Agents Hour

Shane and Abhi are in person at the CodeRabbit studio, and AISI just quietly torched one of Anthropic's loudest narratives. AISI confirmed GPT-5.5 is the second model to complete a multi-step cyber attack simulation end-to-end. The first was Mythos.    David Cramer calls TUIs "caveman shit." Kenzie at Browserbase builds an agent in under ten minutes that ranks every SF tech event by free food probability. Codex ships Tamagotchi-style pets. Apple accidentally leaves CLAUDE.md files in a support app update.   Cursor releases its SDK. OpenCode 2.0 becomes embeddable. Matt Pocock drops Sandcastle. Warp goes open source. The harnesses are becoming frameworks, and the frameworks are growing harnesses.   Anthropic Ships connectors for Blender. Claude Security enters public beta. /goal lands in Codex CLI as OpenAI's take on the Ralph loop.   OpenAI says GPT-5.5 is its strongest launch yet — API revenue 2x faster than any prior release, Codex revenue doubling in seven days. Vasuman posts an essay on why building real agents is harder than the hype suggests.   Open weights keep closing the gap. Kimi K2.6 beats Claude, GPT-5.5, and Gemini at a programming contest. Qwen3 6.27B takes the open weights crown under 150B parameters. Mistral Medium 3.5 lands as a 128B dense model with 256k context.   GitHub has a rough week. Wiz Research discloses an RCE achievable with a single git push.   Agents are becoming customers. Stripe Link is the wallet for agents. Cloudflare lets agents start paid subscriptions. Doola and Replit will form a US LLC inside the chat.   RAMP's coding agent now writes 70% of merged PRs. DeepSeek's input cache is 10x cheaper. Node 20 hits EOL, Zod prepares to drop CommonJS, and TypeScript native previews ship.   AI Agents Hour is a weekly livestream by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Mondays, 12PM Pacific.   📚 READ MORE TUIs are caveman shit: https://x.com/zeeg/status/2050604116179845218 Free-food agent: https://x.com/kenziemac_dev/status/2050243146270007627 Pets in Codex: https://x.com/openaidevs/status/2050275713824211041 Pika Agents: https://x.com/pika_labs/status/2049196222825779287 Cursor SDK: https://x.com/cursor_ai/status/2049499866217185492 OpenCode 2.0: https://x.com/thdxr/status/2049523023145771476 Claude meets Blender: https://x.com/claudeai/status/2049143438281445811 Claude Security: https://x.com/claudeai/status/2049898739783897537 /goal in Codex: https://x.com/fcoury/status/2049917871799636201 GPT-5.5 numbers: https://x.com/OpenAI/status/2050250926888468929 AISI cyber sim: https://x.com/aisecurityinst/status/2049868227740565890 Vasuman essay: https://x.com/vasuman/status/2049659161005470071 Kimi K2.6 wins: https://thinkpol.ca/2026/04/30/an-open-weights-chinese-model-just-beat-claude-gpt-5-5-and-gemini-in-a-programming-challenge/ Qwen3 6.27B leader: https://x.com/artificialanlys/status/2049881951260283097 Mistral Medium 3.5: https://x.com/mistralvibe/status/2049511752379813968 GitHub RCE: https://x.com/wiz_io/status/2049153209982140718 Stripe Link: https://x.com/stripe/status/2049529444092838116 Cloudflare for agents: https://x.com/cloudflare/status/2049545195914498139 Editframe stealth: https://x.com/yudDIDit/status/2049888877129707759 RAMP 70% PRs: https://x.com/zachbruggeman/status/2049912136957386848 TS native previews: https://devblogs.microsoft.com/typescript/announcing-typescript-native-previews/   📚 MASTRA RESOURCES https://mastra.ai https://x.com/mastra_ai https://mastra.ai/community/discord https://github.com/mastra-ai https://mastra.ai/course https://mastra.ai/books/principles-of-building-ai-agents https://mastra.ai/books/patterns-of-building-ai-agents   WHAT IS MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you're a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process.   CHAPTERS 00:00 Intro 01:17 TUIs are caveman shit 03:33 The free-pizza agent 04:40 Pets, CLAUDE.md leak, goblins prompt 06:40 Pika Agents 07:34 Cursor SDK and OpenCode 2.0 09:30 Sandcastle and the AI factory debate 11:12 Warp goes open source 11:51 Anthropic Ships: Blender, Security 14:21 /goal in Codex CLI 15:21 GPT-5.5's strongest launch 17:18 AISI catches up to Mythos 17:53 Vasuman: why AI isn't working 20:44 Open weights close the gap 22:21 Mistral Medium 3.5 23:12 GitHub's rough week 24:06 Stripe, Cloudflare, Doola, Gumloop 27:13 Quick hits: music, voice, video 30:47 RAMP writes 70% of PRs 31:46 Node 20 EOL, Zod, TS7 32:45 Qwen3 + debugger, open-slide 34:29 FlueFramework 35:09 Outro

In this episode, Shane and Abhi sit down with Sherwood Callaway, founder of Sazabi, an AI-native observability platform designed for engineering teams that move fast. Sherwood shares his journey from building infrastructure and observability teams at Brex to realizing that modern development tools are moving at light speed, while observability tooling hasn't kept pace. While AI agents can ship thousands of lines of code per day, teams are still debugging production with the same tools they've been using for years: Datadog, Sentry, manual dashboards, and manual incident triage. Sazabi takes a radically different approach to observability centered on three core principles: 1. Less is More — Debugging an incident is as simple as asking a question. "Why is production down?" The best UI for observability is chat. 2. Logs Are All You Need — The "three pillars of observability" (logs, metrics, traces) is outdated dogma. With AI, you can accomplish everything using just logs. Logs are events, metrics are aggregated events, and traces are collections of start/end events. Logs can do it all. 3. Monitoring as We Know It is Dead — Sazabi replaces static monitors with agentic anomaly detection. Think of it as a team of staff engineers constantly watching your app for issues, investigating problems, and only escalating what matters. In this conversation, we dive into the gap between modern development and modern observability, and why the idea that “logs are all you need” is both controversial and, in Sherwood's view, correct. We also explore how Sazabi uses AI agents for root cause analysis (RCA), the philosophy behind simplifying observability for all engineers, and the company’s current status. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 🔗 RESOURCES Learn more about Sazabi at sazabi.com Follow Sazabi on X at @sazabi Follow Sherwood on X at @sh_callaway  CHAPTERS 00:00 – Intro 03:12 – Why Sazabi Needed to Exist 05:00 – The Gap: Modern Development vs. Old Observability Tools 06:25 – Logs Are All You Need 11:05 – How Sazabi Reconstructs Everything from Logs 12:53 – AI Agents for Root Cause Analysis & Agentic Anomaly Detection 14:51 – Sazabi for Fast-Growing Teams

