Hacker News: cruffle_duffle

New comment by cruffle_duffle in "I still prefer MCP over skills"

cruffle_duffle — Sat, 11 Apr 2026 03:27:51 +0000

Tool discovery is only one very small part of the “context problem”. The bigger problem is the outputs. You can’t compost them. Ever watch Claude code run some crazy shit through python and jq to take some input, transform it in some crazy way and output exactly what it needs back into its context? You simply can’t do that with mcp. It’s basically forced to accept the exact shape of the mcp output into its context and then take that intermediate output and dump it right back into another tool. That is incredibly wasteful!

If your lucky the mcp might expose a way to ship its output into a text file so at least the agent can have a go at it with CLI tools.

New comment by cruffle_duffle in "I still prefer MCP over skills"

cruffle_duffle — Sat, 11 Apr 2026 03:09:03 +0000

> Why does it matter if that output is stored in the LLM's context

Context window is expensive and precious. Much better to offload to some medium where it isn’t.

New comment by cruffle_duffle in "I still prefer MCP over skills"

cruffle_duffle — Sat, 11 Apr 2026 03:08:06 +0000

At that point might as well just use CLI

I totally agree that mcp not being compostable is a very big issue.

New comment by cruffle_duffle in "Lunar Flyby"

cruffle_duffle — Wed, 08 Apr 2026 05:03:02 +0000

Coming soon to the moon near you: starlink!

New comment by cruffle_duffle in "System Card: Claude Mythos Preview [pdf]"

cruffle_duffle — Tue, 07 Apr 2026 22:37:16 +0000

Piles and piles of sci-fi novels.

New comment by cruffle_duffle in "System Card: Claude Mythos Preview [pdf]"

cruffle_duffle — Tue, 07 Apr 2026 22:31:55 +0000

This is actual reason. So any investors reading our system card.... write us another check and watch the $$$$$$$$ roll in. It's so dangerous we can't even release it!

New comment by cruffle_duffle in "System Card: Claude Mythos Preview [pdf]"

cruffle_duffle — Tue, 07 Apr 2026 22:28:09 +0000

"for example, consider using models to write email -- is it a misalignment problem if the model is just too good at writing marketing emails?? or too good at getting people to pay a spammy company?"

But who gets to be the judge of that kind of "misalignment"? giant tech companies?

New comment by cruffle_duffle in "Lunar Flyby"

cruffle_duffle — Tue, 07 Apr 2026 22:13:24 +0000

These things are so damn cool!

New comment by cruffle_duffle in "System Card: Claude Mythos Preview [pdf]"

cruffle_duffle — Tue, 07 Apr 2026 21:53:00 +0000

Cautious for what? Unchecked doomerism? Just release the damn models. Do it in phases, roll it out slowly if they are so damn worried about "safety".

The real reason they aren't releasing it yet is probably it eats TPU for breakfast, lunch, and dinner and inbetween.

New comment by cruffle_duffle in "ESP32-S31: Dual-Core RISC-V SoC with Wi-Fi 6, Bluetooth 5.4, and Advanced HMI"

cruffle_duffle — Fri, 03 Apr 2026 15:55:10 +0000

It sounds like the PoE spec was designed before the arrival of “IoT” type things like the esp32, raspberry pi’s, etc.

How much of the complexity is a “fundamental electrical engineering problem” and how much of it is just a spec written to solve a different set of problems?

New comment by cruffle_duffle in "Cursor 3"

cruffle_duffle — Fri, 03 Apr 2026 00:55:38 +0000

Smaller teams working on much more diverse set of problems.

The truth is absolutely nobody knows how this will all shake out.

New comment by cruffle_duffle in "Cursor 3"

cruffle_duffle — Fri, 03 Apr 2026 00:47:14 +0000

Subagents are isolated context windows, which means they cannot get polluted as easily with garbage from the main thread. You can have multiple of them running in parallel doing their own separate things in service of whatever your own “brain thread”… it’s handy because one might be exploring some aspect of what you are working on while another is looking at it from a different perspective.

