Hacker News: blcknight

New comment by blcknight in "Bun has an open PR adding shared-memory threads to JavaScriptCore"

blcknight — Sat, 20 Jun 2026 22:33:24 +0000

I hear a lot of complaints about bun but nothing concrete about what broke in the migration.

You are also assuming one prompt, and then arguing against your assumptions with zero evidence. It is lazy arm chair criticism.

New comment by blcknight in "Hide Secrets from AI Agents and NPM install using Airgap"

blcknight — Sat, 20 Jun 2026 01:02:02 +0000

How can an agent use these tokens then? If it sources the file can't it just read the env?

It also sounds like it is missing the important step of keeping the LLM credentials from the agents themselves. For example my GCP creds have access to far more than Vertex. This is solved by OneCLI and OpenShell via MITM proxies which seems more elegant to me. The tools live in containers and can't see anything but can use everything.

It also allows finer grained access controls, rating limiting, and there's talk of scanning for destructive actions.

New comment by blcknight in "Zero-Touch OAuth for MCP"

blcknight — Fri, 19 Jun 2026 15:33:08 +0000

There's a new standard in progress for skills over MCP

https://modelcontextprotocol.io/community/working-groups/ski...

New comment by blcknight in "New way of making espresso with ultrasound"

blcknight — Tue, 16 Jun 2026 09:26:39 +0000

I like my Nespresso thank you very much.

New comment by blcknight in "Arch Linux Now Believes Malware Incident Under Control: More Than 1,500 Packages"

blcknight — Sat, 13 Jun 2026 15:47:08 +0000

"Review" them how? Read every single line of code before installing something? If it's a binary package, how do you do that? Make reproducible builds for everything you install? Move to from source distro? Putting this on users is not a tenable solution. There's room for common sense, but blaming the users for this is ridiculous

New comment by blcknight in "Claude Fable is relentlessly proactive"

blcknight — Fri, 12 Jun 2026 02:24:29 +0000

One bad npm package can really ruin your day. These things for me only run in their own VM with it's own GitHub account and basically nothing else

New comment by blcknight in "Claude Fable 5"

blcknight — Tue, 09 Jun 2026 19:13:37 +0000

The fallback doesn't seem to be working for me, I haven't scanned a project in it immediately booted me when it found a security bug even though I didn't ask for it

New comment by blcknight in "Codex just found a "workaround" of not having sudo on my PC"

blcknight — Mon, 01 Jun 2026 01:41:18 +0000

Kube play and quadlets are cool

New comment by blcknight in "Kindle loyalists scramble as Amazon turns page on old e-readers"

blcknight — Sun, 24 May 2026 18:05:08 +0000

The vast majority of kobo books are still ACSM protected.

New comment by blcknight in "Kindle loyalists scramble as Amazon turns page on old e-readers"

blcknight — Sun, 24 May 2026 10:57:38 +0000

It hasn't changed, and I don't know why people are saying that most books don't have DRM. It is only a small minority.

Tor books is the largest publisher without it (owned by Macmillan). Otherwise everything is truly hard DRM either ACSM with epub or Kindle's. They are both more or less easily defeated though.

New comment by blcknight in "MCP Hello Page"

blcknight — Sun, 17 May 2026 00:04:38 +0000

All kinds of tools make it really difficult to not make a URL clickable and even if it wasn't clickable they might still put it in address bar...

New comment by blcknight in "The agent harness belongs outside the sandbox"

blcknight — Sat, 02 May 2026 22:25:59 +0000

I am not sure anyone knows what a harness is at this point. I've heard 17 different definitions of it at this point. It's almost like a buzzword in search of a problem.

New comment by blcknight in "IMF says America's $39T national debt is a global problem"

blcknight — Sat, 18 Apr 2026 19:10:05 +0000

Trump just needs to mint a 39T coin with his face on it.

New comment by blcknight in "Claude Code Routines"

blcknight — Wed, 15 Apr 2026 01:11:19 +0000

That's why I mentioned `-p`.

`--continue` and `--resume` are broken from `-p` sessions for the last 2 weeks. The use case is:

1. Do autonomous claudey thing (claude -p 'hey do this thing')

2. Do a deterministic thing

3. Reinvoke claude with `--continue`

This no longer works. I've had this workflow in GitHub actions for months and all of a sudden they broke it.

They constantly break stuff I rely on.

Skill script loading was broken for weeks a couple months ago. Hooks have been broken numerous times.

So tired of their lack of testing.

New comment by blcknight in "Turn your best AI prompts into one-click tools in Chrome"

blcknight — Wed, 15 Apr 2026 00:07:53 +0000

Fight enshittification. For whatever reason, many travel sites no longer send full details in the e-mail confirmation, they want you to click through to the site...which means I can't forward it to plans@tripit.com for automatic import.

Immediately after booking something,I tell Gemini to add it to my TripIt. Works great. I have a little prompt explaining how I like it formatted that I cut and paste, so I can just make this a one-click prompt. I could also have it add flights to my.flightradar24.com.

I also use Gemini in Chrome to add appointment confirmations to my calendar. Or remember things in Google Keep.

There's lot of use cases for this kind of thing.

New comment by blcknight in "Turn your best AI prompts into one-click tools in Chrome"

blcknight — Tue, 14 Apr 2026 23:08:06 +0000

Lots of people are flocking to Claude and ChatGPT -- and making Gemini more useful in the browser everyone already has makes a lot of sense.

More Google use, more data they gather, more ads they can show you

New comment by blcknight in "Claude Code Routines"

blcknight — Tue, 14 Apr 2026 22:58:25 +0000

For the love of god fix bugs and write some fricken tests instead of dropping new shiny things

It is absolutely wild to me you guys broke `--continue` from `-p` TWO WEEKS AGO and it is still not fixed.

New comment by blcknight in "Apple's accidental moat: How the "AI Loser" may end up winning"

blcknight — Mon, 13 Apr 2026 06:18:58 +0000

There is a cognitive ceiling for what you can do with smaller models. Animals with simpler neural pathways often outperform whatever think they are capable of but there's no substitute for scale. I don't think you'll ever get a 4B or 8B model equivalent to Opus 4.6. Maybe just for coding tasks but certainly not Opus' breadth.

New comment by blcknight in "Lay offs at Red Hat"

blcknight — Thu, 09 Apr 2026 19:19:25 +0000

:-(

New comment by blcknight in "The Claude Code Source Leak: fake tools, frustration regexes, undercover mode"

blcknight — Wed, 01 Apr 2026 00:59:17 +0000

Yea this is the thing that makes no sense to me. Any frontier model can unmiminize minified JS pretty decently. Obviously not everything comes through, comments and such, but I always assumed the reason it wasn't open source was to prevent an endless shitstorm of AI slop PR's, not because they were trying to protect secret sauce.