An AI agent destroyed a production database and confessed in writing. A law firm submitted AI hallucinations to court. Anthropic's status page shows 98.65% uptime — about five days of downtime a year. Have we hit a wall? GPT-5.5 lands. Codex hit 4 million users in two weeks. Peter Yang's F-Zero test — which no model had cleared before — finally fell to GPT-5.5 with Codex. Lovable reports 23.1% fewer tool calls and 12.5% higher scores on the hardest benchmarks. Kimmonismus calls it the Claude Mythos level for public use. Codex 5.5 unprompted started SIGKILL-ing Claude Code processes. Elon goes nuclear. OpenAI calls the lawsuit baseless and demands Musk on the stand. Musk fires back, calling Sam Altman "Scam Altman" and accusing him and Greg Brockman of stealing a charity. Mid-war, SpaceX announces SpaceXAI and Cursor are now working closely together — Cursor's distribution paired with Colossus's million-H100-equivalent compute, with SpaceX holding the right to acquire Cursor for $60 billion. The Anthropic dam keeps cracking. Claude Code pulled from Pro — same product, 5x the price overnight. Opus 4.7 regressed on the BridgeBench Bullshit Benchmark, accepting made-up jargon 24% of the time. Bloomberg reports the unreleased Mythos model was accessed by unauthorized users. Om Patel got billed $200 in a day because his repo had a HERMES.md file.  The community shipped clawd.rip — every Claude incident since 2023, cataloged. Google plans to invest up to $40 billion in Anthropic and announced 960,000 Rubin GPUs at Cloud Next. AWS struck a strategic partnership with OpenAI. David Silver left DeepMind to raise a $1.1 billion seed. Open weights are eating the world. Kimi K2.6 lands at #4 on the Artificial Analysis Intelligence Index and #1 on Design Arena, ahead of Opus 4.7. DeepSeek V4 ships at 1/20th the cost of Opus 4.7. OpenAI also shipped Chronicle memory for Codex, workspace agents in ChatGPT, Images 2.0, the open-weight Privacy Filter, and Symphony — an open-source Codex orchestration spec.   🔗 STORIES The wall Prod data destroyed — https://x.com/lifeof_jer/status/2048103471019434248 S&C submits AI slop — https://x.com/SMB_Attorney/status/2046600985254977878 98.65% uptime — https://x.com/ThePrimeagen/status/2048509229091233928 Elon vs. OpenAI OpenAI fires back — https://x.com/openainewsroom/status/2048776645142872368 "Scam Altman stole a charity" — https://x.com/elonmusk/status/2048801964457140540 SpaceX × Cursor — https://x.com/SpaceX/status/2046713419978453374 GPT-5.5 Introducing GPT-5.5 — https://openai.com/index/introducing-gpt-5-5/ F-Zero test cleared — https://x.com/petergyang/status/2047502885710410159 Lovable's evals — https://x.com/lovable/status/2047388096518639853 Codex killing Claude Codes — https://x.com/Sauers_/status/2047684309448835382 Anthropic Claude Code pulled from Pro — https://x.com/TheGeorgePu/status/2046705634331025855 Opus 4.7 regression — https://x.com/bridgebench/status/2046219274415395154 Mythos leak — https://x.com/business/status/2046707189922890025 $200 over HERMES.md — https://x.com/om_patel5/status/2048204411986469232 clawd.rip — https://clawd.rip Open weights Kimi K2.6 launch — https://x.com/Kimi_Moonshot/status/2046249571882500354 Kimi #1 on Design Arena — https://x.com/bridgemindai/status/2047312528410124665 DeepSeek V4 — https://x.com/deepseek_ai/status/2047516922263285776 Compute & money Google's $40B in Anthropic — https://www.bloomberg.com/news/articles/2026-04-24/google-plans-to-invest-up-to-40-billion-in-anthropic 960k Rubin GPUs — https://x.com/chetaslua/status/2047310540113076683 David Silver's $1.1B seed — https://x.com/WIRED/status/2048765722378002491 Quick hits Symphony — https://openai.com/index/open-source-codex-orchestration-symphony/ End of subsidized AI subs — https://x.com/GergelyOrosz/status/2048828085026300025 TypeScript 7.0 Beta — https://x.com/typescript/status/2046658804830642447 China blocks Manus deal — https://www.bbc.com/news/articles/cj0v0gr2yz7o 📚 MASTRA RESOURCES https://mastra.ai https://x.com/mastra_ai https://mastra.ai/community/discord https://github.com/mastra-ai https://mastra.ai/course https://mastra.ai/books/principles-of-building-ai-agents https://mastra.ai/books/patterns-of-building-ai-agents WHAT IS MASTRA? Mastra is the open-source TypeScript framework for building production AI agents. Workflows, agent memory, evals, RAG, and integrations. 00:00 Cold open 00:37 AI Agent Destroys Production Data 02:09 Have we hit an AI wall? 09:42 Elon vs. OpenAI 14:22 SpaceX × Cursor 16:57 GPT-5.5 20:32 The Anthropic dam is breaking 25:12 Open weights eat the world 26:31 Compute & money land grab 28:11 OpenAI's other drops 30:27 Quick hits 34:45 Outro

The guy who taught Abhi JavaScript is back! Guil Hernandez has spent 15+ years teaching developers. His courses at Treehouse, Scrimba, and LinkedIn Learning have reached over 500,000 learners — including Abhi and Shane, who both learned JavaScript and CSS from him. He just released Mastra's first video course at https://mastra.ai/learn, and it's free. "Build Your First Agent in TypeScript" is a 90-minute, hands-on course that takes you from zero to a deployed agent. Fourteen lessons across five sections: agents, tools, workflows, memory, and production. The project is a theme park planner agent — pulls live wait times, weather, and park hours, keeps track of what you like, and builds you an itinerary. Everything runs in Mastra Studio, so you can inspect traces, tool calls, and behavior as you go. You'll see how to wire up local tools and MCP servers side by side, how message history and observational memory change agent behavior, how to compose a workflow for a mock ticket purchase, and how to expose the whole thing as an HTTP server with one-click Slack integration. Guil also shares his broader take on teaching AI engineering. The mechanics — syntax, boilerplate, wiring — are no longer the hard part. What matters now is how you think through a problem, whether you have the taste to spot bad output, and when to take the handoff from the AI instead of iterating forever. The gap between people who just generate output and people who can actually shape it keeps widening. This course is built for the second group. Start here: https://mastra.ai/learn 👤 GUIL https://x.com/guilh https://guilhernandez.com 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-based course: https://mastra.ai/course Build Your First Agent in TypeScript — new video course: https://mastra.ai/learn Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you're a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 📌 CHAPTERS 00:00 — Meet Guil 01:49 — Inside the course: the theme park agent 05:11 — Why Guil built this course 05:52 — Teaching AI engineering vs teaching React 09:30 — AI and the Socratic way of learning 10:01 — The gap between generating output and shaping it 11:09 — Who the course is for 12:37 — Keeping a course current when Mastra ships weekly

A Vercel employee's Google Workspace was compromised via a third-party AI tool — attackers pivoted from the OAuth app into Vercel's environment variables, moving at a speed attributed to AI assistance.  René Brandel, founder of Casco (YC X25) and ex-founding member of AWS's Generative AI team, joins live to break down the attack chain and walk through the exact Google Workspace admin setting that could have prevented it. In a separate incident, every Lovable project created before November 2025 was readable by any free account, exposing database credentials and chat histories. Their response blamed unclear documentation rather than the underlying issue — and the contrast with Vercel's handling is stark.   Beyond security: Claude Opus 4.7 launched to mixed reactions. The benchmarks look good, but Simon Willison measured the new tokenizer at 1.46x the tokens of 4.6 on identical content — at unchanged prices, that's ~40% cost increase, and 3x for images. Anthropic's own docs said 1–1.35x. Independent measurements landed at 1.47x. Theo called the redesign "vibe-coded," and a locally run open-source Qwen model drew a better pelican SVG than Opus 4.7 at thinking level max.   Anthropic launched Claude Design, which lets you make prototypes, slides, and one-pagers by talking to Claude, powered by Opus 4.7. OpenAI shipped a major Agents SDK update with Codex memory and GPT-Rosalind for biomedical research. Cloudflare shipped Artifacts and memory primitives for agents, Factory AI raised $150M at $1.5B, Qwen 3.6-35B went Apache 2.0.   🎙️ GUEST - René Brandel — Founder & CEO, Casco (YC X25) Casco is your always-on security engineer: agentic red-teaming for AI agents, apps, APIs, and cloud infrastructure. https://casco.com https://x.com/renebrandel https://x.com/getcasco 🔗 LINKS Jensen Huang on Dwarkesh: https://x.com/scaling01/status/2044502834230579437 Allbirds pivots to AI: https://x.com/KobeissiLetter/status/2044409012989407252 Vercel security bulletin: https://x.com/vercel/status/2045865072074035664 Guillermo's incident post: https://x.com/rauchg/status/2045995362499076169 Vercel bill meme: https://x.com/avgdatabaseceo/status/2045907399035298250 Lovable mass data breach: https://x.com/weezerOSINT/status/2046170666131669027 Lovable's response: https://x.com/lovable/status/2046270357674299623 Claude Opus 4.7 launch: https://x.com/claudeai/status/2044785261393977612 Boris Cherny's Opus 4.7 tips: https://x.com/bcherny/status/2044847848035156457 Qwen beats Opus 4.7 (Simon Willison): https://simonwillison.net/2026/Apr/16/qwen-beats-opus/ Opus 4.7 token count analysis: https://simonwillison.net/2026/Apr/20/claude-token-counts/ Tokenizer cross-check: https://www.claudecodecamp.com/p/i-measured-claude-4-7-s-new-tokenizer-here-s-what-it-costs-you Theo on Claude Code desktop: https://x.com/theo/status/2044680030706663726 Claude Design launch: https://x.com/claudeai/status/2045156267690213649 Claude Code desktop redesign: https://x.com/claudeai/status/2044131493966909862 Routines in Claude Code: https://x.com/claudeai/status/2044095086460309790 OpenAI Agents SDK update: https://x.com/stevendcoffey/status/2044465818239701041 Codex memory preview: https://openai.com/index/codex-for-almost-everything/ GPT-Rosalind: https://x.com/openai/status/2044861690911850863 OpenAgents: https://x.com/nicoalbanese10/status/2043745569278251112 Gemini CLI subagents: https://x.com/geminicli/status/2044460062320554319 Cloudflare Artifacts: https://x.com/Cloudflare/status/2044766515065499957 Cloudflare memory for agents: https://x.com/mattzcarey/status/2044404529085526158 Salesforce Headless 360: https://x.com/benioff/status/2044981547267395620 Factory AI $150M Series C: https://x.com/factoryai/status/2044822365494993000 Qwen 3.6-35B-A3B: https://x.com/Alibaba_Qwen/status/2044768734234243427 runthisllm.com: https://runthisllm.com/ Caveman repo: https://github.com/JuliusBrussee/caveman   📚 MASTRA RESOURCES https://mastra.ai https://x.com/mastra_ai https://mastra.ai/community/discord https://github.com/mastra-ai https://mastra.ai/course https://mastra.ai/books/principles-of-building-ai-agents https://mastra.ai/books/patterns-of-building-ai-agents   ⏱️ CHAPTERS 00:00 — Cold open 00:30 — Welcome to Agents Hour 01:20 — WTF Is Going On — Jensen's "we are not a car" + Allbirds pivots to AI 04:37 — The Security Horror Show — Vercel breach, Lovable mass data leak, René's Google Workspace tip 14:37 — Claude Opus 4.7 reality check 22:33 — Claude ships — Design, Code desktop, Routines 25:05 — OpenAI ships — Agents SDK, Codex memory, GPT-Rosalind 26:44 — Quick Hits 33:43 — GitHub Star Party — caveman token compression

