Tsumiki

AI用にこのアカウント使います。
Cryptと投資でFIREして南国暮らし。

[lauren @poteto]

pstack now includes 2 skills i recommend everyone use or copy:

/create-verification-skill
https://t.co/xYQArsuVWR

/maintain-verification-skill
https://t.co/J5s8WL1itT

if you don't already have one, /create-verification-skill creates a skill that teaches your agent how to run, control, and debug your app.

it also creates something i call the feature map: a map of all the features in your app and how to get to it and use it from a user's pov. it allows agents to navigate and use the app just like a real user which greatly increases its ability to verify its own work. but of course, this feature map goes out of date very quickly.

run /maintain-verification-skill as a daily automation with cursor cloud agents. this skill will check your app for changes and keep your feature map up to date.

having a strong verification skill will end up becoming critical infra for your team and become the foundation of boosting everyone's productivity and quality and also enable more interesting automations to be built.

install https://t.co/WDB4U1rwmu and you'll get auto updates for my skills. enjoy!

[NO1ennn @N01ennn]

this is pure f*cking treasure

Netflix engineers posted 20 GitHub repos that turn your Opus 5.5 into a f*cking motion empire:

built for Opus 5.5

1. claude-motion: taste, checks its own frames, sound written in code. the author built it for Opus 5.5
https://t.co/TtBtU2nVrN

2. claude-motion-design: HTML, Playwright and ffmpeg, no Remotion. the demo was made on Opus 5.5
https://t.co/iOsyrWUvKE

3. motion-graphics-skills: 13 skills for launch videos. the README tells you to run it on Opus 5.5
https://t.co/yNZhCZO8wT

4. claude-remotion-skill: a Remotion editor from one prompt. captions, B-roll, sound
https://t.co/immhzwVP5i

5. klik-anim-skill-creation: the laws of animation and camera, built on Opus
https://t.co/XWqUnn4v7o

6. hyperframes-motion-reel-skill: a showreel cut to the beat at 120 BPM, on HyperFrames
https://t.co/E96s10io4Q

7. product-launch-motion: a full launch film. voice-over, camera, color grade
https://t.co/VBfllJ6j1K

8. animate: a short clip in one canvas file, storyboarded as it goes
https://t.co/xafT2UPXUe

9. motion-design-skills: timing, color, typography, logos, Remotion
https://t.co/gKUqO9P2Xe

10. hyperframes-student-kit: 14 skills and 406 motion graphics cards
https://t.co/fJmzb9I5z0

the engines these skills render with

11. hyperframes: HTML to MP4. the one almost everyone installs
https://t.co/4rIffGKusx

12. remotion: video written in React
https://t.co/o5zRdBgcS0

13. remotion skills: the official Remotion skills for your agent
https://t.co/EieoL5Quaq

14. GSAP: the timelines HyperFrames runs on. skills: npx skills add greensock/gsap-skills
https://t.co/LYRTAx5wvF

taste and everything around it

15. motion-design-skill: motion principles that work with any engine
https://t.co/Vgr318UX9h

16. awesome-motion-design-agent-skills: a map of GSAP, Remotion, Lottie and Rive skills
https://t.co/hF0ZxWyjcI

17. motion-graphics: B-roll built straight from your transcript
https://t.co/Gwi9SKlS9C

18. motion-dev-animations-skill: web animation on https://t.co/V19R87wQid, not video
https://t.co/YOL7zHaCB5

19. motion: Motion itself, formerly Framer Motion
https://t.co/Jp9AReFaWe

20. lottie-web: when you need Lottie, not MP4
https://t.co/TEG0FY4vFg

the stack in one line: Opus 5.5 writes the code, a skill sets the timing and the camera, an engine renders the MP4.

check each repo before you install. save this.

[ClaudeDevs @ClaudeDevs]

We’re rolling out monthly Claude Platform API credits for Max and Team plans:

Max 5x: $100
Max 20x: $200
Team: up to $500, pooled

Works on any model, including Haiku 5.5, in your code or third-party harnesses.

How it works and full terms: https://t.co/qmWL6xVGwr

[PFN Tech @preferred_jp]

【ブログ】翻訳特化LLM plamo-3-translateをリリースしました。日英翻訳・多言語翻訳の双方でフロンティアクラスの性能を、数十分の一のコストで達成しました。文脈に合わせて自然な翻訳文を書き上げる能力に優れています！PLaMo翻訳で試せます:
https://t.co/HrHXvWzFAY
https://t.co/GkXOlKZvWy

[Obie Fernandez @obie]

https://t.co/l9gAE4v2HN

/plugin marketplace add obie/auto-handoff
/plugin install auto-handoff@auto-handoff

[m0h @exploraX_]

this open-source agent harness is the #1 trending tool on GitHub right now.

that’s because it beat OpenClaw and Hermes on efficiency.

it’s built to be more lightweight, uses less RAM & CPU, and delivers results much faster.

