Horizon scandal

https://www.theregister.com/2025/10/13/fujitsu_uk_280m_funding/
4•rntn•8m ago•0 comments

I did a funny launch on Product Hunt, but nobody noticed anyway

1•dahra•9m ago•0 comments

The Eames Lounge Chair was always expensive

https://www.echevarria.io/blog/eames-lounge-chair-price-history/index.html
2•ivanech•13m ago•0 comments

The only benchmark that matters is the one that emulates your real workload

http://www.kroah.com/log/blog/2025/10/01/the-only-benchmark-that-matters-is.../
1•LorenDB•14m ago•0 comments

When Values Eat Their Young: How Ideal-Driven Groups Drift into Their Own Shadow

https://kennethreitz.org/essays/2025-08-25-when-values-eat-their-young
1•PaulHoule•14m ago•0 comments

Sandboxing on macOS

https://bdash.net.nz/posts/sandboxing-on-macos/
1•aviramha•15m ago•0 comments

Multi-booting Windows From 1.01 to 11 on real hardware

https://old.reddit.com/r/pcmasterrace/comments/1o4vilb/multibooting_windows_from_101_to_11_all_17/
1•LorenDB•16m ago•0 comments

Causal Profiling

https://khayyamguliyev.substack.com/p/causal-profiling
1•khguliye•16m ago•0 comments

Instantspaces: Remove/Minimize macOS Space Animations

https://github.com/flawnn/instantspaces
2•flawn•17m ago•2 comments

Tauchgang

https://tauchgang.dev
3•amanciero•17m ago•1 comments

The Problem with “You”: How a Pronoun Complicates Our Relationship with AI

https://opuslabs.substack.com/p/the-problem-with-you
1•opuslabs•18m ago•0 comments

24 Years of NASA Satellite Data Suggest the World Is Getting Darker

https://www.iflscience.com/24-years-of-nasa-satellite-data-suggest-the-world-is-getting-darker-an...
2•geox•27m ago•0 comments

Trio win Nobel economics prize for work on technology-driven growth

https://www.theguardian.com/business/2025/oct/13/nobel-economics-prize-technology-joel-mokyr-phil...
1•andsoitis•28m ago•0 comments

Show HN: CrossRun – An Open Standard to Make Software Run on Any Platform

https://github.com/CrossRun/Manifesto
1•Swarde11•30m ago•0 comments

Fentanyl overdoses among seniors surge 9k% – A hidden crisis few saw coming

https://www.sciencedaily.com/releases/2025/10/251012054606.htm
1•Noaidi•30m ago•0 comments

AI Triples Stroke Recovery Rates

https://thedispatch.com/article/donald-trump-third-term-steve-bannon-jd-vance/
1•mhb•31m ago•0 comments

Show HN: Terminal-First Documentation Service

https://www.dodo-doc.com/
1•mkei29•31m ago•0 comments

Ross Sorkin on similarities between Wall Street today and 1929's markets

https://www.youtube.com/watch?v=26WMF7Xdb0E
2•belter•32m ago•1 comments

Show HN: ch, a terminal tool to highlight specific words in your command output

https://github.com/dtonon/ch
1•dtonon•33m ago•0 comments

Show HN: EMDR Bilateral Stimulation App

https://myemdr.app/bls
1•positive-minds•34m ago•0 comments

Apple ups the reward for finding major exploits to $2M

https://www.wired.com/story/apple-announces-2-million-bug-bounty-reward/
3•Brajeshwar•35m ago•1 comments
Open in hackernews

RMPocalypse: Breaking AMD Confidential Computing with a Single Write

https://rmpocalypse.github.io/
3•hasheddan•2h ago

Comments

magicalhippo•1h ago
Under the "How difficult is it to exploit" in the FAQ they mention it's very easy and with 100% success. Yet the text itself says this:

In our analysis of the RMP initialization, we observed that the malicious hypervisor running on the x86 cores can still create dirty cachelines pointing to DRAM. [...] As depicted in (c), the malicious hypervisor can use the primitive to get arbitrary unchecked writes to RMP memory.

So it would seem it's easy as long as you managed to install a malicious hypervisor...

Of course not great, with supply chain attacks being a serious cause for concern. Still, hardly "easy" if it requires hijacking a core piece of infrastructure?