Anthropic Launched their Agents SDK!!🔥

Give your @crewAIInc agents a reference library With Docling Knowledge Sources, agents can: - Ingest web pages or docs - Answer with citations - Work with richer context Example: use Docling to load Lilian Weng’s papers into CrewAI 👇

Tried out @CodeRabbit CLI today 🐇⚡ And honestly… I’m impressed. > The file tree view makes the DX great! > It spotted potential issues in my code. > Even explained + suggested AI agent prompts to fix them Overall, big fan of this workflow 👀

Loving this little detail in the @coderabbitai CLI, the file tree view. 🌲 Such a simple thing, but it makes navigating + reviewing way easier

Awesome AI Apps has Crossed 5k Stars on GitHub! Thanks, everyone, for your support! I'll be adding some more projects soon!

Just tried the new @github Copilot CLI 👀 And wow… It’s quite good. I asked it to fetch issues, explain code, and even suggest project ideas. The coolest part? It actually scanned my repo and gave me solid ideas based on them! This looks Promising!

GitHub just dropped Copilot CLI 👀 With it, you can use Copilot directly in your terminal: - Access repos, issues, and PRs with natural language - Build, edit, debug, and refactor code with AI - MCP-powered, fully controllable Install via npm & code smarter ⚡

Context Engineering >> Prompt Engineering! Totally agree with @jeffreyhuber

It takes just one line to crawl entire websites with AI ✨ @scrapegraphai SmartCrawler can: - Traverse multiple pages & follow links - Extract structured data with LLMs - Convert HTML → clean markdown (80% cheaper) Here’s how 👇

Awesome AI Apps now features top agent frameworks: → Google ADK → OpenAI Agents SDK → @LangChainAI → @llama_index → @AgnoAgi → @crewAIInc → AWS Strands SDK → @pydantic AI Been fun curating this. What should we add next? Drop your favs 👇

Qwen just launched @Alibaba_Qwen 235B A22B VLM 👀 Had to give it a spin on @HuggingFace, first test was a flag 🇪🇺 (it guessed 12 stars instead of 13 😅). Next try? I gave it an image of a Zebra, and it had the correct answer! Solid start for such a massive model!

Simple multi-agent Human in the Loop system using Google ADK. Simple to orchestrate but quite powerful!

The agentic loop is what makes Strands agents so smart. Instead of hardcoding every step, Strands lets the model: 1. Perceive the situation 2. Think about options 3. Act with the right tool It’s a continuous cycle, so your agent can handle complex tasks on its own

Databases have powered the internet for 50 years. Now they might power AI memory too.

Introducing Hotel Finder Agent! 🏨 We built an AI Agent that: - Supports Quick + Advanced Hotel Search - Takes location, price, amenities, and guest info - Finds deals, luxury stays, & more Built with @agnoagi, @nebiusaistudio, & @streamlit 100% Free & Open Source

Learn How to Build Docs QnA Agent in 5mins! Full Video: youtu.be/lZRKKd7MC5A x.com/Arindam_172...

Been Exploring @awscloud Strands Agents lately 👀 It’s actually pretty fun, spins up agents fast, and the Agent Loop is slick. So I recorded a quick tutorial: • How Strands works • Launching your first agent • Agent Loop in action Check it out 👇 www.youtube.com/watch?v=9ry...

Agents don’t need to forget. With @heygibsonai Memori's Auto Ingest Mode: - Each query runs a memory search - Finds the most relevant facts (up to 5) - Injects them into the LLM call Dynamic context retrieval in real time 👇

Finally Langchain launched their New Docs! And believe me, It's much cleaner than the previous versions! Mintlify did the magic again ✨

AI Agents aren’t just another workflow tool! They flip control logic on its head! Instead of rigid programs, agents reason, adapt, and act dynamically. That’s why they’re perfect for messy, real-world problems where fixed workflows break

So, GitHub has its Own MCP Registry!🔥 This looks quite interesting! Find it here: github. com/mcp

Build a Browser Agent that navigates websites with @nebiusaistudio & @browser_use! Just a few lines of code and your agent can: - Open amazon .com - Search for laptops - Sort by best rating - Return the top result in markdown Example 👇

Awesome AI Apps is Trending on GitHub 🌟 It now has 45+ AI Agents, including: - Starter agent templates - Complex agentic workflows - Agents with Memory - MCP-powered agents - RAG examples - Multiple Agentic frameworks Thanks, everyone, for supporting this.

It takes just 15 lines to use @nebiusaistudio with the Vercel AI SDK ✨ Build AI-powered apps in TypeScript, React, or Next.js: - Install the SDK - Use Open Source models like GPT OSS - Generate or stream text instantly Here’s how 👇

The only RAG engine you’ll ever need! ✨📄 RAGFlow is an open-source RAG engine built for deep document understanding, perfect for delivering truthful Q&A with solid citations. 100% Open Source. Ready to deploy!