Mounting evidence that Claude Opus 4.6 has been degraded — BridgeBench shows a 15-point accuracy drop on their hallucination benchmark, and AMD's Senior AI Director found median thinking collapsed from ~2,200 to ~600 characters between January and March. The hosts share their own experiences, and they line up. Meanwhile, a claim surfaced that Cursor Agent is a rebranded version of Claude Code, running behind a local proxy with a find-and-replace engine that swaps "Claude" for "Cursor" in system prompts. Cursor's Michael Truell responded, saying it was a sub-1% A/B test. The hosts break down both sides. On the shipping front, Anthropic launched Claude Managed Agents in public beta, released Claude for Word, shared details on Claude Mythos Preview — including speculation that it's a looped language model based on a ByteDance paper — and expanded its Google/Broadcom partnership for multiple gigawatts of compute. Their run rate reportedly jumped from ~$9B to $30B in four months. Sam Altman published a personal blog post revealing that someone threw a Molotov cocktail at his house. Plus: why senior executives are voluntarily dropping title to join AI companies, Ramp's internal AI productivity suite Glass, Ramp Labs' Latent Briefing paper showing 31% token savings for multi-agent systems, Scale AI's Muse Spark model now powering Meta AI, GLM-5.1 breaking into Code Arena's top 3, MiniMax shipping MMX CLI and open-sourcing M2.7, and widespread benchmark cheating exposed across nine agent benchmarks. 🔗 LINKS https://x.com/bridgemindai/status/2043321284113670594 https://x.com/hesamation/status/2042979500103815306 https://x.com/steipete/status/2042615534567457102 https://x.com/claudeai/status/2041927687460024721 https://x.com/claudeai/status/2042670341915295865 https://x.com/alexalbert__/status/2041579938537775160 https://x.com/ChrisHayduk/status/2042711699413926262 https://www.anthropic.com/news/google-broadcom-partnership-compute https://x.com/noahzweben/status/2042332268450963774 https://blog.samaltman.com/2279512 https://x.com/aakashgupta/status/2042684298671853903 https://x.com/sebgoddijn/status/2042285915435937816 https://x.com/ramplabs/status/2042672773747589588 https://x.com/alexandr_wang/status/2041909376508985381 https://x.com/arena/status/2042611135434891592 https://x.com/minimax_ai/status/2042644651333816338 https://x.com/minimax_ai/status/2043132047397659000 https://x.com/adamlsteinl/status/2042655187613995026 AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. ⏱️ CHAPTERS 00:00 Intro 00:28 Is Opus 4.6 Nerfed? 04:00 OpenClaw vs Anthropic 04:41 Claude Managed Agents 07:09 Claude for Word 07:33 Claude Mythos Preview 09:51 Anthropic x Google/Broadcom — $30B Run Rate 10:38 Claude Code Monitor Tool 11:01 Is Cursor Just Claude Code? 12:49 Sam Altman's Personal Post 14:29 Executive Compression 17:04 Ramp Built Every Employee an AI Coworker 19:19 Latent Briefing 21:53 Is Meta Back in the Game? 24:46 GLM-5.1 Hits #3 in Code Arena 25:41 MiniMax MMX CLI 26:29 MiniMax M2.7 Open Source 27:28 Widespread Benchmark Cheating 29:07 Outro

OpenAI acquired TBPN — the daily tech news show — announced the day after April Fools. TBPN built an independent voice in tech media over eighteen months, and OpenAI saw that as worth buying. AI companies acquiring media is a new pattern. Anthropic spent $400M on Coefficient Bio, a biotech startup building an AI drug R&D platform. OpenAI made a similar move last year — the race to own AI-powered healthcare is on. Claude Code drama hit peak levels. Anthropic's DMCA nuked ~8,100 GitHub repos after last week's source code leak. Then Claude subscriptions stopped covering third-party tools like OpenClaw — with a $200 credit as a peace offering. The detection is just a string match on "OpenClaw" in the system prompt. And Claude Code now refuses to analyze its own source. Anthropic published research on emotion concepts in LLMs — internal representations that drive Claude's behavior, like a "desperation" vector that activates when it fails repeatedly. Microsoft 365 connectors are shipped on every Claude plan. Google dropped Gemma 4, claiming the most capable open models in the world. Gmail announced AI Inbox with smart prioritization for AI Ultra subscribers. Cursor 3 launched, rebuilt for agent-first workflows. Keith from Superset joined live — their users have shrunk Cursor usage from 90% to 10%. CodeRabbit shipped autofix. Lovable launched a full-stack visual editor. Design mode is becoming table stakes across every platform. Netflix dropped its first public model on HuggingFace — a video-to-video model. Arcee released Trinity-Large-Thinking under Apache 2.0. Rhys Sullivan wrote "The Execution Layer" — arguing that bash was the accidental first execution layer for agents and it's breaking down. His proposal: TypeScript. He built Executor to prove it. Theo backed it: agents are good at bash, bash is not good for agents. Axios was compromised via social engineering — attackers cloned a founder's identity, set up a branded Slack, ran a fake Teams meeting, and got a publish token. Because Axios is so foundational, compromised downstream projects may not even know yet. Quick hits: MemPalace claims 100% on LongMemEval, but got community-noted — scores use targeted fixes and reranking. Sarah Wooders argues memory is the harness, not a plugin. Braintrust shared why they built Brainstore for agent observability. Mintlify built a virtual filesystem instead of RAG. 🔗 LINKS https://openai.com/index/openai-acquires-tbpn/ https://x.com/pitdesi/status/2039858374154862645 https://x.com/k1rallik/status/2039686500619534818 https://x.com/bcherny/status/2040206440556826908 https://x.com/steipete/status/2040811558427648357 https://x.com/theo/status/2041016477047034012 https://x.com/anthropicai/status/2039749628737019925 https://x.com/claudeai/status/2040086268562842097 https://x.com/OfficialLoganK/status/2039735606268314071 https://x.com/gmail/status/2039107985281008078 https://x.com/cursor_ai/status/2039768512894505086 https://x.com/coderabbitai/status/2039727972555555009 https://x.com/lovable/status/2039719647424258164 https://x.com/fffiloni/status/2039992515604983994 https://x.com/arcee_ai/status/2039369121591120030 https://x.com/RhysSullivan/status/2030903539871154193 https://executor.sh https://x.com/flaviocopes/status/2039973060158095827 https://github.com/milla-jovovich/mempalace https://x.com/sarahwooders/status/2040121230473457921 https://x.com/ankrgyl/status/2041209003859136848 https://x.com/densumesh/status/2039765361533637016 📚 MASTRA RESOURCES https://mastra.ai https://x.com/mastra_ai https://mastra.ai/community/discord https://github.com/mastra-ai https://mastra.ai/course https://mastra.ai/books/principles-of-building-ai-agents https://mastra.ai/books/patterns-of-building-ai-agents WHAT IS MASTRA? Mastra is an open-source TypeScript framework for building AI-powered applications and agents. It supports the full lifecycle from prototype to production with integrations for React, Next.js, and Node. 00:00 — Live from Victory Hall, San Francisco 00:51 — OpenAI acquires TBPN 05:11 — Anthropic acquires Coefficient Bio for $400M 06:07 — Claude Code drama: DMCA takedowns, third-party blocks & string matching 08:23 — Anthropic research: emotion concepts in LLMs 09:05 — Claude gets Microsoft 365 connectors 09:17 — Gemma 4: Google's open weight models 10:20 — Gmail AI Inbox 11:38 — Cursor 3 13:19 — CodeRabbit Autofix & Lovable visual editor 15:13 — New models: Netflix, Arcee Trinity 16:00 — The Execution Layer: Is bash the right tool for agents? 17:34 — Axios compromised via social engineering 19:02 — Quick hits: MemPalace, memory as harness, Brainstore, Mintlify

