LYRA Capture

Je retweet pour moi, si ça t'aides tant mieux pour toi frero

[Mr. Buzzoni @polydao]

Claude Code tip: once Opus 5.5 is your main model, stop letting your Fable 5.1 quota go to waste

put it on call with /advisor

run /advisor fable

Opus 5.5 keeps doing the work
Fable 5.1 sits on the sidelines, reads the whole session, and steps in at three moments:

→ before a plan: is this right?
→ when the same error comes back: am I going the wrong way?
→ before "done": did I miss anything?

Fable 5.1 advises. Opus 5.5 writes the code

the same idea sits under Jev engineering: the expensive model stops weighing in on every step and only gets called at the moments that change the outcome

• the full setup
> Opus 5.5 on high runs the main session
> subagent one reads code
> subagent two edits and runs tests
> subagent three looks up docs
> all three on medium
> Fable 5.1 on call

hand the tree and this prompt to Claude Code 👇

"Set up my Claude Code to match this tree:

1. Reuse fitting subagents from ~/.claude/agents and .claude/agents.

> Propose new ones only for missing roles
> Set each to model: opus, effort: medium
> Leave any that set a different model alone and list them

2. Set main session effort to high via effortLevel in ~/.claude/settings.json

3. Check for env vars that disable the advisor (CLAUDE_CODE_DISABLE_ADVISOR_TOOL, DISABLE_TELEMETRY, anything that stops flag fetching) and CLAUDE_CODE_EFFORT_LEVEL, which overrides subagent effort. Report them, don't change them

4. Add a rule to ~/.claude/CLAUDE.md: ask the advisor before a big plan, when an error repeats, and before calling a long task done

Show me the changes first. Don't edit files yet."

↳ https://t.co/pkgl3S02TF

[Nico @nicos_ai]

EL CREADOR DE CLAUDE CODE YA NO ESCRIBE PROMPTS

Boris, la persona que construyó la herramienta, soltó esto en una entrevista:

"Ya no escribo prompts. Escribo loops, y los loops hacen el trabajo. Mi trabajo es escribir loops."

Mientras la mayoría de devs sigue puliendo el prompt perfecto, él ya pasó a otra cosa.

En 30 minutos enseña su setup diario real:

→ Claude Code + loops + workflows dinámicos
→ Su flujo de trabajo de cada día
→ Cómo aplicarlo desde hoy

Vale más que cualquier curso de vibe coding de 500$.

Gratis. Subtitulado a español e inglés.

Y después lee esto: todo lo que necesitas saber sobre loops para aplicar lo que cuenta ↓

Guarda los dos. Este es tu plan del finde.

[Mo Elgaraihy @EngMoElgaraihy]

🚨 مهندسة من قلب Anthropic تفجرها وتسفّه كل ما تعلمته عن كتابة الأوامر: "أنت لست هنا لتكتب Prompts لـ Claude.. وظيفتك أن تبني نظاماً يكتب الأوامر لنفسه!" 🤯🧠

​في 45 دقيقة فقط، كشفت سراً تمنع الشركات الكبرى تسريبه:
كيف تبني Anthropic وكلاء أذكياء يمتلكون ذاكرة حقيقية، يصلحون أخطاءهم البرمجية ذاتياً، ويصبحون أكثر ذكاءً بعد كل عملية تشغيل (Run)! 🔥

​💡 لماذا يعتبر هذا الفيديو الأهم هذا العام؟
لأنه يسحق تماماً كل الكورسات المدفوعة التي تباع بآلاف الدولارات عن الـ AI Agents، ويضع في يدك المعمارية الحقيقية لبناء الـ Autonomous Loops!

​توقف عن كتابة البرومبتات اليدوية المملة وابدأ في بناء أنظمة ذكية تدير نفسها بنفسها..

