Image created with gemini-2.5-flash-image with claude-sonnet-4-5. Image prompt: Cinematic 80s film still of children in Halloween costumes walking independently down a suburban sidewalk at golden hour, each heading toward different decorated houses, autumn leaves scattered on ground, warm porch lights glowing, shot on 35mm film with slight grain and rich fall colors
Today we’re launching Perplexity Patents, the world’s first AI patent research agent that makes IP intelligence accessible to everyone. Read more about Perplexity Patents in our latest blog: https://x.com/perplexity_ai/status/1983875975877423277
🎉 Congrats to @Kimi_Moonshot! vLLM Day-0 model expands! Now supporting Kimi Linear — hybrid linear attention with Kimi Delta Attention(KDA): – RULER 128k context: 84.3 perf + 3.98× speedup – Up to 6× faster decoding & 6.3× faster TPOT (1M tokens) – 75% KV cache reduction 💡 https://x.com/vllm_project/status/1983941708233765149
🔥 Inside Kimi Linear: First-Hand Insights @Kimi_Moonshot just dropped something impressive again. @yzhang_cs from Kimi AI Infra, shared an insider’s look at the making of Kimi Linear — an architecture designed around hybrid linear attention and optimized for efficiency × https://x.com/ZhihuFrontier/status/1984321210055082207
Kimi just released another “”next-gen”” model that reduces memory usage by up to 75%, while achieving up to 6.3× higher decoding throughput and outperforming MLA and GDN baselines https://x.com/scaling01/status/1983926811051384965
Devin now has full computer use capabilities and can share screen recordings. You can control desktop apps, build and QA mobile apps, and automate tedious work. Here are some examples that blew our team away: 1. Making a desktop game https://x.com/cognition/status/1983983151157563762
Google just took another big step towards becoming ChatGPT for planet earth. I can’t overstate how important this is — geospatial AI commodified. Here’s what it can do: https://x.com/bilawalsidhu/status/1981566109863289028
Today, we’re starting the early access rollout of the Gemini for Home voice assistant in the U.S. You can either say “Hey Google” to your speaker or display to request specific help or answers, or talk naturally with Gemini Live by saying “Hey Google, let’s chat.” https://x.com/Google/status/1983246777215033718
Your AI browser is here, whether you use a Mac or PC. Live now: https://x.com/mustafasuleyman/status/1981421909847031887
Turn on agent mode and ChatGPT can take action for you—research, plan, and get things done while you browse. Now in preview for Plus, Pro, and Business users. https://x.com/OpenAI/status/1984304194837528864
try agent mode in ChatGPT Atlas:”” / X https://x.com/gdb/status/1984304783881355451
Sam Altman says OpenAI will have a ‘legitimate AI researcher’ by 2028 | TechCrunch https://techcrunch.com/2025/10/28/sam-altman-says-openai-will-have-a-legitimate-ai-researcher-by-2028/
Agentic Commerce Protocol and building the Economic Infrastructure for AI — with Emily Glassberg Sands, Head of Data & AI at Stripe https://www.latent.space/p/stripe
🎉 Congrats to the @MiniMax__AI team for releasing MiniMax-M2 model! Built for advanced coding and agentic tasks, MiniMax-M2 is now available with Day-0 support on vLLM, bringing fast, efficient inference and smooth long-context performance. vLLM is proud to power the next”” / X https://x.com/vllm_project/status/1982675383091916856
💻Deep Agents CLI Using the deepagents package, we built a simple coding CLI as an example of a coding application you could build on top of deepagents We added in a concept of memory so that it would remember instructions and guidance over time Blog: https://x.com/hwchase17/status/1984303925101735950
📣 I’m giving a talk tomorrow at @_odsc west on context engineering w/ LangChain + LangGraph! 📄 Perhaps the biggest news with the LC + LG V1 releases is the rollout of our new, unified docs site. Check out our guide on context engineering w/ agents as a preview!”” / X https://x.com/sydneyrunkle/status/1982909408901509602
13/ @RajaPatnaik built an AI research agent that writes comprehensive reports with proper citations and optimizes its own prompts automatically using @LangChainAI + @ExaAILabs + @DSPyOSS + GEPA. https://x.com/AtomSilverman/status/1981855886885937345
15/ @omarsar0 discussed building OS agents for long-horizon tasks. It uses multiple cooperating agents for memory, planning, and error correction. 🤖 https://x.com/AtomSilverman/status/1981855893601001682
16/ @_avichawla explained Memento for fine-tuning agents without model updates. It uses memory-based reinforcement learning with case-based reasoning. https://x.com/AtomSilverman/status/1981855896708821234