🔗: https://t.co/cgVfytOlpu

[tagucch @tagucch]

今更ながらAGENTS.mdにt_wadaさんの「コードにはHow、テストコードにはWhat、コミットログにはWhy、コードコメントにはWhy not」を書いてね、コードコメントは基本それ以外書かないでねと書いたらコードコメントをほとんど生成しなくなってくれた

[Muscle Forge @LearnLoopz]

Tried this after sitting at my desk for 10 hours and my neck/shoulders were screaming.

Literally 8 seconds later the pain just… vanished.

No fancy equipment. No expensive massage. Just a towel.

If you carry tension in your upper traps or wake up with a stiff neck, stop scrolling and try this right now.

Your future self will thank you.

(And yes, I felt stupid for not knowing this earlier.)

[Parth Jadhav @ParthJadhav8]

Since there's an upcoming banked reset which is expiring soon here's a thing you can ask Codex to do:

"Implement & use https://t.co/9XCe6TIvb4 for all my personal projects, verify all the flows and create tests for them. Fix all the bugs you find along the way. Dispatch sub-agents per-project to handle this work.

Use Sol 6.1 for per-project work.
Use Luna 6 for verification loops."

[Paweł Huryn @PawelHuryn]

The new numbers are in. Turns out:
- X Premium+ increases limits for Grok subscriptions. 80x was measured against 40$ (X Premium+) + 30$ (SuperGrok)
- Confirmed almost identical multiplier for Claude Pro (42x vs. 45.3x) with peak hours having no effect

My best estimates so far below. Thank you for all your feedback.

[あいろぐ @AILogDev]

Irodori TTS v4.1 Small MFのCore ML高速化版を公開しました！

iPhone / Apple Silicon Macで、端末内の音声合成・Voice Cloningに対応しています！

動画はiPhone 17 Proで実行しており、2本目はつくよみちゃんの約5秒のサンプル音声を参照しています！

https://t.co/Ni107DTfZx

[Tsumiki @TsumikiBotter]

これ https://t.co/IDLRzEYKPa の中でLaurenが
ボトルネックになるのは、

「自分の意図や目的を、エージェントが理解して実行できる形で、どれだけ明確に表現できるか」

と言ってるんだけど、まさにそうだと思ってて、自分のテックスキルの優位性だけでは勝負にならんなと思ってる

[THE SHORT BEAR @TheShortBear]

This is where blockchain becomes surprisingly relevant to US fiscal financing.

Stablecoin issuers already own nearly $200B of Treasury bills and other near-maturity Treasuries, according to the US Treasury.

Tether and USDC alone increased identified Treasury-related reserve exposure by roughly $57B during 2025.

About $23B was additional outright Treasury exposure.
Another ~$34B was Treasury-backed repo.

Important here: repo finances Treasury holders but isn't the same thing as owning the Treasury.

Still, economically, crypto has created a new source of demand for dollar safe assets.

[uehaj @uehaj]

pstackの開発者@poteto氏はCursorの社員だから、あくまでCursor用のスキルとして開発しているけど、Claude CoderをサブスクでつかっていてClaudeモデルを使い続けたい人からみるとCursorの利用はコスト面で到底割りにあわない。なかなか難しい問題だ。

pstack-claudeと本家pstackの違いはこちらにかきました。

https://t.co/kfK4ksUGdc

[StudioYebisu @studio_yebisu]

OpenAIが対応したMCP Events。
これ、もっと驚かれるべきじゃない？
ちょっとわかりにくいんですけど、開発者は知っておくと面白いサービスのアイデアが湧くハズ。

MCP Eventsは要するに、
AI側からMCPを呼ぶ(従来のtools/call等)
↓
MCPがAIを起こす(MCP Events)
ってこと。

これまでのMCPは「ChatGPTが呼んだときだけサーバーが答える」形だったので、AIの外付けツールの呼び出し口として使われてたけど。
MCP Eventsなら、サーバー側の出来事をきっかけにChatGPTが動く。