The future of AI Agents is Event-Driven ⚡ Not just bigger models, we need better infra. EDA acts as the “nervous system” for agents: - Async communication - Loose coupling (no rigid APIs) - Real-time data flow & context sharing This is how we scale beyond fixed workflows.

Your @AgnoAgi agent can now maintain persistent memory across chats✨ Just add MemoriTools() and it will remember: - Conversation history - User preferences - Long-term context Here’s the agent setup 👇

The only Prompt Engineering Guide you’ll ever need! 💡 Packed with the latest papers, tools, lectures, and techniques to help you master LLMs from reasoning to tool use. 100% free and perfect for devs & researchers.

Awesome AI Apps is at 4.9k Stars! Can we get 5k by this week?

Started Working on a New Project using @heygibsonai Memori, @composiohq , @nebiusaistudio & @scrapegraphai Will be sharing more details soon!

So finally the day has arrived. Official MCP Registry is here! A universal catalog for MCP servers. → One-click discovery, → metadata via server.json, → support for NPM/PyPI/Docker, → publisher CLI, → REST API + community curation. This changes how agents connect.

I just vibe coded a @vercel-like toolbar!⚡ Instead of screenshots + Slack threads, users can now: - Drop pin comments on the page - See live cursors & presence in real time - Control everything from the bottom Built with @lovable_dev, @windsurf & @veltjs Demo👇

Antonio Gulli just dropped the Agent Design Patterns book we’ve been waiting for. - 406 pages, 21 patterns, zero fluff - Real code, tested - MCP integration (finally!) The other 16 patterns you’re not building 👀

Give your AI agents real-time web superpowers Bright Data MCP Server lets LLMs search, extract & navigate the web seamlessly, without getting blocked. Works with any LLM, IDE, or agent framework. 100% open-source.

The Agentic AI Periodic Table just dropped. 🤯 Prem Natarajan mapped all 66 elements of agent building: • AI Terminology (AGI, LLM…) • Agent Patterns (MCP, PLAN, MAS) • Memory Systems (STM, LTM, DIARY) • Safety Protocols This is Very Well Structured!

"Context Engineering > Prompt Engineering" Nobody explains it better than @karpathy. Prompting is a surface-level interaction, but context engineering is what makes industrial-grade LLM apps work. It’s a blend of: - Science: packing the right inputs like examples, memory, RAG docs, task state

This Channel is a Gem! 💎

NVIDIA recently dropped Universal Deep Research (UDR)! It's a deep research framework that works across any provider. (Yes, @nebiusaistudio too) We forked it & tested with @Kimi_Moonshot Kimi K2 worked really well! Here's the Demo:

Extract structured data from messy text ✨ Google’s LangExtract lets you use LLMs to pull key info from unstructured docs into clean, organized formats, while keeping traceability to the source. 100% open-source.

If You're struggling to choose a Model for your Usecase, Cursor has a Dedicated guide for that! It's really helpful!

Google just launched EmbeddingGemma ✨ - Top open embedding <500M on MTEB - Runs in <200MB RAM - Dynamic dims 768 → 128 - 2K context length - 100+ languages Built on Gemma 3 (270M). Perfect for on-device AI! x.com/GoogleDeepM...

Choosing the right LLM isn’t about knowing every option; it’s about asking the right questions! The answers will point you to the perfect fit.

Crazy to imagine 🤯 Awesome AI Apps just doubled its GitHub stars in only 25 days! 🌟 Super thankful for all the love & support, this community is amazing 🙌

SmolDocling uses DocTags to keep text and structure separate! It makes it way easier for image-to-sequence models to parse docs without getting confused. Unlike direct HTML/Markdown, it keeps layout details intact, uses fewer tokens, and exports cleanly to HTML, Markdown, or JSON.

Introducing Talk to Docs! 📄 We built an AI Agent that lets you chat with any documentation in plain English. - Plug in your hosted MCP server - Ask questions - Get detailed answers with source links Built with @agnoagi, @nebiusaistudio, @mintlify MCP & @streamlit Watch here:

AWS's Strands Agents SDK is a simple, straightforward way to build agents. 15 lines to define an agent with access to an MCP server. I found it fun to work with and appreciated how simple it was to define functions.

Wish me luck!

A good LLM ≠ a good agent. To run production-grade agents, you need a framework. One that handles: > Multi-step orchestration > Tool usage > Error recovery > External integrations LLMs generate, but frameworks orchestrate.

Tired of coding agents redoing the same steps? Try this instead: - Save the current state in progress.md - Continue from there with your new prompt It Cuts down so much back & forth. Simple but powerful ⚡

Microsoft just dropped VibeVoice-Large!🔥 I tested it on @huggingface for podcast generation, and wow, it’s impressive! Feels like we now have an open-source NotebookLM. 100% Open Source (MIT License)