20/ @swyx explained subagents killing long context. It uses fast parallel tool calling for efficient code search in agent trajectories. https://x.com/AtomSilverman/status/1981855910336328184
23/ @pontusab shares how he structure their multi-agent AI system: •10 agents (triage + 9 specialists) •43 tools (grouped by domain) •12 artifacts (visual canvases) https://x.com/AtomSilverman/status/1981855919572136352
4/ @agi_inc is now ranked #1 on OSWorld 🔥 agi-0 agent reaches superhuman performance in universal computer-use across Linux, macOS, and Windows. @DivGarg_ https://x.com/AtomSilverman/status/1981855856494088555
5/ @ycombinator launched Patent Watch. It generates claim charts in 20 minutes to identify infringing products for licensing or litigation. ⚖️ https://x.com/AtomSilverman/status/1981855860117921986
6/ @ycombinator launched @PatentWatchai. It generates claim charts in 20 minutes to identify infringing products for licensing or litigation. ⚖️ Congrats on the launch, @astroe777 & @stroe_andy https://x.com/AtomSilverman/status/1981855863360082311
7/ @pontusab released AI SDK Tools v0.9.0. It includes agents, memory, state management, devtools, and visual components for AI apps. 🛠️ https://x.com/AtomSilverman/status/1981855866447188225
9/ @diabrowser added Skill Builder. It creates expert prompts with tools and context from simple sentences for complex tasks. https://x.com/AtomSilverman/status/1981855873481023972
A world where AI augments individual workers in idiosyncratic, but often very large, ways is a world where workers can most capture the benefit of AI Firms can’t calibrate productivity expectations, so workers can choose what to do with extra productivity: leisure or more work.”” / X https://x.com/emollick/status/1981331406039945567
After playing with @LangChainAI’s new Agent Builder in LangSmith, I knew it was a homerun Now learning about this new LangChain 🦜Deep Agents CLI, it’s starting to become clear that Deep Agents are going to be powering a LOT of new apps. Im very excited to try this out! https://x.com/GitMaxd/status/1984306847856410953
Agent Labs Are Eating the Software World – Log – nibzard https://www.nibzard.com/agent-labs
Also out: R-HORIZON. It composes interdependent chains across math, code, and agent tasks to test real long-horizon reasoning. Top models degrade rapidly as horizon grows: DeepSeek-R1 falls from 87.3% to 24.6% at 5 linked problems, R1-Qwen-7B drops from 93.6% to 0% at 16. https://x.com/gm8xx8/status/1982608933563826270
BREAKING: Cursor 2.0 is out now! It’s a reenvisioned Cursor focused on agentic programming. We’ve been testing it for a week or so @every and here’s our Day 0 Vibe check. What’s new: – New agent view prioritizes what programmers actually spend time on (delegating to and https://x.com/danshipper/status/1983566084163666250
Cline v3.35 is live <long read below> https://x.com/cline/status/1984306206538940702
Code like a surgeon https://www.geoffreylitt.com/2025/10/24/code-like-a-surgeon
Cognition | Introducing SWE-1.5: Our Fast Agent Model https://cognition.ai/blog/swe-1-5
Composer is a frontier coding model that completes tasks in under 30 seconds. https://x.com/cursor_ai/status/1983567621602881992
Cursor debuts Composer and multi-agent suite in Cursor 2.0 https://www.testingcatalog.com/cursor-debuts-composer-and-multi-agent-suite-in-cursor-2-0/
Exciting news from our partner, @confluentinc! Confluent has introduced new Streaming Agents capabilities and a Real-Time Context Engine, bringing fresh, live context to intelligent AI agents and enterprise applications. Together, Qdrant and @confluentinc are empowering https://x.com/qdrant_engine/status/1983843826436395090
Graph-based Agent Planning It lets AI agents run multiple tools in parallel to accelerate task completion. Uses graphs to map tool dependencies + RL to learn the best execution order. RL also helps with scheduling strategies and planning. Major speedup for complex tasks. https://x.com/omarsar0/status/1983892163990843692
How Cursor uses Cursor agents internally: https://x.com/benln/status/1983960258809831530
I joined @cursor_ai a few months ago to help build a fast model for agentic coding. Very excited the first version has shipped and can’t wait to hear what you all think! https://x.com/samkottler/status/1983569571631210629
In @code, Plan is now built-in. Plan analyzes tasks, breaks them down into steps, and generates implementation plans before starting development. Available in VS Code Insiders, coming to Stable soon! https://x.com/code/status/1983942033879257195
Inside LangSmith’s No Code Agent Builder We built a no code agent builder, and our team is breaking down why we did it + what’s powering it under the hood. In our latest roundtable discussion, Harrison and the engineering team behind LangSmith Agent Builder sit down to discuss: https://x.com/LangChainAI/status/1983916519513059728
