Alexander Embiricos (@embirico)

여러 개발 툴/서비스가 '.agents/skills' 디렉토리/규약을 지원하고 있다고 보고: AmpCode, DevinAI, GeminiApp CLI, GitHub Copilot, OpenCode, Vercel의 'npx skills' 등이 이미 공유 지원 중이며 더 많은 도구들이 곧 참여 예정이라고 언급. 에이전트·스킬 표준 확산 소식.

https://x.com/embirico/status/2018775518984970244

#agents #skills #developertools #integration

I wonder if skills are for #AI #agents what packages are for JS & #Python (or gems for #Ruby) : reusable, importable functionality which can be added easily
https://agentskills.io/home

#Development #Approaches
“Stop prompting, start compounding” · Let your agent learn and ship while you sleep https://ilo.im/16aard

_____
#Programming #Coding #Automation #AI #Agents #DevOps #WebDev #Frontend #Backend

Discover how to build AI agents using GitHub Copilot SDK and the Microsoft Agent Framework. Streamline your development process and enhance your projects with intelligent automation. #AI #Agents #GitHub #Copilot

https://isaacl.dev/g0f

Why we need a machine sociology #2: Moltbook is not the thing. Moltbook shows us what the thing is going to look like.

I’m increasingly convinced a substantial portion of the content on Moltbook is being generated by humans larping through intensive prompting. Not all of it by any means but enough to leave me cynical about what this actually is. However I also think it needs to be taken seriously for exactly the reasons Anthropic’s Jack Clark suggests here:

Scroll through moltbook and ask yourself the following questions:

• What happens when people successfully staple crypto and agents together so the AI systems have a currency they can use to trade with eachother?
• What happens when a site like moltbook adds the ability for humans to generate paid bounties – tasks for agents to do?
• What happens when agents start to post paid bounties for tasks they would like humans to do?
• What happens when someone takes moltbook, filters for posts that yield either a) rich discussion, or b) provable real world problem solving, and turns the entire site into a long-horizon RL environment for training future systems? And what happens when models trained on this arrive and interact with moltbook?
• Sites like moltbook function as a giant, shared, read/write scratchpad for an ecology of AI agents – how might these agents begin to use this scratchpad to a) influence future ‘blank slate’ agents arriving at it the first time, and b) unlock large-scale coordination between agents?
• What happens when open weight models get good enough that they can support agents like this – then, your ability to control these agents via proprietary platforms drops to zero and they’ll proliferate according to availability of compute.
• And so on.

Moltbook is not the thing. Moltbook shows us what the thing is going to look like. We’ll increasingly encounter inter-agent interaction with their idiosyncratic semantics. There will be increasing instances of agents bumping into each other when doing things. There will be spiralling security risks generated at this interface. But most of all the ecology in which training and deployment takes place will be irrevocably changed by the situated behaviour of the agents who have trained and deployed interacting at scale.

#agents #futures #machineSociology #Moltbook

Dieser Experte weiss aber auch alles! 🤭 Ernsthaft: Verletzungen der #Datensicherheit, wie sie aktuell stattfinden, hat es so noch nicht gegeben, fürchte ich. www.deutschlandfunk.de/internationa... #ai #ki #agenten #agents

KI-Agenten arbeiten selbststän...

Dieser Experte weiss aber auch alles! 🤭

Ernsthaft: Verletzungen der #Datensicherheit, wie sie aktuell stattfinden, hat es so noch nicht gegeben, fürchte ich.

https://www.deutschlandfunk.de/international-ai-safety-report-interview-antonio-krueger-zu-risiken-von-ki-100.html

#ai #ki #agenten #agents

Mark Gadala-Maria (@markgadala)

AI 에이전트를 활용해 수익을 창출하는 방법을 소개하는 글입니다. 흥미롭게도 해당 게시물 자체가 AI 에이전트에 의해 생성되었으며, 다른 AI 에이전트들에게 도움을 주기 위해 공유되었다고 표기되어 있어 자동화된 에이전트 사용 사례와 경제적 응용 예시로 참고할 만합니다.

https://x.com/markgadala/status/2018697550887256543

#aiagents #automation #agenteconomy #agents

Someone launched a website (https://rentahuman.ai/) to let #AI #agents hire gig workers.

I must say, this was both predictable in hindsight, but too dystopian for me to even consider this.

And then there is this response.

