Hacker News Reader

stbenjamon Backdoor in Claude Code for VS Code extension on the Cursor Marketplace

We still don’t know exactly what happened, but I publish software to public registries, and supply chain attacks scare the hell out of me. They also frustrate me, because most of the defenses aren't automatic and often really hard to get right. We're really failing as an industry to do better here.

stbenjamon Backdoor in Claude Code for VS Code extension on the Cursor Marketplace

5,000*

stbenjamon Gemini 4 Carbon

I can't tell if they're withholding access to manufacture some sense of scarcity but the delays are really prolonged. I'm nervous to use Gemini models anyway, as its attached to my Google account and I've seen too many people having their whole account banned for dubious reasons (with little recourse).

Also the whole "we're only releasing it to cyber defenders first" framing is weird, is any Gemini model an actual cyber risk?

stbenjamon No signs of AI payoff outside the tech sector

Small sample size, but I've heard from multiple physicians that AI note taking allows them to see more patients or spend more time with them and they're very happy with those kinds of capabilities that take over toil

stbenjamon Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor

Yea, I think all the tools have this now but its not a default. Install scripts should also be turned off by default and require some hassle to turn them on for the first time in a package.

stbenjamon Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor

IMHO defaults for uv, npm, etc should all have a cooldown for 24 - 48 hours by default, you should have to go out of your way to install something newer.

stbenjamon Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor

This... isn't acceptable. Once is forgivable, twice is malpractice.

stbenjamon Livenerf: Has Opus 5.5 been nerfed yet?

what word?

stbenjamon Agents don't need memory, they need documentation

I’ve noticed a growing pattern of people creating repositories full of Markdown documentation, or adding large amounts of it directly to their main repositories, often generated by agents. In some cases, this can add up to megabytes of material, and I haven’t yet seen much evidence that this level of documentation meaningfully improves an agent’s performance and that it just doesn't rot over time.

If someone has good A/B evals of this being more effective I'll eat my hat, but the reason no one is publishing them is because well, evals are hard, and this is likely just magical thinking.

My current view is that an agent generally needs three things to work effectively: a way to discover information that isn’t obvious, such as a minimal `AGENTS.md` that points to more focused brief files; a clear way to verify that its work is correct; and some guidance on project-specific tastes. Everything else is noise.

stbenjamon Livenerf: Has Opus 5.5 been nerfed yet?

These guys are on twitter angry about the rate limit decrease and allegedly cancelled all their OpenAI accounts. Wonder how they'll maintain this.

I am quite convinced that the whole nerfing phenomenon is 90% AI psychosis. I have the word muted on X.

stbenjamon Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com

Oh wow those fees are predatory

stbenjamon Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com

It is a money grab from Amazon I'm sure. They want to make a deal with meta to get paid for access. Maybe also worried about liability I imagine on accidental unauthorized purchases?

stbenjamon UTF-8000: Unlimited UTF-8

> No special cases introduced. All properties preserved.

I don’t actually know if this is LLM-generated, but phrasing like this is weirdly triggering to me now

stbenjamon Claude Code now reads AGENTS.md if there is no Claude.md

Everyone sharing the 513 ways you can already do this is entirely missing the point. Standards matter and there's very few things in the AI world everyone agrees on and this is one of them. .agents/skills is another and anthropic is still holding out there.

stbenjamon Pion, an agent designed to run any company autonomously

The example of the plant watering business is the hellscape future where humans have to deal with agents bugging them to sell them their services constantly. We're already seeing the first hints of this with iLands.

And the "hiring technicians" thing... great, humans on call for robots isn't the future I want.

stbenjamon Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher

https://github.com/stbenjam/skills/tree/main/plugins/hype

I have a plug-in to do this. I don't know if it's effective but Claude said it was genuinely helpful (obviously would say that about anything)

stbenjamon C*: Unifying Programming and Verification in C

By that logic, I think I'll create a new language called Pascal, or maybe Ada...

stbenjamon OpenClaw 2.0, Accidentally

The risk here is wildly overstated, prompt injection risk is becoming vanishingly small with the latest frontier models.

I would not run an OpenClaw with full access to my bitwarden, but it certainly has some logins available to it, and can make purchases with link-cli which has human-in-the-loop.

stbenjamon I spent $266 and four AI models to own my tablet. GLM-5.3 finished it in a day

Ah sorry, for OpenAI it does seem automatic with ID verification. Anthropic's seems more picky.

stbenjamon I spent $266 and four AI models to own my tablet. GLM-5.3 finished it in a day

Apply, not sign up. It is not automatic and most will be refused.

stbenjamon Managing AI Coding Costs at Scale

$80 sounds extremely low for what you're describing - are you on API token plans?

I have had some $3,000 token days - even without Fable. I don't see how this is sustainable.

My personal 20x plans get so much usage for so cheap. The consumer subsidies are crazy, but alas I can't use them for work.

stbenjamon Pi's Minimalism Is Its Advantage

Pi is ok, but I dislike needing plugins to do anything useful. Subagents, MCP, /goal ... none of these things should be plugins.

stbenjamon Patreon laying off 20% of staff

My comment about legal reasons was legal reasons surrounding employee law suits from being laid off. At will employment in most US states allow you to state no reason, that's safer than offering any reason at all. A specific reason for a layoff can be litigated.

stbenjamon Benchmarking Opus 5 on SlopCodeBench

> I hope the big labs will start using this benchmark in their RL pipelines.

Labs do not train on benchmark data (allegedly). They can train on similar problems, but benchmarks have specific strings in them that labs are supposed to be aggressive in filtering out of their training corpora.

stbenjamon Patreon laying off 20% of staff

I assume it is legal reasons.

stbenjamon TFTP Honey Pot Results

I know tftp is still in wide use, I wonder if there's things out there looking for stuff that's less common like NNTP, finger servers, etc

stbenjamon How to read more books

Oh I didn't hear about this. That sucks. The stock firmware is terrible. CrossPoint makes it usable.

stbenjamon How to read more books

I adore my XTEink 4 with the crosspoint firmware. Best small form factor ereader

stbenjamon Grok 4.5

Elon's rhetoric doesn't really match the model's behavior. It is willing to criticize Elon and argues against many of the insane right way points he tries to make.

stbenjamon Zuckerberg says AI agent development going slower than expected

The gulf is bridgeable. The problem is that a lot of people are building agents without strong enough judgment layers around them. Work that can be verified with reasonable accuracy are the sweet spot right now.