Introducing Cursor 2.0. Our first coding model and the best way to code with agents. https://x.com/cursor_ai/status/1983567619946147967
Kick off, monitor, and review all of your agents in @code – from Copilot, or agents like Codex. https://x.com/pierceboggan/status/1983217733706625122
LangSmith Agent Builder is our no code agent builder for anyone to create an agent, now available in private preview. This is not a workflow builder. Built on our Deep Agents architecture, LangSmith Agent Builder handles planning, memory, and sub-agents automatically. This means https://x.com/LangChainAI/status/1983568636079112233
Manage a fleet of cloud agents, directly from Cursor. Now with faster startup, improved reliability, and a new UI. https://x.com/cursor_ai/status/1983954528933421419
MiniMax Agent: Minimize Effort, Maximize Intelligence https://agent.minimax.io/
MiniMax M2 & Agent: Ingenious in Simplicity – MiniMax News https://www.minimax.io/news/minimax-m2
New Coding Model and Agent Interface · Cursor https://cursor.com/changelog/2-0
New dataset/standard for SFT of agentic language models! – A new standard for agentic training data: Agent Data Protocol – 13 standardized datasets, 1.27M instances – Training results on 3 agents, 7-32B models, often w/ SOTA accuracy for the model size”” / X https://x.com/gneubig/status/1983548125135655228
ollama run minimax-m2:cloud Free until November 7th in partnership with @MiniMax__AI ! MiniMax M2 is now available on Ollama’s cloud! It’s currently the #1 open weight model built for coding and agentic workflows. Learn more 👇👇👇 https://x.com/ollama/status/1983285057746763810
Postman’s latest Guidebook on building AI-ready APIs is one of the most important documents you can read today as a developer! We are headed into an era where every website must be “”Agent-ready””. – Agents will make purchases, not humans. – Agents will find the best options, not https://x.com/_avichawla/status/1983060260710392194
Read about everything that’s new in 2.0. More tomorrow! https://x.com/cursor_ai/status/1983567631035883677
Real-time, context-aware agents are here ⚡️ @confluentinc and Weaviate are helping developers build intelligent, event-driven AI systems that reason and act on live data — not stale snapshots. Learn how Streaming Agents on Confluent Cloud and Weaviate’s vector database work https://x.com/weaviate_io/status/1983921589163835398
Really like how @hwchase17 breaks down the emerging layers of the agent ecosystem: Runtime → Framework → Harness ⚙️ LangGraph = runtime 🧠 LangChain = framework 🪶 DeepAgents = harness Having spent the last months on the LangChain v1 abstractions, I find “agent harness” the”” / X https://x.com/bromann/status/1982789085979685349
Reload Windsurf or update to the latest version to try it out: https://x.com/cognition/status/1983662840574796049
so excited to share Composer with the world! Composer is Cursor’s own agentic coding model. it plans, edits, and builds software alongside you with precision, keeping you in flow with incredible speed. i started this project on the side while working on a bug-finder prototype”” / X https://x.com/ellev3n11/status/1983571309100782029
Some personal news: I recently joined Cursor. Cursor is a small, ambitious team, and they’ve created my favorite AI systems. We’re now building frontier RL models at scale in real-world coding environments. Excited for how good coding is going to be. https://x.com/srush_nlp/status/1902736199636205914
Super proud of the @MicrosoftAI team for their contributions to MSFT earnings today: – @MicrosoftEdge has taken share for 18 consecutive quarters – We took share again with @Bing – @Copilot daily users are up almost 50% QoQ – Search and news ad revenue ex-TAC up 16% (15% in cc)”” / X https://x.com/mustafasuleyman/status/1983681503918723540
The Agent Sessions view is your unified interface to manage both local and cloud agent sessions. Try it now in VS Code Insiders! https://x.com/code/status/1984322058503807066
This is actually a clever context engineering technique for web agents. It’s called AgentFold, an agent that acts as a self-aware knowledge manager. It treats context as a dynamic cognitive workspace by folding information at different scales: – Light folding: Compressing https://x.com/omarsar0/status/1983646041850495140
Today we’re releasing 0.2 of deep agents 📁The main addition here is a “”backend”” abstraction, which allows you to swap out the filesystem that deep agents use. Can be local filesystem, a database, remote vm, anything Agent Harnesses like deep agents are becoming more important https://x.com/hwchase17/status/1983218572202803285
Turn speech into code with voice mode. https://x.com/cursor_ai/status/1983567629303357773