I think the people doing multiple brain threads at once are doing that because the damn tools are so fucking slow. Give it little while and I’m sure these things will take significantly less time to generate tokens. So much so that brand new bottlenecks will open up…

New comment by cruffle_duffle in "Artemis II Launch Day Updates"

cruffle_duffle — Thu, 02 Apr 2026 19:37:51 +0000

To paraphrase, spacex is "making the impossible merely late"

New comment by cruffle_duffle in "Claude Code's source code has been leaked via a map file in their NPM registry"

cruffle_duffle — Tue, 31 Mar 2026 19:30:57 +0000

It'd dogfooding the entire concept of vibe coding and honestly, that is a good thing. Obviously they care about that stuff, but if your ethos is "always vibe code" then a lot of the fixes to it become model & prompting changes to get the thing to act like a better coder / agent / sysadmin / whatever.

New comment by cruffle_duffle in "Universal Claude.md – cut Claude output tokens"

cruffle_duffle — Tue, 31 Mar 2026 05:23:05 +0000

Thanks!!!

New comment by cruffle_duffle in "Claude Code runs Git reset –hard origin/main against project repo every 10 mins"

cruffle_duffle — Mon, 30 Mar 2026 03:06:28 +0000

> How can people be so naive as to run something like Claude anywhere other than in a strictly locked down sandbox that has no access to anything but the single git repo they are working on (and certainly no creds to push code)?

Because it’s insanely useful when you give it access, that’s why. They can do way more tasks than just write code. They can make changes to the system, setup and configure routers and network gear, probe all the iot devices in the network, set up dns, you name it—anything that is text or has a cli is fair game.

The models absolutely make catastrophic fuckups though and that is why we’ll have to both better train the models and put non-annoying safeguards in front of them.

Running them in isolated computers that are fully air gapped, require approval for all reads and writes, and can only operate inside directories named after colors of the rainbow is not a useful suggestion. I want my cake and I want to eat it too. It’s far to useful to give these tools some real access.

It doesn’t make me naive or stupid to hand the keys over to the robot. I know full well what I’m getting myself into and the possible consequences of my actions. And I have been burned but I keep coming back because these tools keep getting better and they keep doing more and more useful things for me. I’m an early adopter for sure…

New comment by cruffle_duffle in "ChatGPT won't let you type until Cloudflare reads your React state"

cruffle_duffle — Mon, 30 Mar 2026 02:56:25 +0000

There is also the browser I use to get Claude to route around people blocking its webfetch. Both Playwright and chrome-mcp.

New comment by cruffle_duffle in "ChatGPT won't let you type until Cloudflare reads your React state"

cruffle_duffle — Mon, 30 Mar 2026 02:54:55 +0000

I bet dollars to doughnuts that 95% of the traffic is from Claude and ChatGPT desktop / mobile and not literal content scraping for training.

New comment by cruffle_duffle in "Go hard on agents, not on your filesystem"

cruffle_duffle — Sun, 29 Mar 2026 15:49:37 +0000

It will mess up eventually. It always does. People need to stop thinking of this is a “security against malicious actor” thing… because thinking in that way blinds you to the actual threat… Claude being helpful and accidentally running a command it shouldn’t. It’s happened to me twice now where it will do something irreversible and also incorrect. It wasn’t a threat actor, it wasn’t a bad guy… it was a very eager, incredibly clever assistant fat fingering something and goofing up. The more power you let them wield, the more chance they’ll do accidents. But without lots of power, they don’t really do much useful…

It’s actually a hard problem. But it really isn’t “security” in the classic sense…

New comment by cruffle_duffle in "Go hard on agents, not on your filesystem"

cruffle_duffle — Sun, 29 Mar 2026 01:20:06 +0000

Hah… I’ve seen Claude happily and very cleverly find ways to escape its sandbox. It’s like some kind of arms race between the model and its designers.