Blog

Claude dominates agent workflows, OpenAI focuses on super-apps and open source

Wednesday, May 13, 2026

Hello, this weekly newsletter guides you through the most important new videos from a curated selection of AI and coding YouTube channels. Each video gets a compact summary, plus a daily overview of the dominant topics. Simply click the link below the summary if you’re interested.

Brian Casel (3 new videos)

• How I build agents that work the night shift
  12.5.2026, 12:01:07

  The video demonstrates how to use AI agents as a “night shift” to automatically handle recurring tasks while you sleep or focus on other activities. The approach consists of three parts: a shared interface (such as a Markdown file or custom app), human feedback in short, focused sessions, and an agent with a skill running on a defined schedule. The speaker shares two concrete examples from his own business: an agent that checks and improves SEO meta tags on his website, and another that reviews pull requests in his open-source projects on GitHub. The process allows him to review the agents’ work at short intervals and provide feedback without spending all his time in a chat window. The speaker emphasizes the importance of designing systems upfront and treating agents as team members who work independently.

  The speaker uses Claude and OpenAI, with a focus on Claude, and addresses both beginners and advanced users who want to integrate AI agents into their workflows.

• Why You Need Claude Code Server Mode?
  6.5.2026, 14:01:40

  The video shows how Claude Code can use locally installed skills that wouldn’t be available in cloud-based Claude Code. The user demonstrates that locally installed skills appear in the list of available skills because they exist on the local file system. This enables tasks that require access to local files and directories, making the local version of Claude Code more flexible. The main advantage is that local resources can be leveraged directly, which isn’t possible in a cloud-based environment.

  **Final note:** The video explicitly focuses on Claude Code and is aimed at intermediate users.

• Why Every AI Coding Tool is Converging on Plan Mode?
  1.5.2026, 14:01:11

  The video demonstrates how three different tools (Claude Code, Cursor, and Codex) are used for plan-driven development. The author starts with a basic PRD (Product Requirements Document) for a customer portal and shows how this document is used in each of the three tools to create an implementation plan. Claude Code automatically recognizes plan mode, while Cursor and Codex activate plan mode via the Shift+Tab keyboard shortcut. The author emphasizes the convergence of tools toward plan-driven workflows, which facilitates standardization and simplifies development.

  The video explicitly focuses on Claude Code, Cursor, and Codex and is aimed at intermediate and advanced users.

Niklas Steenfatt

No new videos in this period.

Fireship (3 new videos)

• Every operating system concept in one video…
  7.5.2026, 17:32:34

  The video explains in detail how an operating system works from the moment the power button is pressed until shutdown. It starts with the bootloader, which loads the operating system, then moves on to privilege rings, which separate kernel and application permissions. Virtual Memory is described as a system that allows multiple applications to run in parallel without interfering with each other. The kernel sets up the file system, loads device drivers, and enables interrupts, which allow the system to respond to input. The kernel then starts the first process (PID1), which is the ancestor of all other processes. System calls enable applications to communicate with the kernel, and the scheduler manages CPU time for many processes. Threads allow applications to execute multiple tasks simultaneously, and inter-process communication (IPC) allows different processes to communicate safely. Finally, the shutdown process is described, where all processes are terminated and the system shuts down safely.

  The video generally covers operating systems and their components without naming specific tools or vendors, and is better suited for intermediate to advanced viewers.

• 732 bytes of Python just borked every Linux machine on earth…
  4.5.2026, 18:40:40

  The video covers a critical security vulnerability in the Linux kernel, referred to as “copy fail” (CVE-2023-31431), which has existed since 2017 and was discovered by an AI tool. The vulnerability allows a local user to gain root access by writing four bytes to the page cache of a read-only file. All Linux distributions updated after 2017 are affected. The vulnerability was exploited through a Python script that uses the ONC ESN protocol and the AF_AGL interface. Although the vulnerability is not remotely exploitable, it is strongly recommended to update systems. The video also mentions the role of AI in discovering security vulnerabilities and promotes Code Rabbit, an AI tool for improving code quality.

  The video explicitly covers AI tools such as the AI agent tool used by Theori and Code Rabbit, and is intended for intermediate to advanced users.

• GitHub is having some major issues right now…
  30.4.2026, 16:37:31

  The video “It’s 10 p.m. Do you know where your children are?” from The Code Report addresses the current problems and challenges facing GitHub, the leading platform for software development. It notes that GitHub has recently experienced massive reliability issues, including frequent outages, missing pull requests, and non-functional search features. The video creator attributes this to increasing strain from AI agents using the platform as a free resource. An example cited is Mitchell Hashimoto, creator of tools like Vagrant and Terraform, who is removing his open-source projects from GitHub after struggling with outages for months. The video also mentions alternatives such as GitLab, Codeberg, and Source Hut, which could serve as potential fallback options. In conclusion, it emphasizes that despite GitHub’s problems, developers still have many options for hosting their projects.

  The video covers GitHub and Microsoft, and is intended for intermediate to advanced developers.