​احفظ التغريدة، شاهد الشرح فوراً، بالأسفل! 🧵👇

[Sprytix @Sprytixl]

ANTHROPIC LEAKED A FILE WHERE 4 AGENTS BUILD AN AI COMPANY IN 60 MINUTES THAT DOES $100K A MONTH

i hunted this file for over 5 months, and inside it $0 of development gives you an AI business that pulled 70k+ in its first month

product → growth → support → ops

four departments, four pinned versions, and no agent ever reaches into another one's branch

product on Sonnet 5 eats $2.57, growth on Opus 5.5 takes $1.00, support on Haiku 4.5 costs $0.22, ops on Opus 5.5 closes the hour for $0.80

the full hour across all four costs $4.59, and the most expensive department is cheaper than a coffee

in 60 minutes twelve things exist that did not an hour earlier: the app, the API, the docs, the landing page, the ads, the emails, the help centre, billing, metrics and alerts

three per department, and every one is a file, a page or a rule, not a slide with a plan

payroll for four departments like this is $31,000 a month, the tokens are $450

the agents do the typing, deciding what to build is still on you

120 people at $99, 200 at $299, 29 at $999 make $100,651 a month

the middle plan carries more than half the month, and it is what turns this into $1.2M a year

the hour builds the company, the first customers decide whether it is one

save this and paste it into Claude Code - let it stand up these 4 departments for you ↓

[Innopro @innopro_]

THIS IS F**KING UNREAL

a 19 year old makes $453 per DAY, that’s $13,540 per month!!

All from Claude + YT shorts
100% of what he does is in the video.

full breakdown in the article below.

[beamnxw ./ @beamnxw]

this is pure f*cking treasure

20 useful GitHub projects for a system that captures sources, connects ideas, remembers context, and helps agents act on it

your second brain should get sharper after every session

CAPTURE THE INPUT
01 Chubby Skills
▸ https://t.co/InxyRwMZtz
02 OpenWiki
▸ https://t.co/bLeUouBLuu
03 Agent Second Brain
▸ https://t.co/WANfN9tVP4
04 DocMason
▸ https://t.co/rQsMrIDGh0
05 DocsAgent
▸ https://t.co/R8DF9NI0tt

CONNECT THE IDEAS
06 claude-obsidian
▸ https://t.co/uhwe1f5n6X
07 SwarmVault
▸ https://t.co/QNxk0QvLYH
08 sage-wiki
▸ https://t.co/OQIkSucHAA
09 Vault Curate
▸ https://t.co/FeR1r3vLle
10 COG Second Brain
▸ https://t.co/1gtpmSyJuy

REMEMBER WHAT MATTERS
11 Hindsight
▸ https://t.co/3cXNIFAyDN
12 memU
▸ https://t.co/iCoOwoBIT3
13 TencentDB Agent Memory
▸ https://t.co/i2DaJnwGhu
14 agentmemory
▸ https://t.co/3OlbQk5CPq
15 OpenViking
▸ https://t.co/Ydr8K54XxW

PUT IT TO WORK
16 Open Second Brain
▸ https://t.co/DiOBxmK4Gh
17 Second Brain Cloudflare
▸ https://t.co/WlokLe12JV
18 makerskills
▸ https://t.co/Q1RQFhStsL
19 Row-Bot
▸ https://t.co/I1KBQwA7hk
20 MateClaw
▸ https://t.co/N9vGfry4no

the loop:

capture the source → turn it into linked, checkable knowledge → recall the right context → act with it → write the correction back

3 builds I'd explore:

creator: Chubby Skills → SwarmVault → Hindsight → makerskills
researcher: DocsAgent → sage-wiki → memU → Row-Bot
team: DocMason → COG Second Brain → TencentDB Agent Memory → MateClaw

save this, then build a second brain for your business ⭣

[kev @kvm_69]

ESTO ES UNA MINA DE ORO

Encontré un sitio con MILES de skills y plugins gratuitos para Claude.

→ ventas y marketing
→ contenido y programación
→ automatización y trading

Plugins que de normal costarían dinero, aquí gratis.

Listo para usar hoy si usas Claude Code.

