amyst ∴

notes, bugs, ideas, and other suspicious activity.

[amyst ∴ @amyst007]

This is great

[amyst ∴ @amyst007]

@shafu0x Is this from personal experience or just from looking at other people's laptops?

[amyst ∴ @amyst007]

@mattyp is that zed code editor?

[amyst ∴ @amyst007]

Not the 2-3x some people are claiming, but still decent.

Anyone else testing it? How are your numbers looking?

[amyst ∴ @amyst007]

Quick question for Codex + Claude Code users:

Have you tried Nerfguard yet?

Want to know:
- How much token savings you're actually seeing
- Whether it affects quality/reasoning on complex work
- Any issues with reliability

It’s still pretty new so real feedback would be useful.

https://t.co/nbHYTTTxDC

[amyst ∴ @amyst007]

@kai_fell Never had any such issue, been using it for a long time as my daily driver now.

[Sakana AI @SakanaAILabs]

Introducing Sakana Fugu: A full multi-agent orchestration system accessible via a single model API.

Our ‘Fugu Ultra’ model matches the performance of Fable and Mythos, delivering frontier capability without the risk of export controls.

Try it: https://t.co/hhO6qTawgb 🐡

[amyst ∴ @amyst007]

Is Claude Code actually usable on the Pro Plan? Just 1-2 prompts already take up 40-50% of the token limit on Opus 4.8 Medium. I've been implementing skills & MCPs like Cavemen and Serena to save tokens, but it still feels heavy. Am i doing something wrong?

[Indie Game Joe @IndieGameJoe]

These indie devs are making an open-world survival FPS where you plant alien seeds to reshape the world

- Must have the right gear for cold weather
- Unlock powerful alien abilities
- Hunger slows you down

It's called HAEX. Would you play this?

[andi (twocents.com) @Nexuist]

Every year I sit down with my mother to explain how to use her phone and every year Apple sends 750 engineers into their little labs underneath their demonic Cupertino crop circle to come up with new and exciting ways to confuse her

[amyst ∴ @amyst007]

@mattpocockuk Hi Matt, does /teach work better with Claude code or codex?

[Anthropic @AnthropicAI]

Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor.

It’s happening faster than we thought, and the implications deserve greater attention. https://t.co/OVVPJO7VQx

[amyst ∴ @amyst007]

Beautiful.

[ThePrimeagen @ThePrimeagen]

I am remembering right now for my own sanity. There is so much peace in deep thought instead of some insane frenetic agent orchestration fever dream

[amyst ∴ @amyst007]

@claudeai I hope it also makes those sounds on interaction as seen in the clip.

[amyst ∴ @amyst007]

starting from zero.
please lower your expectations accordingly.