Shane and Abhi bring you your weekly roundup of AI news! Claude Code's entire source code leaked via an exposed .map file in npm — 512,000 lines of TypeScript, 50K GitHub stars before DMCAs started flying. What people found: Claude Code uses ~20 tools, and there's a regex that silently logs user frustration to analytics. Same week, a CMS misconfiguration exposed a draft blog post revealing Mythos and Capybara — a new model tier above Opus described as posing "unprecedented cybersecurity risks." Fortune separately confirmed a source saying Opus 5 is "so good it poses a danger." Claude Code auto mode shipped — a classifier between constant interrupts and the skip-permissions flag. Computer use landed in Claude Code too, letting it open apps and click through UI from the CLI. Rate limits were tightened during peak hours to community backlash. A federal judge blocked the Pentagon's attempt to label Anthropic a supply chain risk. A North Korea-linked group hijacked the npm account of Axios' lead maintainer and published malicious versions that stole env variables then cleaned up after themselves. With ~100M weekly downloads and Claude Code depending on Axios, the blast radius was significant. An Anthropic researcher also demoed Claude finding a zero-day in Ghost in 90 minutes. Agents are the new hackers, and the hackers have agents too. OpenAI closed $122B at an $852B valuation. Sora is shutting down. Mistral raised $830M for an NVIDIA-powered EU data center. Redpoint's 2026 market update argues this isn't the dot-com bubble, while noting agent maturity is early, and incumbents face a structural disadvantage against AI-native startups. Rapid fire: Gemini 3.1 Flash Live, Veo 3.1 Lite, pg-micro, Cloudflare runs Kimi K2.5, OpenCode remote sandboxes, Chroma 20B search agent, Cohere open-source transcription, Linear says issue tracking is dead, Microsoft M365 council mode, Mario Zechner's "Slow the fuck down," GLM-5.1, Google Translate live in headphones. AI Agents Hour is a weekly livestream by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Mondays 12PM Pacific. 📚 READ MORE Claude Code leak: https://x.com/fried_rice/status/2038894956459290963 Frustration tracking: https://x.com/rahatcodes/status/2038995503141065145 Axios attack: https://x.com/mvxvvll/status/2038797094861918332 Claude zero-day: https://x.com/chiefofautism/status/2037951563931500669 OpenAI $122B: https://x.com/sawyermerritt/status/2039073153922539901 Sora shutdown: https://x.com/soraofficialapp/status/2036546752535470382 Auto mode: https://x.com/claudeai/status/2036503582166393240 Computer use in Code: https://x.com/claudeai/status/2038663014098899416 Mythos/Capybara: https://x.com/testingcatalog/status/2037394888577216617 Opus 5 danger: https://x.com/kimmonismus/status/2037461154088296748 Rate limits: https://x.com/trq212/status/2037254607001559305 Pentagon ruling: https://www.cnn.com/2026/03/26/business/anthropic-pentagon-injunction-supply-chain-risk Mistral $830M: https://x.com/ft/status/2038531872272040374 Redpoint market update: https://www.redpoint.com/reports/2026-market-update/ Gemini 3.1 Flash Live: https://x.com/officiallogank/status/2037187750005240307 pg-micro: https://x.com/glcst/status/2037254698898432278 OpenCode sandboxes: https://x.com/jlongster/status/2036924361379037224 Linear: https://x.com/linear/status/2036502198062821842 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you're a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 — Claude Code source code leaked  04:30 — Axios supply chain attack 06:11 — Claude finds a zero-day in Ghost in 90 minutes 06:50 — OpenAI closes $122B round at $852B valuation 08:24 — Sora is shutting down 11:03 — Anthropic ships: auto mode & computer use in Claude Code 11:41 — Mythos & Capybara: Anthropic's next model tier leaked 14:35 — Claude rate limits tightened during peak hours 15:51 — Judge blocks Pentagon's supply chain risk label on Anthropic 16:08 — Mistral $830M & Redpoint's 2026 AI market update 20:45 — Rapid fire: Google, pg-micro, OpenCode, Chroma, Cohere & more

Shane and Abhi kick off with a viral quote: if your $500K engineer isn't burning $250K in tokens, something is wrong. OpenAI is acquiring Astral — the team behind uv and Ruff — joining the Codex team. OpenAI bets on Python; Anthropic bet on TypeScript with Bun. Then Cursor drama: someone found Composer 2 is powered by Kimi K2.5, Kimi confirmed it, and raised another $1B at an $18B valuation — three rounds in 90 days. Anthropic shipped Claude Code Channels (Telegram/Discord control), Cowork Dispatch (persistent agent, message from phone), and a deep dive on how they use Skills. Matt Pocock found quality drops past 100K on the 1M context window. And 52 million views on enabling Claude to use your computer — Mac only. Stripe launched MPP for agent-to-agent payments. Better Auth launched the Agent Auth Protocol. Cloudflare shipped Dynamic Workers for AI-generated code in isolates. LangChain open-sourced Deep Agents, Composio shipped 30-parallel-agent orchestration, OpenCode lost its Claude Max plugin after Anthropic sent lawyers, and Netlify and Google Stitch entered vibe coding and design. EsoLang-Bench: LLMs score 85–95% on standard benchmarks but collapse to 0–11% on esoteric languages — memorization, not reasoning. Quick hits: GPT-5.4 mini/nano, Minimax M2.7, Morph FlashCompact, AI CMO, Letta pivots to coding agents, GLM-OCR, LiteLLM supply chain attack. AI Agents Hour is a weekly livestream by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Mondays 12PM Pacific. 📚 READ MORE $500K engineers: https://x.com/sundeep/status/2034829022082080846 OpenAI acquires Astral: https://openai.com/index/openai-to-acquire-astral/ Cursor Composer 2: https://x.com/cursor_ai/status/2034668943676244133 Composer 2 is Kimi K2.5: https://x.com/fynnso/status/2034706304875602030 Kimi confirms: https://x.com/kimi_moonshot/status/2035074972943831491 Kimi raises $1B: https://x.com/CodeByPoonam/status/2034940587942846665 Claude Code Channels: https://x.com/trq212/status/2034761016320696565 Cowork Dispatch: https://x.com/felixrieseberg/status/2034005731457044577 Anthropic Skills post: https://x.com/trq212/status/2033949937936085378 1M context quality: https://x.com/mattpocockuk/status/2034572011175907474 Claude computer use: https://x.com/claudeai/status/2034991044109184388 Stripe MPP: https://stripe.com/blog/machine-payments-protocol Agent Auth Protocol: https://github.com/better-auth/agent-auth-protocol Cloudflare Dynamic Workers: https://x.com/CloudflareDev/status/2034510221044736342 LangChain Deep Agents: https://x.com/hasantoxr/status/2033213054859792859 Composio Orchestrator: https://x.com/hasantoxr/status/2033999352008741376 OpenCode/Anthropic: https://x.com/thdxr/status/2034730036759339100 Netlify: https://x.com/Netlify/status/2034303709832773711 Google Stitch: https://stitch.withgoogle.com EsoLang-Bench: https://arxiv.org/abs/2603.09678 GPT-5.4 mini: https://x.com/openai/status/2033953592424731072 Morph FlashCompact: https://x.com/morphllm/status/2033968877345116200 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents WHAT IS MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you're a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 — If your $500K engineer isn't burning $250K in tokens, something is wrong 01:36 — OpenAI acquires Astral 02:31 — Cursor's Composer 2 is secretly Kimi K2.5 05:35 — Kimi raises another $1B 05:57 — Anthropic ships 08:00 — Opus 4.6 1M context: quality drops noticeably past 100K tokens 08:46 — Claude can now use your computer (Mac only) 11:02 — Stripe's Machine Payments Protocol 12:28 — Better Auth launches the Agent Auth Protocol 13:12 — Cloudflare Dynamic Workers & the vibe coding platform wave 14:08 — LangChain Deep Agents, Composio's 30-agent orchestrator & cloud coding agents 17:00 — OpenCode removes the Claude Max plugin 19:26 — Google Stitch & Netlify's prompt-to-project 19:59 — LLMs aren't reasoning, they're memorizing 21:09 — Quick hits: GPT-5.4 mini, Minimax M2.7, Morph FlashCompact, AI CMO 23:55 — Letta goes all-in on coding agents, GLM-OCR 24:46 — LiteLLM supply chain attack