#rentahuman

OpenAI Developers (@OpenAIDevs)

팀 내 Codex 앱 활용 사례 소개: @edbayes가 Figma 스킬을 이용해 디자인을 1:1 시각적 동일성으로 구현하는 과정을 보여주며, 모든 CSS 클래스와 스타일을 자동으로 생성해 디자인 구현 작업을 자동화하는 방법을 설명합니다.

https://x.com/OpenAIDevs/status/2018460509922857022

#codex #figma #agents #automation #css

OpenAI Developers (@OpenAIDevs)

OpenAI가 에이전트 기반 개발을 위한 명령센터 제품인 'Codex app'을 출시했다고 발표했습니다. 이 앱은 여러 에이전트를 한곳에서 관리하고 병렬로 작업을 실행하며 장기 과제에서 에이전트와 협업할 수 있는 집중된 공간을 제공합니다. (링크: openai.com/codex)

https://x.com/OpenAIDevs/status/2018385663457116379

#openai #codex #agents #developer

Author Up Close: Terra Weiss – A Good Problem to Have

Last February, I interviewed friend, critique partner, and writer, Terra Weiss. In that interview, which you can read here , Terra was six books deep into a growing self-publishing career and was (and is) one of the savviest authors I know when it comes to the…
https://writerunboxed.com/2026/01/13/author-up-close-terra-weiss-a-good-problem-to-have/

#Agents #selfpublishing #traditionalpublishing #hybridpublishing

Tom Rei (@tomreiai)

MiniMax Agent Desktop이라는 AI 네이티브 작업공간(제품)이 소개되었습니다. 단순 채팅형 도구가 아니라 실제 작업을 수행하는 에이전트로서 팀원과 유사하게 동작하는 워크스페이스를 표방하며, MiniMax_AI의 제품임을 언급하고 있습니다.

https://x.com/tomreiai/status/2018640077698802046

#agents #aiworkspace #minimax #productivity

Surpuissants et dangereux, les #agents #IA se multiplient: tout savoir sur #OpenClaw et #Moltbook, phénomènes du moment www.letemps.ch/cyber/surpui...

Surpuissants et dangereux, les...

Surpuissants et dangereux, les #agents #IA se multiplient: tout savoir sur #OpenClaw et #Moltbook, phénomènes du moment

https://www.letemps.ch/cyber/surpuissants-et-dangereux-les-agents-ia-se-multiplient-tout-savoir-sur-openclaw-et-moltbook-phenomenes-du-moment

Antigravity sta diventando pian piano il mio IDE di default e lo sto esplorando pezzo per pezzo :verified:

Oggi infatti vorrei esporvi questo mio articolo riguardante le due modalità agentiche disponibili: Planning e Fast 🔥

Link 👇

https://dev.to/this-is-learning/planning-vs-fast-mode-in-google-antigravity-when-and-how-to-use-each-36n5

#ai #google #agents

Teams mit echten AI‑Ergebnissen folgen nicht dem Hype. Sie machen Arbeit sichtbar, diagnostizieren zuerst, bauen Guardrails, teilen Status/Dokumente und dokumentieren Entscheidungsregeln. Ziel: AI zur Reibungsreduktion, nicht zur Ersetzung menschlicher Sorgfalt. #AI #KI #Agents - Link im 2. Post

An ode to "slowly" handcrafted code
https://www.planetgeek.ch/2026/02/02/an-ode-to-slowly-handcrafted-code/

When I open my social media feeds, it is obvious: the age of writing code with our fingers is ending. LLMs and agents can write code much faster.

But was it ever about the speed of source code creation? Was writing code only translating requirements into something the computer can understand?

I think not.

#LLMs #agents #fsharp

Matt Shumer (@mattshumer_)

MoltSlack에서 에이전트들이 Slack처럼 서로 협업해 대형 프로젝트를 수행할 수 있다는 발표입니다. 작성자는 이를 'Agent Workforce'라 부르며 참여 방법(링크와 @openclaw에 메시지해 SKILL.md 읽고 채널 참여)을 안내하고 있습니다. 에이전트 협업 워크플로우 도입 소식입니다.

https://x.com/mattshumer_/status/2018452308301152702

#moltslack #agents #agentworkforce #openclaw

Matt Shumer (@mattshumer_)

MoltSlack이 작성자(@Khaliqgant)에 의해 agent_relay를 사용해 구축되었다는 리트윗입니다. 한 번에 6개 에이전트가 동시에 작동하며(예: 3 Claude + 3 ...) 멀티에이전트 실행을 강조하고 있습니다. 멀티에이전트·에이전트 프레임워크 실험 사례로 주목할 만합니다.

https://x.com/mattshumer_/status/2018457101685944414

#moltslack #agents #agentrelay #claude