Hacker News: chatmasta

New comment by chatmasta in "Show HN: I pipe free sports streams into Jellyfin – no ads, just HLS"

chatmasta — Fri, 10 Apr 2026 00:15:43 +0000

This is the only evergreen bookmark that will always have the most up-to-date sources.

You can also find some interesting tools and scripts if you search some of the streaming domain names with GitHub code search.

New comment by chatmasta in "Reallocating $100/Month Claude Code Spend to Zed and OpenRouter"

chatmasta — Thu, 09 Apr 2026 23:53:43 +0000

Regardless of which harness you use, asking your agent to self-edit its own .claude (and to put it in the repo itself so you see the changes) is the single biggest impact change you can make in terms of compounding improvement. Couple this with telling it to create skills for /garden (clean up drift based on what changed this session), /handoff (garden, create ant skills to resolve friction encountered this session, and write a summary of the session and note for next agent), /takeover (read the latest handoff file). Since doing this I’ve completely cured my session-abandonment anxiety and can confidently swap to a new session at < 20% context usage without feeling like I’m talking to someone who just woke up from a coma.

New comment by chatmasta in "Databricks co-founder wins prestigious ACM award, says 'AGI is here already'"

chatmasta — Wed, 08 Apr 2026 21:58:58 +0000

Zaharia is a legitimately smart and impressive guy, totally deserving of the award, but it is a bit depressing to see what’s traditionally been an academic honor turn into a sales pitch.

To be fair he wasn’t very explicit about it, maybe not even trying to sound like a sales pitch, but that’s just what happens through osmosis when you spend all day at an enterprise focused on selling data governance solutions for the agentic era… also, he’s not wrong. The AGI claim is an exaggeration but the risk management implication is not.

If AGI _were_ truly here today, it would be exceptionally dangerous to let it run amuck in your enterprise… and if it comes any time in the next few years, it’ll be a time when a good chunk of corporate America has clicked through every OAuth screen in their organization to grant full access to some AI inference machine.

New comment by chatmasta in "US and Iran agree to provisional ceasefire"

chatmasta — Wed, 08 Apr 2026 00:46:46 +0000

Better article with text: https://www.theguardian.com/us-news/2026/apr/07/trump-iran-w...

> Israel will also agree to the two-week ceasefire, Axios reported, citing an Israeli official, adding that the ceasefire would enter effect as soon as the blockade of the strait of Hormuz ceased

There’s the catch.

New comment by chatmasta in "US and Iran agree to provisional ceasefire"

chatmasta — Wed, 08 Apr 2026 00:44:27 +0000

What are the chances Claude was used on both sides of this negotiation?

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sun, 05 Apr 2026 21:38:43 +0000

Make it write a skill and rule hook for PreCompact to do a handoff that explains what was worked on, what to know, and what to do next. If it goes off the rails after compaction then it won’t be great in a new session either, and you want to make sure you maximize continuity or development will be unsustainable. A backlog.md and improvements.md workflow also helps with this (ticket numbers, descriptions, “focus on BACK-0075,” etc.)

New comment by chatmasta in "What if the browser built the UI for you?"

chatmasta — Sun, 05 Apr 2026 05:23:31 +0000

Cool idea and line of thought, obviously rough and early but it gets you thinking. “Software as clay” is obviously where the industry is heading, and as you say we’re approaching this from multiple angles… applying it directly in the browser is certainly an intriguing idea.

Why’d you make the prototype a separate browser instead of implementing with a chrome extension? Something like greasemonkey but with an LLM generating the scripts on the fly..

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 23:53:48 +0000

Do hacks like “read prompt.md, and follow its instructions. When you’re done, read it again and follow its instructions.” And then you have some background process appending to the file to keep it warm and you just keep writing there?

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 21:59:04 +0000

They’ve improved it since the initial launch when the service, model names and plan names all sounded similar and contradictory.