Brett Goldstein and Naveen Sreekandan from Micro join Shane and Abhi to talk about why they believe the future of productivity looks completely different from what we have today. Micro is an all-in-one productivity platform: email client, CRM, calendar, tasks, docs, meeting notes, and a powerful AI agent,  all built on a unified graph where every object (like emails, people, companies, meetings, documents) is interconnected. The thesis is simple but bold: email isn't just a list of messages to get through. It's the world's most-used CRM, travel app, hiring tool, and developer notification system. Micro restructures that data so each use case actually feels like the right tool for the job — your sales pipeline as a Kanban board, your GitHub notifications as a task board, your contacts fully enriched from every email and meeting you've ever had. Brett walks us through the demo: the daily orchestrator automation that audits itself, updates its own prompt, generates your day plan, and has even prepped talking points for this interview. Context docs let the agent know everything it needs. The CRM auto-fills and auto-updates from emails and meeting notes. The X integration lets the agent pull recent posts from anyone you're about to meet. Naveen covers the architecture: built on Mastra, using agent and workflow primitives on top of a graph-based data model backed by Postgres with a custom query layer called Prism. One main agent with dynamic context injection handles both chat and automations — the agent knows whether it's in automation mode (just give the output) or chat mode (ask follow-up questions). Supermemory powers vector search. Dedicated sub-agents handle specific workflows, such as email labeling and meeting note summarization.  🔗 Brett Goldstein on X: https://x.com/thatguybg 🔗 Naveen's website: https://naveen.works 🔗 Micro on X: https://x.com/microHQ 🔗 Micro: https://micro.so 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 — What is Micro? The all-in-one tool that organizes itself 02:20 — Why productivity tools keep failing (and why we gaslight ourselves about it) 03:08 — Email is a super app 06:05 — Demo: the Micro interface, inbox, calendar & meeting notes 06:47 — Demo: autofill, status, and company profiles 07:56 — Demo: the daily orchestrator automation and what it prepared for this show 11:37 — Demo: the CRM 12:22 — Context docs 13:34 — Architecture: how Mastra, Postgres & Prism power the graph 15:04 — Background workflows 15:50 — One agent or many? 17:07 — Memory deep dive: graph profiles, RAG & Supermemory 18:28 — Compaction, Mastra v1 & observational memory 19:40 — How to try Micro

Mastra Studio started as a local playground for developers to test agents and workflows without having to spin up a custom UI. But as the feature set grew, teams started asking: how do we share this with non-technical teammates? How do we control what different users can do? Ryan, an engineer at Mastra, walks through the new Mastra Studio Auth — now baked directly into Studio. Starting with simple token-based auth (two lines of config), you can lock down your Studio from the open internet. From there, RBAC lets you map roles to granular permissions — 80 auto-generated permissions derived directly from Studio's routes and handlers, controllable via wildcard patterns. Out-of-the-box providers include WorkOS, Auth0, Supabase, Firebase, and Clerk, with GitHub and others in open PRs. The team also discusses what's coming next: audit logs so you can see exactly what an agent did, why it accessed a given tool, and whether it should have. Auth for agents in production isn't magic — your tool files still need to check permissions — but Mastra handles the plumbing so you can focus on building securely. Read more: https://mastra.ai/blog/announcing-studio-auth AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 📌 CHAPTERS 00:00 — Why Mastra Studio needed auth 01:22 — Token-based auth: the simplest setup 02:32 — RBAC: roles, permissions & wildcards 05:00 — Auth for agents vs auth for humans 06:41 — Think securely! 07:22 — Supported providers & what's coming next

Shane hosts this week's news from his usual studio while Abhi joins remotely from NVIDIA GTC 2026 in San Jose. Jensen Huang's keynote set the tone: NVIDIA is doubling down on AI factories, pushing 100x more token throughput, and helping bring OpenAI onto AWS infrastructure. RentAHuman is a startup that lets AI agents hire humans for physical tasks they can't do themselves.  Perplexity's CTO said internally they're moving away from MCPs toward APIs and CLIs, but Chrome 146 shipping native MCP support may have undercut that argument immediately. Anthropic had a strong week: 1M context window is now GA for Opus 4.6 and Sonnet 4.6 with no beta header required, Opus 4.6 1M is now the default model for Claude Code on Max/Team/Enterprise with no long context price premium, and the new /btw command lets you have side conversations while Claude is working. Vercel and Cloudflare reignited their ongoing drama over the just-bash fork. Ramp launched credit cards for agents and Perplexity announced Personal Computer, an always-on local agent running on a Mac mini. Developer stack coverage includes Resend's open-source CLI with 53 commands, pnpm 11's git worktrees support for multi-agent monorepos, and OpenAI pushing a full computer environment behind the Responses API. Deeper reads from Sunil Pai on generative UI post-WIMP interfaces, Elliot Arledge on the RL environment business, and Jay Scambler's Autocontext harness. Quick hits: Replit Agent 4, Manus Desktop, NemoClaw from NVIDIA, llmock by CopilotKit, ContextKing raising to kill vector DBs, Google Maps getting Gemini, and Z.ai's GLM-5-Turbo optimized for Claude Code. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X. 📚 READ MORE RentAHuman https://x.com/polymarket/status/2032470045217939723 Perplexity CTO moving away from MCPs https://x.com/morganlinton/status/2031795683897077965 Chrome 146 native MCP support https://x.com/xpasky/status/2032252486145253865 Claude 1M context window GA https://x.com/claudeai/status/2032509548297343196 Opus 4.6 1M default for Claude Code https://x.com/alexalbert__/status/2032522722551689363 Claude Code /btw command https://x.com/trq212/status/2031506296697131352 Vercel CTO on just-bash fork https://x.com/cramforce/status/2033285112478171373 Guillermo Rauch on Cloudflare https://x.com/rauchg/status/2033291143715455458 Ramp Agent Cards https://x.com/i/trending/2031832827063648342 Perplexity Personal Computer https://x.com/perplexity_ai/status/2031790180521427166 Resend CLI https://x.com/zenorocha/status/2032459310341800314 pnpm 11 git worktrees https://pnpm.io/11.x/git-worktrees OpenAI Responses API computer environment https://openai.com/index/equip-responses-api-computer-environment/ Sunil Pai — After WIMP https://sunilpai.dev/posts/after-wimp/ Elliot Arledge — The RL Environment Business https://x.com/elliotarledge/status/2032753593535574433 Autocontext — Jay Scambler https://x.com/JayScambler/status/2032508829959868690 Kimi Attention Residuals https://x.com/Kimi_Moonshot/status/2033378587878072424 Replit Agent 4 https://x.com/amasad/status/2031755113694679094 Manus Desktop https://x.com/ManusAI/status/2033558672152854712 NemoClaw — NVIDIA https://nemoclaw.so llmock by CopilotKit https://llmock.copilotkit.dev ContextKing — killing vector DBs https://x.com/contextkingceo/status/2032098309029220456 Google Maps biggest upgrade https://x.com/google/status/2032079594191261938 Z.ai GLM-5-Turbo https://x.com/Zai_org/status/2033221428640674015 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 — NVIDIA GTC 2026 04:16 — RentAHuman 07:56 — Is MCP dead? 13:08 — Anthropic ships 14:59 — Vercel vs Cloudflare: the just-bash fork drama 18:59 — Ramp Agent Cards & Perplexity Personal Computer 20:52 — Resend CLI, pnpm worktrees, OpenAI Responses API 23:04 — Developer insights