[Noha_Ai @Noha_Coder]

80+ AI tools to complete months of work in minutes

1. Research

- ChatGPT
- Copilot
- Gemini
- Abacus
- Perplexity

2. Image

- Fotor
- Dalle 3
- Stability AI
- Midjourney
- Microsoft Designer

3. CopyWriting

- Rytr
- Copy AI
- Writesonic
- Adcreative AI

4. Writing

- Jasper
- HIX AI
- Jenny AI
- Textblaze
- Quillbot

5. Website

- 10Web
- Durable
- Framer
- Style AI

6. Video

- Klap
- Opus
- Eightify
- InVideo
- HeyGen
- Runway
- ImgCreator AI
- Morphstudio .xyz

7. Meeting

- Tldv
- Otter
- Noty AI
- Fireflies

8. SEO

- VidIQ
- Seona AI
- BlogSEO
- Keywrds ai

9. Chatbot

- Droxy
- Chatbase
- Mutual info
- Chatsimple

10. Presentation

- Decktopus
- Slides AI
- Gamma AI
- Designs AI
- Beautiful AI

11. Automation

- Make
- Zapier
- Xembly
- Bardeen

12. Prompts

- FlowGPT
- Alicent AI
- PromptBox
- Promptbase
- Snack Prompt

13. UI/UX

- Figma
- Uizard
- UiMagic
- Photoshop

14. Design

- Canva
- Flair AI
- Designify
- Clipdrop
- Autodraw
- Magician design

15. Logo Generator

- Looka
- Designs AI
- Brandmark
- Stockimg AI
- Namecheap

16. Audio

- Lovo ai
- Eleven labs
- Songburst AI
- Adobe Podcast

17. Productivity

- Merlin
- Tinywow
- Notion AI
- Adobe Sensei
- Personal AI

18. Social media management

- Tapilo
- Typefully
- Hypefury
- TweetHunter

Must follow @Noha_Coder for more.

[Maya @mayaislam_ai]

🚨 STOP USING CLAUDE LIKE A CHATBOT. START USING IT LIKE A COMPLETE AI WORK SYSTEM.

Most people use Claude in one simple way:

Open Claude → write a prompt → get an answer → start another chat.

That’s useful, but it treats a powerful AI ecosystem like a basic question-and-answer tool.

A better approach is to understand the different parts of Claude and use each one for the type of work it is designed to handle.

Think of it as moving from:

“Ask AI a question.”
to:

“Give AI a role inside my workflow.”
Here’s the professional breakdown 👇
Follow @mayaislam_ai for more

[Mo Elgaraihy @EngMoElgaraihy]

🚨 المهندس الذي بنى Claude Code بنفسه يخرج عن صمته في فيديو تاريخي مدته 28 دقيقة يشرح كيف تكتب Prompts تعمل بكفاءة حقيقية! 🤯⚡️

​رأيت كورسات بـ 300 دولار لا تغطي حتى ما كشفه في أول 10 دقائق فقط:
​💡 أسرار حصرية من قلب الكود:
📂 ملفات CLAUDEmd وطريقة استغلالها
🧠 اختصارات الذاكرة
⚡️ الجلسات المتوازية
🎯 أنماط هندسة الأوامر الحقيقية

​المفاجأة؟ الشرح مجاني بالكامل، وموجّه للمطور المحترف، المبتدئ، ولكل من يستخدم Claude منذ أشهر لكن بطريقة خاطئة!

​📌 28 دقيقة س��غير طريقة برمجتك للأبد.. احفظ التغريدة وشاهده الآن! 🧵👇

[Neeraj @neeraj_here15]

Anthropic engineer:

"99% of people use Claude Code like Google, and only 1% are running swarms of self-learning Claude agents

I'm running 100+ agents in a loop. I have Chief agent, PM agents - they manage the whole team"

in a 30-minute workshop, an Anthropic engineer revealed how to get max value from Claude Code at min. cost

this is worth more than another $500 vibe-coding course

watch today, then read how to build self-improving agentic systems with Fable in the article below

[Charlie Hills @charliejhills]

Claude Opus 5.5 can (finally) generate motion graphics

16 motion graphics in Claude Code (my 3 steps):

1. Pick one motion, like a chart that morphs.
2. Show Claude a reference and name every state.
3. Ask for HTML and SVG, then fix it round by round.

Download them all here:

https://t.co/pjMv8ExvWx

Repost ♻️ to help someone in your network.

[CyrilXBT @cyrilXBT]

Your CLAUDE. md is making Opus 5.5 DUMBER.

Lines like "double check your work" and "think step by step" were written for OLD models.

On Opus 5.5 they slow it down and make the output worse.

Lance Martin at Anthropic just shipped the fix.

Open Claude Code and type:

/claude-api prompt-audit

It scans your skills, CLAUDE. md and prompts.

Cuts the lines that hold the model back.