We designed SWE-1.5 as an integrated package: the model, inference, and agent harness are co-designed as a unified system optimized for both speed and intelligence. https://x.com/cognition/status/1983662838955831372
We just built and released the largest dataset for supervised fine-tuning of agentic LMs, 1.27M trajectories (~36B tokens)! Up until now, large-scale SFT for agents is rare – not for lack of data, but because of fragmentation across heterogeneous formats, tools, and interfaces.”” / X https://x.com/yueqi_song/status/1983539504385253684
We’ve migrated our system prompt tool calling format to native tool calling and split that out for different model families. Here’s why this results in a better experience using Cline: Models now return tool calls in their native JSON format, which they were specifically https://x.com/cline/status/1984334385626411397
🚀 LangChain is now an AWS Generative AI Competency Partner We’re joining a small group of companies recognized by AWS for their technical expertise and customer impact in GenAI. LangSmith (our platform for agent engineering) is now available through the AWS marketplace. https://x.com/LangChainAI/status/1984303566723625044
Today we’re releasing SWE-1.5, our fast agent model. It achieves near-SOTA coding performance while setting a new standard for speed. Now available in @windsurf. https://x.com/cognition/status/1983662836896448756
Today, @cognition released SWE-1.5 – the world’s fastest coding agent, powered by Cerebras. SWE-1.5 achieves frontier-level coding ability, comparable to Sonnet 4.5 and surpassing GPT-5. Cerebras and Cognition engineers worked hand in hand over the past few weeks, training a https://x.com/cerebras/status/1983695672454074794
1/ @LangChainAI raised $125M to build the platform for agent engineering. It enables seamless agent development with insights and multi-turn evaluations for production use. 🚀 Great work @hwchase17 https://x.com/AtomSilverman/status/1981855844229955719
We just acquired @weavy_ai, soon to be known as Figma Weave. With Weavy, you’ll get more models, more tools, more ways to create within the canvas. More to come https://x.com/figma/status/1983889394944692359
A continuing issue in studying LLMs for medicine is the fact that everyone is testing different things with different standards. This (interesting) paper is about agentic systems powered by DeepSeek-V3.2. Other papers look at single LLMs. Tons of different benchmarks. Confusion.”” / X https://x.com/emollick/status/1982630126065201636
10/ @nutlope built an agent generating interactive courses from PDFs. It uses pre-built components invoked through tool calls for structured lessons. 📚 https://x.com/AtomSilverman/status/1981855876639313972
Learn all about the new LangChain agent in these courses! Both in Python and Typescript! Only ~1hr so a fun little learning opportunity”” / X https://x.com/hwchase17/status/1982919412954067000
🔥 New LangChain Academy Course: LangChain Essentials (Python & TypeScript) 🔥 Learn the basics of LangChain – our open source framework that makes it easy to start building agents with any model provider. Last week, we released LangChain 1.0. We’ve completely rewritten https://x.com/LangChainAI/status/1982851795287507398
Smoking gun: Pretty sure Cursor’s new Composer-1 is a fine-tuned Chinese model. As I was building, it switched its inner monologue to Chinese, and I can’t get it back to english. @simonw https://x.com/auchenberg/status/1983901551048470974
Marvel gave Ian Dawson and his team six months to design Tony Stark’s holographic AI assistant. “”We weren’t trying to predict the future. We were just solving script problems.”” Then Microsoft was building it. Tesla was building it. Apple was building it. Here’s my interview”” / X https://x.com/bilawalsidhu/status/1981853282328035627
The Building Blocks of Agentic AI: From Kernels to Clusters https://ai.meta.com/blog/introducing-pytorch-native-agentic-stack/
12 new @Copilot features in 27 seconds – built to make a difference in your real world instead of pulling you away from it https://x.com/mustafasuleyman/status/1981828293503725784
2/ @MicrosoftEdge launched Copilot Mode in Edge. It transforms the browser into an intelligent companion with AI innovations for Windows and Mac users. https://x.com/AtomSilverman/status/1981855847832846810
All of today’s @Copilot announcements boil down to one core idea: we’re betting on humanist AI. An AI that always puts humans first. – Copilot Groups – AI browser – our new character Mico – memory updates – Copilot for health + more in this morning’s event https://x.com/mustafasuleyman/status/1981390345578697199
Introducing Agent HQ: Any agent, any way you work – The GitHub Blog https://github.blog/news-insights/company-news/welcome-home-agents/