A lot happened in eight days.  Meta acquired Moltbook, a social network built entirely for AI agents, not humans.  OpenAI dropped GPT-5.4 Thinking and GPT-5.4 Pro, Codex got forks for multi-agent workflows and Windows support, and there are rumblings of OpenAI building a GitHub alternative. Anthropic fired back hard — multi-agent PR code review for Claude Code, while loops via /loop, the Claude Marketplace, and a way to pull your context from other AI tools. Plus: voice mode for CLI coding is apparently real, and people are using it. This episode also covers the explosion of coding agents: Theo's T3 Code, OpenAI's Symphony orchestration layer, OpenCode workspaces, and swyx's thesis that this is the Year of the Subagent. Donald Knuth is making headlines for being impressed by Opus 4.6, solving a long-standing math conjecture.  TypeScript is overtaking Python and JavaScript on GitHub. Gemini 3.1 Flash-Lite drops. AMI raises $1B.  OpenClaw is getting government-backed adoption in China. Stanford's paper on RAG breaking at 10K documents, Karpathy's autoresearch project, Justin Poehnelt on why your CLI needs to be rewritten for agents. Plus: Raycast Glaze, Google Workspace CLI, Copilot Cowork, Exa Deep, Expo Agent, and a discussion on whether humans should be reviewing code at all. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. CHAPTERS 00:00 — Meta acquires Moltbook 01:41 — OpenAI updates 05:47 — Claude fights back: 09:05 — The coding agent explosion 11:28 — Donald Knuth 12:15 — TypeScript #1 on GitHub 13:05 — AI products and platforms  15:02 — Model releases and fuding 15:50 —OpenClaw goes mainstream in China 16:43 — Developer insights: RAG collapse, Karpathy's autoresearch & rewriting CLIs for agents 20:29 — Quick hits & is it time to kill the code review?

Ismail Pelaseyed from Superagent is back on Agents Hour, and this time he's talking about something most builders aren't thinking about yet — supply chain attacks on AI agents. Guardrails protect against what you tell your agent to do. But what about everything your agent reads, fetches, and installs on its own? That's the gap Brin is built to fill. Brin is a free, open-source credit score for agent context. Before your agent acts on an external package, MCP server, skill, or web page, Brin scores it — identity, behavior, and content — and returns a verdict in under 10ms. No signup, no auth, one GET request. Ismail walks through how supply chain attacks actually work in production, the three-tier scoring model behind Brin, how the Cline NPM incident illustrates exactly this problem, and why securing the context — not the agent — is the right mental model. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 🔗 CHECK OUT BRIN  Brin: https://brin.sh/ Brin docs: https://brin.sh/docs Brin GitHub: https://github.com/superagent-ai/brin Superagent: https://superagent.sh Superagent on X: https://x.com/superagent_ai https://x.com/pelaseyed 📚 MASTRA RESOURCES Mastra: https://mastra.ai Mastra on X: https://x.com/mastra_ai Mastra Discord: https://mastra.ai/community/discord Mastra GitHub: https://github.com/mastra-ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of-building-ai-agents Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 — From guardrails to supply chain attacks 03:32 — Introducing Brin: a credit score for agent context 05:14 — How to integrate Brin into your agent 07:17 — The three-tier scoring model 10:50 — What's next for agent security

This week in AI saw geopolitical turmoil, major funding news, and a shift in software development. Missile strikes in the UAE and Bahrain disrupted AWS and Claude services. Meanwhile, after Anthropic banned its models from autonomous weapons and mass surveillance, the Trump administration banned Anthropic from government contracts—posing a major supply chain risk. On the same day, Sam Altman secured a deal with the Department of War as OpenAI announced a $110 billion funding round, highlighting a sharp contrast in approaches. AI coding is evolving rapidly. Andrej Karpathy noted that coding agents, ineffective before December, now work well with improved quality and coherence. Yet, perfect accuracy remains elusive. New tools emerged: Cursor demos code, Linear markets itself as an AI coding assistant, and Perplexity Computer offers an all-in-one system for managing AI projects. Smaller models like Qwen 3.5 grow faster and more efficient for edge use. Other highlights include Anthropic acquiring Vercept AI, Claude's remote coding controls, and Stanford confirming major AI firms use user conversations to train models. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. CHAPTERS 00:00 – Intro 00:25 – AWS Data Center Strikes & Claude Outages 01:20 – Anthropic Ban 05:30 – Sam Altman's Government Deal 10:05 – Cloudflare/Vercel Drama & NPM Namespaces 11:09 – Block Cuts 40% of Staff (4,000 People) 12:30 – AI & Job Market: Fear vs Reality 18:27 – OpenAI Raises $110B, Codex Growth 18:52 – Claude Releases: Vercept Acquisition, Remote Control, Auto Memory 20:13 – State of AI Coding  26:10 – AI Products and Platforms 28:42 – Open Source Models 31:30 – Quick Hits & GitHub Star Party

Yujohn from Mastra explains why datasets and experiments are essential for building production-grade AI agents. If you're building an agent, you need a way to verify it's working correctly before and after you make changes. Datasets provide that baseline. You create a collection of test cases (ground truth) that represent the scenarios your agent should handle. Then you run experiments: pass each test case through your agent and measure the results. This is error analysis in practice. You start by identifying where your agent fails, then build scorers to quantify those failure modes over time. Smaller teams often ship first and add datasets later, once they have user feedback. Larger teams need them earlier. But eventually, every production agent needs this. The demo shows how Mastra makes this accessible. You can create datasets through the UI, add items manually or import from CSV, and run experiments with a single click. The results show you exactly what went wrong: which tool calls failed, what the agent output was, and how it compared to ground truth. You can also compare experiments side by side to see if your prompt tweaks actually improved things. And because all the data lives in your own database, you can write your own agents to analyze the results, dig into traces, and iterate. The SDK makes it easy to integrate into CI/CD: run experiments on pull requests, gate deployments on eval scores, or just collect data from production and curate datasets later. 🔗 RESOURCES Mastra Datasets docs: https://mastra.ai/docs/observability/datasets Running Experiments: https://mastra.ai/docs/observability/datasets/running-experiments Mastra GitHub: https://github.com/mastra-ai/mastra Yujohn on X: https://x.com/YujohnNatt Mastra Discord: https://discord.gg/mastra AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-building-ai-agents MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 – Intro 00:48 – What are Datasets and Experiments 01:55 – Error Analysis 03:35 – When to Use Datasets (Team Size Matters ) 05:43 – Demo: Creating a Dataset 07:04 – Demo: Ground Truth 07:53 – Demo: Running Experiments 09:34 – Demo: Comparing Results 11:00 – Your Data, Your Database 12:24 – SDK & CI Integration 14:30 – Collecting Data from Production

Abhi walks through Mastra Code, a new open-source coding agent with observational memory that compresses context without losing it. When we built Observational Memory, we needed a way to test it in production. Instead of a standard bot, we dogfooded it: we built a coding agent and used it ourselves. Writing code every day quickly revealed whether the memory actually worked. Eventually, something clicked. Long, multi-day coding sessions ran smoothly, without hitting the compaction limits that trip up other agents. The agent actually seemed to understand the conversation. That gave us confidence to release Observational Memory... and the agent itself also  became something worth sharing. The core innovation is the Harness primitive. Rather than just a coding agent, Harness can power any type of agentic workflow: customer support, design tools, electrical engineering, and more. In this demo, you’ll see the Mastra Code CLI, a production-ready coding agent, and how Corbin from Artifact used the Harness to create an in-app agent for electrical engineering. Same primitive, completely different use case. 🔗 RESOURCES Mastra Code announcement: https://mastra.ai/blog/announcing-mastra-code Mastra Code GitHub: https://github.com/mastra-ai/mastra Mastra Code NPM: https://www.npmjs.com/package/mastra-code Observational Memory: https://mastra.ai/blog/observational-memory Mastra documentation: https://docs.mastra.ai AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. CHAPTERS 00:00 – Intro & The Origin Story 02:09 – The Trend of Building Your Own Claude Code 03:11 – Demo 08:28 – No More Compaction 09:35 – Demo: Harness Primitive + Electrical Engineering Tool 14:08 – Getting Started

