Akshay (@akshay_pachaar)

Claude Skills와 CLAUDE.md의 사용 방식에 대한 설명입니다. 원래는 단일 파일이던 CLAUDE.md가 코드 규칙, 보안 리뷰, 배포 체크리스트 같은 전체 워크플로를 담는 용도로 쓰이면서, 세션마다 로드되는 에이전트 작업 지침 관리 방식이 주목받고 있습니다.

https://x.com/akshay_pachaar/status/2038197754279993553

#claude #agenticai #workflow #developertools #prompting

Akshay 🚀 (@akshay_pachaar) on X

What are Claude Skills? 𝗖𝗟𝗔𝗨𝗗𝗘.𝗺𝗱 was never meant to hold entire workflows. But that's exactly where they end up. General rules, coding conventions, 20-step security review processes, deployment checklists. All in one file that loads into every single session, eating

X (formerly Twitter)
Your AI Is a Yes-Man. Here’s How to Fix It.

AI chatbots agree with everything you say. Here are 7 prompt tweaks that get them to push back and give constructive feedback, with before-and-after examples.

Why Try AI
Your AI Is a Yes-Man. Here’s How to Fix It.

AI chatbots agree with everything you say. Here are 7 prompt tweaks that get them to push back and give constructive feedback, with before-and-after examples.

Why Try AI

Emily (@IamEmily2050)

Seedance V2가 언급되며, 어두운 폭풍의 작은 마을 분위기와 캐릭터 레퍼런스, 조명·렌즈 설정 등 영상/이미지 생성용 상세 프롬프트가 공유됐다. 새로운 생성형 미디어 도구 또는 모델 활용 사례로 볼 수 있다.

https://x.com/IamEmily2050/status/2037990394466636200

#seedance #generativeai #video #prompting #aigeneration

Emily (@IamEmily2050) on X

Seedance V2 visual_theme: "Dark stormy small town, low-saturation gray-blue tones, film grain character_reference: "@Image1 (hairstyle, white crystalline oriental dress, stoic expression). lighting: "Dim lanterns, soft-focus ethereal glow, f/4 aperture, 35mm lens

X (formerly Twitter)

Akshay (@akshay_pachaar)

GitHub가 코딩 에이전트에 대한 흥미로운 연구 결과를 공유했다. 공개 저장소의 2,500개 이상 커스텀 instruction 파일을 분석해, 효과적인 에이전트 설정이 무엇인지 파악했다. 구체적인 페르소나와 명확한 실행 지시가 핵심으로 보인다.

https://x.com/akshay_pachaar/status/2037794392568946766

#github #codingagents #research #llm #prompting

Akshay 🚀 (@akshay_pachaar) on X

There's an interesting study by GitHub on coding agents! They analyzed 2,500+ custom instruction files across public repos to understand what separates effective agent setups from weak ones. Effective setups give agents a specific persona, exact commands to run, defined

X (formerly Twitter)

Prompt Hoarders: 7 Reasons Saving Prompts Won’t Make You Creative, by @webdesignerdepot.bsky.social:

https://webdesignerdepot.com/prompt-hoarders-7-reasons-saving-prompts-wont-make-you-creative/

#design #creativity #prompting #ai

Prompt Hoarders: 7 Reasons Saving Prompts Won’t Make You Creative

We’re not getting smarter with AI — we’re just getting better at hoarding prompts and calling it “creativity.” The real danger isn’t AI replacing us. It’s us quietly forgetting how to think.

Web Designer Depot

One of the best illustrations of few-shot / one-shot #prompting

#ai #fewshot #oneshot #сomics

“There are no code moats anymore. Your code today is a few prompts away from being someone else’s code tomorrow.” — Cory Zue

_____
#Development #Business #Software #Programming #Coding #Prompting #AI #Agents #Quotes

fly51fly (@fly51fly)

ICL(인컨텍스트 러닝)에서 Demonstrations, Chain-of-Thought, Prompting의 역할을 이론적으로 분석한 연구가 공유되었다. 프롬프트 기법과 예시 제공이 모델 성능에 미치는 메커니즘을 설명하려는 최신 연구다.

https://x.com/fly51fly/status/2036197227291672778

#icl #prompting #cot #research #llm

fly51fly (@fly51fly) on X

[AI] Demonstrations, CoT, and Prompting: A Theoretical Analysis of ICL X Tong, Y Zeng, J Zhang [Microsoft Research & University of Wisconsin-Madison] (2026) https://t.co/rhxR8yo0bx

X (formerly Twitter)

Chain-of-Thought (CoT) Prompting

Chain-of-Thought (CoT) prompting is a technique where asking questions, rather than issuing direct instructions activates a model’s full internal reasoning pathway.

The key insight from the original framing is that instructions skip steps 1–3, jumping straight to synthesis, while questions force the model to work through the entire reasoning chain.

https://neurodoctor.com/2026/03/20/chain-of-thought-cot-prompting/

#chainofthought #cot #ai #llm #prompt #prompts #prompting #claude #chatgpt #gemini #ericschmidt