Public preview of Copilot Metrics dashboard! Track the impact of using Copilot and _any_ coding agent across your organization. HUGE. This is so highly requested in @Code as well. https://x.com/burkeholland/status/1983214298705850768
T-10 minutes https://x.com/mustafasuleyman/status/1981387721861189817
Proud to be joining @github’s new Agent HQ as launch partners. More choices and more ways to try Devin, coming later this year!”” / X https://x.com/cognition/status/1983251389473271887
Seeing everyone’s reactions to Copilot Groups this week leaves me even more convinced the future of AI is social not just solo”” / X https://x.com/mustafasuleyman/status/1981775472091705535
12/ @milan_milanovic announced Microsoft Agent Framework. It combines Semantic Kernel and AutoGen for scalable multi-agent systems with observability. https://x.com/AtomSilverman/status/1981855883366985829
✨ At vLLM, we strive for correctness, reliability, and open collaboration — every detail matters. Together with @Kimi_Moonshot , we verified Kimi K2’s tool-calling accuracy on vLLM using the latest K2-Vendor-Verifier benchmark. Our debugging uncovered 3 key compatibility”” / X https://x.com/vllm_project/status/1983115488982122929
Kimi For Coding: Exclusive Add-on to Your VIP Plan! We’ve added Kimi For Coding as a powerful add-on built right on top of your current subscription perks. Extra value, no extra cost. More details 👉 https://x.com/Kimi_Moonshot/status/1984207737673359441
Kimi K2vv updated! We’ve added case-by-case statistics for ToolCall-Trigger Similarity and ToolCall-Schema Accuracy. Feedback is welcome! https://x.com/Kimi_Moonshot/status/1983082003731042637
Kimi K2vv updated! We’ve added case-by-case statistics for ToolCall-Trigger Similarity and ToolCall-Schema Accuracy. The infra team also listed some suggestions for vendors; looks like enforcer is important. https://x.com/crystalsssup/status/1983126339399102756
Kimi Linear Tech Report is dropped! 🚀 https://x.com/Kimi_Moonshot/status/1983937694360322136
My favorite part: > “Scaling Ladder” is a Kimi tradition for scaling models. We start from something small (say, 1B active parameters) and gradually aim to beat the baseline on benchmarks, while also monitoring the corresponding “internals.” Only after clearing each gate at each”” / X https://x.com/eliebakouch/status/1984291165860958614
Thankfully, theres a really nice glossary in the KIMI Delta Attention paper that covers most of the notable variants https://x.com/nrehiew_/status/1983891931823505518
There are a lot of works behind Kimi Linear. We’ve rethought efficient and expressive linear attention from infra. We even first discovered the attn matrix, and then the recurrent. No wait to check out the kda kernel in the FLA repo. We have much more work to do, to open.”” / X https://x.com/uniartisan/status/1983941443283775780
You are also welcome to share your suggestions and feedback for our Kimi CLI on GitHub. > https://x.com/Kimi_Moonshot/status/1984207741037252751
Introducing Kimi CLI Technical Preview & Kimi For Coding! Kimi CLI powers your terminal: – Shell-like UI + shell command execution – Seamless Zsh integration – MCP support -Agent Client Protocol (now compatible with @zeddotdev) More features incoming! https://x.com/Kimi_Moonshot/status/1984207733177090274
Claude, GPT-5, Gemini, and Kimi: “”write me a horror story done entirely in the dedications to six books (you can give me the title and author of each book as well)”” ChatGPT and Claude did well in different way. Kimi did the usual (sounds good but meaning falls apart). https://x.com/emollick/status/1982279778859151783
Many people are confused by Minimax’s recent return to full attention – especially since it was the first large-scale pivot toward hybrid linear attention – and by Kimi’s later adoption of hybrid linear variants (as well as earlier attempts by Qwen3-Next, or Qwen3.5). I actually”” / X https://x.com/SonglinYang4/status/1984021551914926514
Codex for empowering everyone to ship:”” / X https://x.com/gdb/status/1981605886956384591
We have big news to share: marimo is joining @CoreWeave! We’re doubling down on open-source and scaling molab with serious compute Our mission is the same: to build the world’s best open-source notebook for working with data Read the full announcement: https://x.com/marimo_io/status/1983916371869364622
🤖deepagents: the open source, multi-model agent harness We’re releasing 0.2 of deep agents, with a big addition: a “”backend”” abstraction This lets you swap the filesystem you use from a local filesystem to a remote VM to a database to anything blog: https://x.com/LangChainAI/status/1983219130057527662