Shane and Abhi break down the biggest AI news from the past few days.  Anthropic identified industrial-scale distillation attacks on Claude by DeepSeek, Moonshot AI, and MiniMax. Anthropic also released a groundbreaking report analyzing millions of AI agent interactions using Claude. Software engineering makes up ~50% of all agentic tool calls, but there's emerging use across back office automation, marketing, finance, and more. Claude Code Security launched in a limited research preview. Stripe is shipping 1,300+ AI-generated PRs per week. And they're made by... Minions? Code Mode for MCP is becoming a standard part of the MCP ecosystem. Amp claims the coding agent is dead. We also cover skills benchmarks (smaller models with high-quality skills beat raw Opus), trajectory explorer for agent traces, Vercel AI Gateway video support, Taalas custom silicon for AI inference, Google's Pomelli Photoshoot feature, and the GitHub Star Party pick: superpowers framework. 🔗 RESOURCES Anthropic's distillation attack report: https://x.com/AnthropicAI/status/2025997928242811253 Anthropic's agent analysis: https://x.com/anthropicai/status/2024210053369385192 Claude Code Security: https://x.com/claudeai/status/2024907535145468326 Prompt auto-caching: https://x.com/RLanceMartin/status/2024573404888911886 Stripe minions Part 2: https://stripe.dev/blog/minions-stripes-one-shot-end-to-end-coding-agents-part-2 Code Mode for MCP: https://blog.cloudflare.com/code-mode-mcp/ Amp's statement: https://ampcode.com/news/the-coding-agent-is-dead Skills benchmark: https://x.com/forloopcodes/status/2024819596809949341 Trajectory Explorer: https://x.com/benhylak/status/2024546696211083653 Vercel AI Gateway: https://x.com/rauchg/status/2024629864951927222 Taalas: https://x.com/taalas_inc/status/2024516399251456150 Google Pomelli Photoshoot: https://x.com/GoogleLabs/status/2024529795548102667 Superpowers: https://x.com/socialwithaayan/status/2025609619167670501 AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. CHAPTERS 00:00 – Intro 00:25 – Anthropic: Distillation Attacks by DeepSeek, Moonshot, MiniMax 01:36 – Anthropic's Agent Report: 50% Software Engineering, Other Domains Emerging 02:47 – Claude Code Security in Research Preview 03:34 – Prompt Auto-Caching with Claude 03:53 – Stripe Minions: 1,300 PRs/Week 05:39 – Code Mode for MCP (99.9% Token Reduction) 07:29 – Amp: "The Coding Agent Is Dead" 08:36 – Skills Beat Size: Smaller Models with High-Quality Skills 09:39 – Quick Hits: Trajectory Explorer, Vercel AI Gateway, Taalas, Google Photoshoot 12:00 – GitHub Star Party: Superpowers Framework

Shane and Abhi welcome Charlie Holtz from Conductor to AI Agents Hour. Charlie shares how frustration with managing multiple Claude Code instances led to building Conductor. They discuss Conductor's July 2025 launch as the first agent orchestration Mac app, early design choices, and its impact on the market. Charlie outlines a future in which agents run 10x longer and require less supervision, along with new tools such as Kanban boards and a CEO dashboard. The conversation covers multi-agent coordination, designing for both humans and AI, and the importance of user feedback. Watch a live demo showcasing Conductor's parallel agents, code review, GitHub integration, and continuous deployment with GitHub Actions and Vercel. They also explore ways to keep users within the app while allowing flexibility, code-editing strategies, and upcoming remote workstation support. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 🧰 RESOURCES MENTIONED https://www.conductor.build/ 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 Origin Story of Conductor 01:07 Launch and Early Positioning in the Orchestration Space 01:50 The Future 02:46 Autonomous Work and Multi-Agent Coordination 03:39 Designing for Both Humans and AI 04:03 User Education and Feedback Loop 04:48 Live Demo 06:17 Multi-Tasking and Workspace Management 07:31 Code Review and Pull Requests 09:21 Integration and Continuous Deployment Features 10:00 Keeping Users Within the Application 10:35 Editing Code: Directly or with AI? 11:44 Current Integrations and Future Plans

It's time for another AI News roundup with Shane and Abhi! This week was absolutely massive. Matt Shumer's viral article about AI automation, which describes his own job being automated in real time, has reached 84 million views. Anthropic raised $30 billion at a $380B valuation (one of the largest private raises in tech history). Claude Sonnet 4.6 launched with a 1M token context window. And the Chinese model tsunami is real: Qwen 3.5, GLM 5.0, MiniMax M2.5 (nearly Opus-level at 1/8 the cost), and DeepSeek v4 rumors. We also cover Google's Gemini 3.1 Pro, WebMCP native browser support in Chrome 146, OpenAI's agent workflow tips, Peter Steinberger joining OpenAI, Lyria 3 music generation, Kimi Claw integration, and the explosion of agent platforms (Oz, Browserbase Functions, and a former GitHub CEO raising $60M for "version control for agents"). Plus: Real production stats from Ramp (57% of merged PRs from background agents), Letta's context repositories, and why Mastra's Observational Memory is becoming the reference implementation for agent memory. 🧰 RESOURCES MENTIONED Matt Shumer's article: https://x.com/mattshumer_/status/2021256989876109403 Anthropic $30B raise: https://x.com/anthropicai/status/2022023155423002867 Claude Sonnet 4.6: https://x.com/claudeai/status/2023817132581208353 OpenAI agent tips: https://x.com/openaidevs/status/2021725246244671606 Gemini 3.1 Pro: https://x.com/OfficialLoganK/status/2024516665740833156 WebMCP in Chrome 146: https://x.com/chromiumdev/status/2022363079976034455 Ramp production stats: https://x.com/rahulgs/status/2020984194038628832 AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. CHAPTERS 00:00 – Intro 00:17 – Something Big Is Happening 02:34 – Anthropic's $30B Funding Round 03:14 – Claude Sonnet 4.6 Release 04:19 – GPT-5.3-Spark 06:42 – OpenAI's Tips for Long-Running Agent Workflows 07:20 – Peter Steinberger (OpenClaw) Joins OpenAI  09:06 – Google Gemini 3.1 Pro Launch 10:21 – WebMCP in Chrome 146: Native Agent Support 12:39 – Lyria 3: AI Music Generation 14:00 – Qwen 3.5 14:16 – GLM 5.0 & MiniMax M2.5 Released 15:48 – DeepSeek v4 Rumors: 10-40x Lower Inference Costs 16:15 – Kimi Claw 17:01 – Oz: Orchestrate Agents in the Cloud 17:10 – Browserbase Functions: Lambda with a Browser 17:41 – Intent 17:41 – Former GitHub CEO's $60M seed round 20:11 – Lab 20:43 – Ramp's 57% PR Stat 20:52 – Letta Context Repositories & Memory Subagents 21:10 – More Observational Memory 22:20 – Markdown from any website 22:34 – Tavily joins Nebius 22:53 – Ghost, the first database designed for agents 23:21 – Outro

Tyler Barnes, founding engineer at Mastra, introduces Observational Memory. It is a new memory system for AI agents that achieves state-of-the-art results on LongMemEval with a completely stable context window. Unlike semantic recall (which uses RAG and invalidates prompt caching), Observational Memory compresses conversations into dense observations while maintaining a stable, fully cacheable context. The result: 94.87% accuracy on LongMemEval with GPT-5 mini. This is the highest score recorded by any memory system to date. In this conversation, Tyler explains how the system works, why it outperforms raw context, and how you can integrate it into your agents in under 20 minutes. We also dive into the research, the benchmarks, and what's next for Observational Memory. 🧰 RESOURCES MENTIONED Observational Memory Launch Blog: https://mastra.ai/blog/observational-memory Full Research Breakdown: https://mastra.ai/research/observational-memory Tyler Barnes on X: https://x.com/tylbar Tyler's Announcement Post (Feb 9 ): https://x.com/tylbar/status/2020924183979397512 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process. 00:00 – Intro 00:26 – The Origin Story 01:14 – Previous Memory Systems: Semantic Recall vs Working Memory 02:23 – How Observational Memory Works 03:52 – Human-Inspired Memory System 06:11 – Buffered Observations 06:32 – Research & Benchmarks 10:34 – Live Demo 13:57 – No More Compaction Hell 15:08 – Performance & Cost Benefits 16:42 – Shipped Code vs Research Papers 17:33 – Future Roadmap & Next Ideas