Rewrites the rest with the official Opus 5.5 guidance.

5 minutes. Run it TONIGHT.

Bookmark this before you forget.

[Dan McAteer @daniel_mac8]

Claude Coders: Run this prompt for Opus 5.5 *NOW*.

'/doctor prompt-audit'

Different from a viral post I had earlier this week because it prioritizes stale paths, obsolete commands, and conflicting instructions.

[Fran Pradas | Ingeniero de IA @franpradasAI]

🔴 TU PAGINA WEB HECHA CON IA DA ASCO

Haz que se vea x1000 veces mejor así: ↓↓

[Mo Elgaraihy @EngMoElgaraihy]

🚨 مطور غضب من بطء تشغيل الـ Local AI على كارت 12GB.. فقام ببناء محرك استنتاج (Inference Engine) خاص به من الصفر وحقق قفزة مرعبة! 🤯💻⚡️

​كان يحصل على سرعة متواضعة ~15 tps فق�� عند تشغيل نموذج Qwen3.8-Flash-Next على كارت RTX 5070.. الوضع لم يعجبه إطلاقاً، فقرر نسف القواعد وبناء محركه المفتوح المصدر "Strata"!

​النتيجة صدمت مجتمع الـ Local AI بالكامل:
🐌 llama.cpp العادي ⬅️ ~15 tps
🚀 محرك Strata الجديد ⬅️ يقفز حتى 65.1 tps! (أكثر من 4 أضعاف السرعة!) 🔥

​والصدمة الأكبر؟ يعمل على جهاز ألعاب منزلي عادي جداً وليس Workstation خارقة:
🎮 كارت: RTX 5070 12GB
🧠 رام: 64GB DDR5
⚙️ معالج: Ryzen 5 7600 على نظام Windows!

​الأرقام الجنونية على سياق ضخم (128K Context):
⚡️ كوانت Q2_0: سرعة 65.1 tps وسرعة قراءة برومبت 543 tps!
⚡️ كوانت IQ2_XS: سرعة 52.0 tps وقراءة 472 tps!
⚡️ كوانت IQ3_XXS: سرعة 44.8 tps وقراءة 414 tps!

​المطور صمم المحرك خصيصاً للتناغم بين ذاكرة الكارت ورام النظام عبر تقنية RCO-GSQ quants.. والأجمل أنه طرح المشروع مفتوح المصدر (Open-Source) للجميع مجاناً! 🔓

​عصر تشغيل النماذج الضخمة محلياً وبسرعة البرق على كروت عادية بدأ رسمياً..

​رابط المستودع في الـتعليقات، احفظ التغريدة فوراً وجرب المحرك بنفسك! 🧵👇

[Médéric @Med_34000]

🥳 𝗡𝗩𝗜𝗗𝗜𝗔 𝘁𝗲 𝗱𝗼𝗻𝗻𝗲 𝗮𝗰𝗰𝗲𝘀 𝗴𝗿𝗮𝘁𝗼𝘀 𝗮 𝟰 𝗴𝗿𝗼𝘀 𝗺𝗼𝗱𝗲𝗹𝗲𝘀 𝗜𝗔 𝗲𝘁 𝘁𝘂 𝗽𝗲𝘂𝘅 𝗹𝗲𝘀 𝗯𝗿𝗮𝗻𝗰𝗵𝗲𝗿 𝗱𝗶𝗿𝗲𝗰𝘁𝗲𝗺𝗲𝗻𝘁 𝗮 𝗛𝗲𝗿𝗺𝗲𝘀 𝗼𝘂 𝗻 𝗶𝗺𝗽𝗼𝗿𝘁𝗲 𝗾𝘂𝗲𝗹 𝗼𝘂𝘁𝗶𝗹 𝗰𝗼𝗺𝗽𝗮𝘁𝗶𝗯𝗹𝗲 𝗢𝗽𝗲𝗻𝗔𝗜.

Pas de CB. Juste un compte NVIDIA Developer et une cle API.