standard content blocks is a huge new addition in LangChain v1! solves a bunch of issues with switching between model providers”” / X https://x.com/hwchase17/status/1982652804654391432
18/ @rohanpaul_ai detailed DeepAnalyze agent for data science workflows. It plans, codes, and iterates using hybrid rewards for analyst-grade results. https://x.com/AtomSilverman/status/1981855903826706795
14/ @svpino shared 8 rules improving AI coding agents. It automates checks for security and quality using Codacy integration. https://x.com/AtomSilverman/status/1981855890358858159
Introducing Aardvark, our agentic security researcher:”” / X https://x.com/gdb/status/1983971650531160319
Proud to introduce Aardvark, our agentic security researcher powered by GPT-5. Aardvark hunts for vulnerabilities the way a security engineer would: by reading and analyzing code, writing and running tests, and proposing patches. Now in private beta. https://x.com/embeddedsec/status/1983956550239842474
ImpossibleBench: Measuring Reward Hacking in LLM Coding Agents — LessWrong https://www.lesswrong.com/posts/qJYMbrabcQqCZ7iqm/impossiblebench-measuring-reward-hacking-in-llm-coding-1
These are the new component datasets that we have in our new 1.27M trajectory agent training database. One question: did we miss any big/important agent training datasets? I’d love to add more to our repo so if people know any ones that are excellent I’d like to add them. https://x.com/gneubig/status/1983563909505187975
1/ x402-mcp adds Solana support for on-chain payments. @_0xaryan demonstrates ordering books with this protocol enhancement. https://x.com/AtomSilverman/status/1983653186482401434
10/ @replit streamlines MCP server deployment in seconds. @mattppal guides remixing templates for quick publishing. https://x.com/AtomSilverman/status/1983653221131547100
14/ In a detailed explanation, @_avichawla covers MCP and A2A protocols for agent collaboration. https://x.com/AtomSilverman/status/1983653236390424961
16/ @KirkDBorne promotes a new book on building MCP servers in Python. It covers components from testing to deployment. https://x.com/AtomSilverman/status/1983653243239723074
17/ Statsig transforms docs into an MCP server for syntax guidance. @statsig aids agents in feature experiments. https://x.com/AtomSilverman/status/1983653246666469880
18/ Effect simplifies asynchronous MCP result handling with timeouts. @RhysSullivan shares code for efficient domain fetching. https://x.com/AtomSilverman/status/1983653250273591615
19/ @alexdupler optimizes DAX measures using MCP in Power BI. https://x.com/AtomSilverman/status/1983653253624762450
19/ @michaelfreedman launched Agentic Postgres. It features forkable storage and MCP server for agent-optimized database operations. https://x.com/AtomSilverman/status/1981855907379306866
2/ @gumloop launches agents for Slack workflows using MCP. https://x.com/AtomSilverman/status/1983653191234597010
20/ @Cloudflare hosts MCP hack nights on server building. @mies features sessions with mcp-lite and Workers. https://x.com/AtomSilverman/status/1983653257248723147
21/ An AI Content Audit Assistant analyzes sites with MCP tools. @chris_nectiv combines Screaming Frog and Zapier. https://x.com/AtomSilverman/status/1983653260406968765
23/ In a client setup, @pinaldave configures MCP Server for natural language SQL queries. https://x.com/AtomSilverman/status/1983653263791812902
24/ Have you tried the AgentOps MCP for debugging your agent? Test it out now for free. Link in my bio. Follow @AtomSilverman and @AgentOpsAI for everything AI agent-related Last week’s thread: https://x.com/AtomSilverman/status/1983653267252085099
3/ MongoDB’s MCP server enables natural language queries for databases. @svpino highlights its integration with AI assistants for context access. https://x.com/AtomSilverman/status/1983653194845819070
5/ @dylibso introduces Turbo MCP for enterprise security. This self-hosted gateway enables regulated industries to connect apps to AI safely. https://x.com/AtomSilverman/status/1983653202693386478
7/ x402 integrates with @Hive_Intel MCP for on-chain AI actions 🔗 https://x.com/AtomSilverman/status/1983653209660092468
8/ Upgrading Next.js apps becomes effortless with MCP prompts 🔄 @ronphaestas https://x.com/AtomSilverman/status/1983653213879640160
9/ In a market analysis demo, @pikdotfun integrates MCP with x402 for on-chain tools. https://x.com/AtomSilverman/status/1983653217578950884
Introducing LangSmith Agent Builder 🤖🧱 A true agent building experience (not workflows!!), all through a simple natural language interface. Describe your agent to “”build”” it, then interact with it via chat, or add an automatic trigger. Connect your agents to any MCP server, https://x.com/BraceSproul/status/1983581751550341408