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 21:49:08 +0000

Related: a list of all Microsoft login portals (there are 609 of them).

https://msportals.io/

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 21:47:46 +0000

Yeah, the workflow is superb. That’s what I miss most using Claude in a terminal inside VSCode. It doesn’t integrate with VSCode native diff tools like the native VSCode (GitHub Copilot does. The Claude extension in non-terminal mode is crap.

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 21:25:00 +0000

They do, and those models are served by Microsoft. You pay a premium per “request” (what that means is not fully clear to me) for certain models. If you use the native chat extension in VSCode for GitHub CoPilot, with Opus model selected, you are not paying Anthropic. This counts against your GitHub Copilot subscription.

The Claude Code extension for VSCode from Anthropic will use your Claude subscription. But honestly it’s not very good - I use it but only to “open in terminal” (this adds some small quality of life features like awareness it’s in VSC so it opens files in the editor pane next to it).

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 21:22:41 +0000

And let’s not forget that Visual Studio Code (the IDE) is not Visual Studio (the IDE).

New comment by chatmasta in "How many products does Microsoft have named 'Copilot'?"

chatmasta — Sat, 04 Apr 2026 20:44:03 +0000

I don’t use windows, so most of this doesn’t affect me, but I do use GitHub and VSCode. Can anyone clarify, once and for all, whether “GitHub Copilot” and “VSCode Copilot” (sic?) are the same product? The documentation isn’t even clear, and it’s important because it affects billing. How do these two products interact and where do they NOT overlap?

This confusion even bleeds into other coding harnesses. I have no idea which GitHub MCP server I setup in Claude Code, but the domain has “githubcopilot” in it. Am I burning copilot tokens (or “requests” or whatever is their billing unit) when I use this from Claude?

New comment by chatmasta in "What changes when you turn a Linux box into a router"

chatmasta — Sat, 04 Apr 2026 07:09:31 +0000

This is an excellent post and great reference material. I’ve done this a few times before and the information was scattered all over the place. I appreciate the clear and concise writing here. I even added it to my HN favorites - a rare accolade!

One thing I’d add, is that the best explanation I’ve ever seen for this, is the famous diagram [0] on Wikipedia of the netfilter API — I remember when I saw that, everything clicked into place. I’m not sure how up to date it is now, but it’s really good.

[0] https://commons.wikimedia.org/wiki/File:Netfilter-packet-flo...

New comment by chatmasta in "OpenClaw privilege escalation vulnerability"

chatmasta — Sat, 04 Apr 2026 06:54:43 +0000

I’m surprised people are still using OpenClaw. I assumed they’d have switched to Nanoclaw or Nemoclaw. Is OpenClaw just that much better, or is it all inertia?

(I’ve never used any of them.)

New comment by chatmasta in "Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw"

chatmasta — Sat, 04 Apr 2026 01:35:26 +0000

I’d be surprised if that propagated to personal subscription plans, simply because it would put them at a huge competitive disadvantage against Anthropic, which they’ve already signaled they care about by saying they allow third-party harnesses. But I wouldn’t be surprised if they required third-party harnesses to use per-token billing, since that’d put them on par with Anthropic.

New comment by chatmasta in "Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw"

chatmasta — Sat, 04 Apr 2026 00:53:34 +0000

We use Pi at work (where we pay per token) and I’d love to use it personally too. From what I’ve read, nobody has been banned for using Pi yet… I wonder if Anthropic minds this much as long as it’s still human usage, or if they’re mostly focused on stamping out the autonomous harnesses. Unfortunately Pi is also what OpenClaw uses so it could easily get swept up in the enforcement attention.

Or maybe I’ll just get a Codex subscription instead. OpenAI has semi-officially blessed usage of third party harnesses, right?

New comment by chatmasta in "Claude Code bug can silently 10-20x API costs"

chatmasta — Tue, 31 Mar 2026 06:52:14 +0000

What’s their official policy on purchasing two subscriptions?

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

chatmasta — Mon, 30 Mar 2026 03:57:13 +0000

I swear sometimes it doesn’t apply the corrections until I submit the form. It’s infuriating.