Les modeles disponibles :

- 𝗗𝗲𝗲𝗽𝗦𝗲𝗲𝗸 𝗩𝟰.𝟭 𝗙𝗹𝗮𝘀𝗵
- 𝗚𝗟𝗠 𝟱.𝟯
- 𝗚𝗟𝗠 𝟱.𝟯 𝗙𝗹𝗮𝘀𝗵
- 𝗞𝗶𝗺𝗶 𝗞𝟯

Pour les utiliser :

1. Va sur https://t.co/0KEwIF8BcO
2. Cree ton compte developpeur gratuit
3. Genere une cle API
4. Utilise cette adresse comme base URL :
https://t.co/JNwFSuDLbD
5. Choisis ton modele et c est parti

Le gros avantage c est que l API est compatible avec le format OpenAI. Donc pas besoin de reconstruire tes outils pour chaque modele.

Pour des agents perso, des tests, des contre-lectures, des syntheses ou des taches repetitives c est franchement un super plan.

Par contre il faut connaitre la limite : NVIDIA presente ca comme des endpoints gratuits pour le prototypage, la recherche et le developpement. Pas pour faire tourner un service client en production.

Le debit est limite, aucun SLA et un modele peut etre retire. Mais aucune date de fin generale n est annoncee : l acces reste lie au programme developpeur gratuit tant que NVIDIA maintient l offre.

Le bon usage c est pas de jeter tes modeles payants.

C est de faire tourner le travail courant gratuitement et de garder les modeles chers pour les taches ou ils font vraiment la difference.

[THISMA @thismacapital]

La vidéo a été réalisée par Opus 5.5 et c'est bluffant : 15mn pour la réalisation avec 2-3 modifications

Le prompt : "make a dynamic 15-second motion graphics video that shows what an incredible motion designer you are, like it's your showreel for a résumé. go all out."

Et ce qui est encore plus fou : il a tout fait en HTML JS CSS!
Aucun outil utilisé à part du natif web, qu'il a ensuite joué image par image pour générer la vidéo

Incroyable

[santi @santtiagom_]

si usás Opus 5.5 en Claude Code, podés sumar a Fable como advisor con:

/advisor fable

Opus sigue escribiendo el código. Fable lee el contexto de la sesión y aporta una segunda mirada: revisar un plan, destrabar un error recurrente o detectar algo que faltó antes de terminar.

es un patrón de executor + advisor:
un modelo lleva la tarea y consulta a otro en momentos clave, sin cederle la ejecución.

el post incluye un prompt para configurar este flujo 👇

[Vox @Voxyz_ai]

Claude Code tip: before you ship a vibe-coded site, have Opus 5.5 send out 𝟱 𝘀𝘂𝗯𝗮𝗴𝗲𝗻𝘁𝘀 to check 𝘁𝗵𝗲𝘀𝗲 𝟮𝟬 𝘁𝗵𝗶𝗻𝗴𝘀, then fix them all in one pass:

Consistent design
→ Write a DESIGN.md first, and make every color, font size, spacing and corner radius follow it
→ Keep the palette restrained and the type hierarchy clear
→ Make every button, card and input of the same kind look alike
→ Too flat? Add emphasis. Too crowded? Cut
→ Keep text readable on both dark and light backgrounds

On mobile
→ No sideways scrolling, and nothing spilling off the screen
→ Add a mobile menu
→ Make buttons big enough to tap
→ Keep the layout intact when the text is enlarged

Every state
→ Give loading, empty and error states a proper screen
→ Give every button feedback, with its own hover, pressed and disabled look
→ Show form errors right where they happen, plus submitting, success and failure messages
→ Add transitions to modals, dropdowns and tab switches

Use it like a real user
→ Click through the core flows, like sign-up and checkout, from start to finish
→ Find buttons that do nothing and links that are broken
→ Make sure the whole thing works with just a keyboard

Before launch
→ Say what you do in one line on the home page
→ Give each page one main button you want people to click
→ Give every page a title, a description and a favicon
→ Delete all placeholder text

For effort, follow Anthropic's cost guide from yesterday: medium is for well-scoped daily work. The default medium is enough for these 20 fixes; raise it for the harder ones. Thariq tested it: turning effort up mostly buys more verification and edge-case testing.

Send this whole post to Claude Code. If you need to, add this line at the end 👇
"Send one subagent per group to check it without editing anything. Then fix everything yourself from their lists, so no two agents edit the same file. If a group is more complex, raise that subagent's effort. Show me the list of issues first, don't change anything until I confirm, and when you're done, give me before-and-after screenshots."