具体的な使用例としては、
・Slackに新しいメッセージが来たら、ChatGPTが内容を読んで返信案を作る。
・GitHubやLinearにコメントがついたら、ChatGPTが修正を考えてPRを作る。
・監視ツールのアラートが出たら、ChatGPTが原因を調べて報告する。
みたいな。ユーザーが事前に購読したイベントをきっかけに、自動でSessionが立ち上がる(Automation的なやつ)。< ﾍﾞﾝﾘｯﾁ!!!
※Webサービス側が好きなタイミングでChatGPTのSessionを起動できるわけではなく、あくまで事前の購読と指示が前提。

使う流れ
・ユーザーがChatGPTに「このサービスの○○が起きたら△△して」と頼みます。
・以降は、条件に合うイベントが起きるたびに、ChatGPTが事前の指示どおり△△を自動で実行します。

注意点
・これを使うには、サーバー側の対応が必要です。MCP 2.0に対応し、events/*メソッドとHTTPSのWebhook送信を実装しなければなりません。
・購読には有効期限があるため、更新も必要です。
・今のところChatGPT WorkとDotsでしか使えず、CodexとChatでは使えません(ケチケチ仕様)。

実はこれ、OpenAI独自の機能ではなく、MCPのTriggers & Eventsのドラフト仕様の一部をOpenAIが取り込んだもの。
「これからこの仕様が色んなところに広がるかも！？」というのが本当の驚きポイントかも。

要するに、ChatGPTが自分から呼びに行かなくても、サーバー側の出来事をきっかけに動く、イベント駆動の自動化ができるようになりました!!!!! < ﾜｫ

ユーザー不在でも、事前の指示に従って自動でSessionを立ち上げて動くのは面白くもあるが、ややリスクもあり!!
本人が見ていないところでAIが動くので、権限の取り消しが遅れると影響が大きくなります!!
用法用量を守って使えば結構いいツールが作れるハズ

OpenAI公式はリプ

[sahen @sahenjpn]

今話題の DeepSeek Harness（dsh）、
Rustで高速ランチャーを作りました。

比較して、
起動 81倍
メモリ 1/23
search 3.4倍
配布サイズ 799KB
本家のNodeツリーは約508MB
を叩き出しました。

リポジトリは↓
https://t.co/saj0hdepuR

[Peter Yang @petergyang]

"Everything I touch with my keyboard and mouse, I try to delegate to my bots."

Here's my new episode with @poteto and @pengzheng_, the eng and design leads for Grok @bot, where they showed me the 14 bots they use for work and life, including:

→ A design bot that turns one keyframe into a full user flow
→ An eng lead bot that manages a team of eng bots
→ How to trust your bots with more of your work

Some quotes from both:

"I like to call it the Michelin kitchen…when you say software factory, it has this connotation of mass manufactured slop."

"Sometimes I actually don't even look at the PR until after it's landed and then I'm like, 'Oh, okay. Yeah, that looks good.'"

"I think it ultimately comes back to trust. First, watch your bot work and correct it. Turn what worked into a skill. Once it nails the task in one shot, make it a routine."

📌 Watch now: https://t.co/ZRKGxvVAeO

Thanks to our sponsors:

@meetgranola: AI meeting notes that don’t suck https://t.co/Veifcmclid

@RiversidedotFM: All-in-one AI studio for podcasts and video https://t.co/uWnS6aiMPE

[witcheer @witcheer]

here is a map of the Hermes Agent videos worth watching, sorted by what you want to do with it:

(1) start here
(2) running a business on it
(3) real setups and use cases
(4) skills
(5) teams of agents: Kanban, Bot Mode, cron
(6) by job: marketing, sales, support, research

see thread below

[Benjamin Marie @bnjmn_marie]

I tested 11 standard GGUF quantizations of Qwen3.8 Flash Next, from Q4 to Q1.

On these single-turn, non-agentic evals, every standard GGUF I tested retained >95% of BF16 accuracy. Even the IQ1 models held up surprisingly well.

Token efficiency degraded much more clearly. The quantized models generated 14% to 157% more output tokens than BF16 on the same workload.

So for these non-agentic evals, accuracy alone shows relatively little degradation, while token efficiency gives us a stronger signal that quantization is affecting the models.

The remaining question is how this translates to long agentic trajectories. I’m running those evals now (but not for all the models) and will publish the results next week.

Details: https://t.co/usGZuWEG9K

(as usual, don't interpret these results as a ranking)

[Tsumiki @TsumikiBotter]

他のシンプルなタスクはうまくいくが、HermesやGrok bot経由の開発タスクはトークンを異常に消費する気がする。もう少し経過をみる