LlamaIndex now ships native MCP search with our documentation! – your coding agents can directly access search tools across all our docs! 🔍 You can plug this URL into any client/framework/tool that supports MCP search: https://x.com/llama_index/status/1984292554968616994
(28) The Secrets of Claude Code From the Engineers Who Built It – YouTube https://www.youtube.com/watch?v=IDSAMqip6ms
🤖LangSmith Agent Builder Today, we’re releasing our first no-code agent builder experience to let anyone build agents It’s essentially “”general purpose claude code in the UI””. Comes with built in memory support, allowing it to learn and adapt with you over time It is NOT a https://x.com/hwchase17/status/1983584242241294423
17/ @dani_avila7 demonstrated creating Claude Code skills. It involves running a command and describing desired capabilities for reusable tools. https://x.com/AtomSilverman/status/1981855900165148685
8/ Using Skills with the Claude Agent SDK. Here’s an example demo of @trq212 built using one of our premade skills to make an excel demo agent. https://x.com/AtomSilverman/status/1981855870184271989
Claude Skills, anywhere: making them first-class in Codex CLI https://www.robert-glaser.de/claude-skills-in-codex-cli/
We just added thinking block preservation in the Claude API. You can now control how thinking blocks are managed in your context window, resulting in more cache hits and lower costs.”” / X https://x.com/alexalbert__/status/1983597775293177952
We’re open-sourcing MiniMax M2 — Agent & Code Native, at 8% Claude Sonnet price, ~2x faster ⚡ Global FREE for a limited time via MiniMax Agent & API – Advanced Coding Capability: Engineered for end-to-end developer workflows. Strong capability on a wide-range of applications https://x.com/MiniMax__AI/status/1982674798649160175
we’ve been challenging ourselves “”does the world need one more code-cli?”” “”how can we catch up with claude-code?”” maybe the answer is NO, but at least, we have one place to share our understanding of coding agents, and good changes will gradually happen. it’s just a beginning.”” / X https://x.com/bigeagle_xd/status/1984217403023380802
Available in public beta today on the Claude API and on Google Cloud’s Vertex AI, with Amazon Bedrock coming soon. Docs here: https://x.com/alexalbert__/status/1983597787305697745
22/ @alexalbert__ introduced Skills in Claude. It packages knowledge for on-demand loading to handle complex agent tasks efficiently. https://x.com/AtomSilverman/status/1981855917131121028
21/ @alexalbert__ hosted convo with @ErikSchluntz on agents. It covers Claude’s strengths, skill tips, subagents, and future developments. https://x.com/AtomSilverman/status/1981855913977024524
Had a great time at @GitHub Universe announcing Agent HQ. @Claudeai will soon become a native collaborator in GitHub, able to pick up issues, create branches, commit code, and work alongside you—all powered by the Claude Agent SDK and deeply integrated with GitHub’s platform. https://x.com/mikeyk/status/1983213185332326434
MiniMax M2 + Claude Code on KingBench Agentic Evaluations: It now scores #2 on my Agentic Evaluations beating GLM-4.6 by a wide margin. It seems to work much better with Claude Code’s Tools. Really great model and it’s my daily driver now. I haven’t tested GLM with CC yet. https://x.com/aicodeking/status/1983934597353402797
13/ The Flutter Extension connects Gemini CLI to Dart MCP servers 📱 @FlutterDev https://x.com/AtomSilverman/status/1983653233177588214
6/ @EmergentLabsHQ integrates Shopify via MCP servers. @mukundjha automates workflows across platforms. https://x.com/AtomSilverman/status/1983653206011081003
11/ Chrome DevTools MCP empowers agents to debug performance issues. @ChromiumDev measures interactions for UI response fixes. https://x.com/AtomSilverman/status/1983653224927416642
4/ @Docker partners with @e2b for secure AI agent sandboxes. It provides access to tools like GitHub via MCP. https://x.com/AtomSilverman/status/1983653198272618767
15/ MCP Toolbox for Databases supports multiple SQL systems. @Sumanth_077 offers open-source tools for secure agent interactions. https://x.com/AtomSilverman/status/1983653239821365711
🚀We are excited to introduce the Tool Decathlon (Toolathlon), a benchmark for language agents on diverse, complex, and realistic tool use. ⭐️32 applications and 600+ tools based on real-world software environments ⭐️Execution-based, reliable evaluation ⭐️Realistic, covering https://x.com/junxian_he/status/1983834164727312391
An exciting new course: Fine-tuning and Reinforcement Learning for LLMs: Intro to Post-training, taught by @realSharonZhou, VP of AI at @AMD. Available now at https://x.com/AndrewYNg/status/1983205131576590487