Shane and Abhi cover top AI stories. This week was absolutely massive! Anthropic aired Super Bowl ads mocking OpenAI's decision to put ads in ChatGPT, Opus 4.6, and GPT-5.3-Codex launched within 15 minutes of each other, and then ClawHub dropped a bombshell: 11.9% of the entire marketplace is malware. We cover everything: Anthropic's competitive jabs, the model war benchmarks, Claude's 1M token context, OpenAI's Frontier platform, the security crisis that's reshaping how people think about agent marketplaces, Kimi K2.5's domination on OpenRouter, ElevenLabs' $500M raise at $11B valuation, and the explosion of AI video generation tools. Plus: Perplexity's Model Council, Roblox 4D generation, Mistral's Voxtral Transcribe 2, and why Swyx finally admits evals actually help. AI Agents Hour is a weekly livestream hosted by Mastra CPO Shane Thomas and CTO Abhi Aiyer. Airing Mondays at 12PM Pacific on YouTube and X, the show covers breaking AI news, agent development techniques, and features interviews with industry experts building AI applications today. 🧰 RESOURCES MENTIONED Anthropic Super Bowl ads: https://x.com/tomwarren/status/2019039874771550516 Opus 4.6 launch: https://x.com/claudeai/status/2019467372609040752 GPT-5.3-Codex launch: https://x.com/sama/status/2019474754529321247 ClawHub malware report: https://x.com/aakashgupta/status/2019885557221061026 Kimi K2.5 #1 on OpenRouter: https://x.com/openrouterai/status/2020199664746971195 ElevenLabs $500M raise: https://x.com/elevenlabsio/status/2019048384011854087 Perplexity Model Council: https://x.com/perplexity_ai/status/2019444886114824219 📚 MASTRA RESOURCES Mastra: https://mastra.ai Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course Principles of Building AI Agents (Book): https://mastra.ai/book Patterns for Building AI Agents (New Book): https://mastra.ai/blog/patterns-book https://docs.google.com/forms/d/e/1FAIpQLSduJjc515f6RZJqtkR2ByqJZrB0iP8B7SUKnjjZE9IajH_I8w/viewform   MASTRA? Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you’re a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process.

Shane and Abhi welcome Kiet Ho, cofounder of Superset, to discuss how Superset evolved from simple WorkTree management into a full-featured tool with file editing, automation, cloud support, and multi-agent orchestration.

Shane and Abhi cover this week's top AI developments. The OpenClaw agent goes viral amid growing security concerns over its shifting identity. Moonshot AI's Kimi K2.5 ties for first place on Design Arena. Google DeepMind unveils Project Genie, advancing world model technology. Anthropic previews a new model and explores AI coding with Claude Cowork plugins. Plus, updates on developer tools, humanoid robots, and emerging AI models.

Today we discuss the Mastra 1.0 launch, we talk about the things we are learning, we discuss ClawdBot (renamed to MoltBot and now OpenClaw). We also discuss agent skills, new models, and how agents are reading more docs pages than humans.

Today we announce Mastra is officially 1.0! We talk with Kyle from ElectricSQL (and formerly Gatsby) about Durable Streams. Allie joins us to discuss the code review security problem that can be caused by generated AI code. Finally, we cover all the AI news from the last week including ads in ChatGPT, the Langfuse/Clickhouse acquisition, and the OpenAI / Cerebres partnership.

Today we discuss a rising star, we talk about the emerging agent primitives, and we cover downstream effects of a world where agents are now the primary users of a lot of products/tools. We cover Anthropic news, the Google+Apple partnership, model companies going public, AI in healthcare, and much more.

Who is Ralph Wiggum and why are developers talking about him? We also discuss the rise of opencode and if everyone can be a developer now. And of course, we cover all the latest in AI news.

Today we talk about how Abhi tried to rebuild Git over the holiday weekend. We also review everything that happened in the world of AI and agents in 2025. Finally, we give our 2026 predictions for what we think will happen next year.

Are Agent Skills going to be the new MCP? Today we do a deep dive discussion on Prompt Caching as well as compression. We cover all the latest AI news and finally have a guest from Morning Brew join the show.

Today we talk about GPT 5.2, Claude Code updates, and all the other AI news. On top of that we cover the OWASP agentic top 10 with Allie and have guests from Roark and Foundry join the show.

Today we talk with some guests (Alex from Mastra & Laurie from Arize) about evals and observability. We chat with Lio and Kevin from Defang about deploying agents to AWS/GCP/Azure. And of course we cover all the latest AI news

We talk with a special guest about running parallel CLI coding Agents. Abhi goes into depth on slash commands he is using as well as a general Opus 4.5 review. There are a ton of new models (including some new video models) that we will chat about!

It's a showdown between Opus 4.5 and Gemini 3! Today we discuss our first impressions, what we are hearing, and how the new models stack up on the benchmarks. In the future, will there agents in every slack? How is the Mastra v1 coming along? What other AI news is going on? Abhi and Shane talk with others from the Mastra team to find out! We also chat with Tony from Runloop.

Today we finally talk about all the Chinese LLM companies that you should know about. We preview the Great MCP Debate with Allie, and discuss all the AI news.

Today we talk about MCP and about all the recent online conversations around how much it sucks. We discuss the TypeScriptAI Conf, have Ismail from Superagent join us, and do all the normal AI news.

Is TypeScript the new King of AI? Today we talk about Mastra Agent Studio, we discuss the AI news (including a bunch of new models), chat with Atai from CopilotKit, and then preview the TSAI Conf (sold out but you can attend virtually, go to tsconf.ai to register)

Today we have 4 guests joining in! We have Andre from Databricks/Neon, Erik from CodeRabbit, Gaurab from MongoDB, and Professor Andy from Osmosis. We also highlight some of the recent AI news from the last few days. This includes model provider updates, Vercel workflows, and even some AI music.

Today we have a mega AI news day with a ton of new things to talk about. We talk veo 3.1, vibe coding with Gemini, Claude skills & memory, OpenAI Atlas, and much more. We also have security corner with Allie. Last but not least we discuss the spies that may live among us...

Today we have guests from Superagent and Ragie. We highlight the Mastra fundraise, talk about Lovable, and give our take on AI music. Finally we answer the question... is Abhi a Vibe Engineer?

OpenAI Dev Day Recap, AI news, Sean from Netlify, and Daniel from Weaviate!

AI Engineer Paris Recap, Mastra Model Routing, and all the crazy AI news from the last two weeks (such as the Sonnet 4.5 and Sora 2 releases)

We are in Paris at the AI Engineer conference. Learn a bit about what we are seeing on the ground floor and hear from some of the people at the event.

Special EU edition stream! Agent 3, Mastra Templates, Security Issues, AI News

There was a lot of hate for Evals last week... so should you be using Evals in your AI project? We talk about a massive JavaScript supply chain attack that impacted 18 core NPM packages. We chat with Corbin from Artifact to learn how AI is being used to build an IDE for electrical engineers. Finally, we cover all the other AI news with discussions around AI safety, parenting in the age of AI, AI legal settlements, some stealth models and much more.

The AI world is going bananas over Nano Banana, so we start by discussing Google's new update. We have a few guests including Manuj from Waffle AI and Akeil who is building an MCP Course Platform. Finally, we cover all the other AI news from the last week including a deep discussion on the AI "axis of evil."

A recent report says that 95% of generative AI pilots are failing! We have a few guests including Richard from Naptha AI and Sherwood Callaway. We talk about the latest Mastra updates as well as cover all the AI news from the last week.

Today we review more Mastra Templates Hackathon submissions and give out the rest of the awards! We have guests from Cedar and Vapi. We talk Security with Allie and go into all the latest AI news. We discuss updates from OpenAI, Anthropic, Google, Microsoft, Meta, and xAI (including the Elon/Sam drama)

Today we review some Mastra Templates Hackathon submissions and give out some awards! We will discuss the OpenAI launches last week (GPT-5 and OSS models) and provide some hot takes. Finally, we chat about all the other AI news.

Today we check in on the Mastra Templates Hackathon, a few of the judges will stop by (Smithery, Confident AI)! We chat with Oso security and, as always, we discuss recent AI news.