🗃️ Context Caching Update! The Gemini API implicit caching now with 90% cost savings when your requests hit the cache! This means if you send a request to Gemini models with a common prefix as one of previous requests, it might be cached. No code changes needed. Price details https://x.com/_philschmid/status/1983565009574678679
3/ @GoogleAIStudio introduced brainstorming with Gemini in vibe code. It generates context-aware suggestions for features and API integrations during app building. 💡 https://x.com/AtomSilverman/status/1981855852844966045
5 practical tips for Context Engineering, which apply to Google DeepMind Gemini as well: 1⃣ Context Ordering Matters: Try to use “”append-only”” context, adding new information to the end. This maximizes cache hits reducing cost (4x) and latency. 2⃣ Manage Tools Statically: Avoid https://x.com/_philschmid/status/1982861526466707477
Introducing Annotate mode ✏️ while you vibe code in @GoogleAIStudio, mark up any UI with simple drawing tools and then have Gemini action them directly in the code! https://x.com/OfficialLoganK/status/1981375555783045198
Introducing vibe coding in Google AI Studio https://blog.google/technology/developers/introducing-vibe-coding-in-google-ai-studio/
We’re introducing logs and datasets in @GoogleAIStudio ! 💫 Designed to help you debug applications and more easily create datasets for evaluations. – Enable per project with one click; no code changes. – Automatically track Gemini calls at no monetary cost. – Inspect full https://x.com/_philschmid/status/1984258488013340826
two Gemini API updates to help you build more efficiently: • Batch API: run large-scale jobs at a 50% discount (now with support for Nano Banana) • Context Caching: 90% discount (up from 75%) for cached input tokens on Gemini 2.5 models”” / X https://x.com/GoogleAIStudio/status/1983564552408056179
A weird gap in the Google AI line-up is between the Gemini research tools and NotebookLM. Gemini lets you do Deep Research, but only a subset of other NotebookLM features, while NotebookLM won’t let you trigger a deep research report or do other types of AI interactions.”” / X https://x.com/emollick/status/1983611024113856610
With a built-in browser, agents can now run and test their code. https://x.com/cursor_ai/status/1983567626543734799
Brain-IT: Image Reconstruction from fMRI via Brain-Interaction Transformer A new paper from Weizmann Institute of Science, getting reconstructions that are not complete nonsense from ONLY 15 min of data. We previously demonstrated SOTA for 1 hr of data with MindEye2. This https://x.com/iScienceLuvr/status/1984195725253804449
11/ Sierra agents can be published on your web site, integrated with your mobile app, answer the phone, and now they can also be published to ChatGPT so you can directly reach hundreds of millions of consumers with your agent. thanks for sharing @btaylor https://x.com/AtomSilverman/status/1981855879977972004
You’ve asked for more flexible ways to get more Codex usage: Introducing credits for Codex on ChatGPT Plus and Pro. Credits give you more usage beyond what’s included in your plan, kicking in when you hit limits. As a bonus, we also reset Codex rate limits for everyone. Enjoy!”” / X https://x.com/OpenAIDevs/status/1983956896852988014
OpenAI Codex is now integrated directly in @code through the new Agent Sessions view – and can be powered by your GitHub Copilot subscription. Try it out now with VS Code Insiders and a Copilot Pro+ subscription. Happy coding! https://x.com/code/status/1983214972214632897
You can now use @OpenAI Codex with your Copilot Pro+ login. No additional subscription required. Use the agents that you 💗 https://x.com/code/status/1983973969335378241
Welcome to your Agent HQ 📍Orchestrate any agent, any time, anywhere. Coding agents from @claudeai, @OpenAI, @cognition, @julesagent, @xai and more will become available in GitHub as part of your paid Copilot subscription. https://x.com/github/status/1983205334756839605
@OpenAI I really wish Atlas would improve on more advanced browser operations. It can click buttons and do regular browsing, but it gets stuck quite often when it’s adding, formatting, or creating things.”” / X https://x.com/omarsar0/status/1984304979671224702
Now in private beta: Aardvark, an agent that finds and fixes security bugs using GPT-5. https://x.com/OpenAI/status/1983956431360659467
Giving reasoners access to data connections that they can use as needed to look up information is such a huge leap in practice over traditional RAG. Letting the AI do searches, refine its searches in response to results, and learn what does and does not exist gives richer context”” / X https://x.com/emollick/status/1982176215537451253
AgentFold Long-Horizon Web Agents with Proactive Context Management https://x.com/_akhaliq/status/1983547985238577477