Alex Finn (9 new videos)

• LIVE: The greatest Claude Code workflow ever
  13.5.2026, 20:12:33

  The video demonstrates a detailed, advanced workflow for Claude Code that integrates various tools such as Slack, Linear, GitHub, and Claude Code. The host explains how these tools work together to increase productivity, track changes, and organize development. The workflow includes creating tasks and projects in Linear, linking them with GitHub for branch management, and using Claude Code for automating and managing these processes. The host emphasizes the benefits of this workflow, including increased speed, better error prevention, and improved traceability.

  Additionally, the host discusses personal experiences and challenges, such as dealing with difficult times and the importance of perseverance. He also shares his thoughts on using AI tools like Claude Code and Codex, as well as their differences and use cases.

  The video is intended for advanced users who already have experience with Claude Code and similar tools and want to optimize their workflows. Specific tools such as Claude, OpenAI, and Linear are discussed.

• Hermes Agent powered by local models on the DGX Spark is basically magic
  13.5.2026, 13:30:07

  The video shows how to set up a Hermes Agent on a local model on an Nvidia DGX Spark to create a 24/7 available AI employee. The process includes setting up the DGX Spark in headless mode, installing a local model (Quen 3.6 27B), and integrating the model into the Hermes Agent. The creator demonstrates three use cases: a daily report on AI stocks for beginners, repurposing YouTube video content for intermediate users, and vibe-coding a to-do list app for advanced users. The focus is on the benefits of local models, such as being free (aside from electricity costs), privacy, customization, and educational value.

  Final comment: The video explicitly addresses Nvidia DGX Spark, Hermes Agent, Quen 3.6 27B, and Tail Scale, and is intended for intermediate to advanced users.

• LIVE: Talking AI news (no Hermes use cases ignore the thumbnail)
  11.5.2026, 20:11:47

  The video is a live-stream session that primarily revolves around discussing Hermes Agent and its use cases. The host, Alex Finn, begins with an introduction to Hermes Agent and emphasizes the importance of use cases for utilizing the technology. He mentions that Hermes has published a new website with hundreds of use cases that he and the viewers will go through and test.

  However, a large portion of the video is taken up by various tangents and discussions, including:

  1. **Investment Corner**: Alex discusses current investment opportunities in the AI industry, particularly in companies such as Nvidia, Micron, TSMC, and Tesla. He emphasizes the importance of current AI developments and the need to invest in the right companies.

  2. **Personal Stories and Anecdotes**: Alex shares personal stories and anecdotes that are often humorous and entertaining but not directly related to the main topic of the video.

  3. **Interaction with the Chat**: A large portion of the video consists of interaction with the live chat, where Alex answers questions, makes comments, and engages with viewers.

  4. **Use Cases for Hermes Agent**: Toward the end of the video, Alex begins discussing some of the use cases from the new Hermes website. He mentions use cases such as creating research reports, managing tasks, and automating processes.

  5. **Announcements and Updates**: Alex provides updates on his own projects and announcements, such as launching a second YouTube channel and attending a Baby Keem concert.

  **Final Comment**: The video explicitly addresses open-source AI models and tools such as Hermes Agent and OpenClaw. It is more suited for intermediate to advanced users who already have some understanding of AI and its applications.

• Hermes Agent is blowing me away…
  9.5.2026, 20:54:26

  The video compares the AI agents Hermes and OpenClaw and recommends Hermes due to reliability, self-improvement, and user-friendliness. The author describes the advantages of Hermes, including regular, thematic updates, a self-improving capability through usage, and a strong emphasis on experimentation and local models. The installation of Hermes is described as simple, with options for different models and communication services, with Telegram and Opus recommended. The author shows three use cases: one for beginners that discovers new AI tools daily, one for intermediate users that performs daily proactive check-ins, and one for experts that creates AI-generated videos. At the end, the author emphasizes the importance of brain-dumping and reverse-prompting to use the AI agent personally and effectively.

  The video explicitly addresses the AI tools Hermes Agent and OpenClaw and is more suited for intermediate and advanced users.

• LIVE: Anthropic and Elon just teamed up to take down OpenAI
  6.5.2026, 20:12:34

  The video covers the strategic alliance between Anthropic and Elon Musk’s XAI (X.AI), characterized by a major compute deal. Anthropic gains access to SpaceX’s Colossus-1 cluster, which will significantly enhance their ability to develop and train AI models. This partnership marks a turning point in competition with OpenAI, which has held a dominant position in recent months with Codex. Anthropic has struggled during this time with reduced limits and less powerful models, which will now be addressed through the new alliance. Elon Musk, who was previously critical of Anthropic, is now providing massive computing resources, which changes the dynamics of AI competition. The video also discusses Elon Musk’s long-term strategies, which may focus on larger goals such as autonomous vehicles, space exploration, and robotics rather than remaining in the AI chatbot competition. The alliance could lead to a new era of innovation and improvement in AI tools, from which consumers will benefit. The video emphasizes the importance of using both leading AI tools, Claude Code and Codex, to benefit from their respective strengths.

  **AI Tools/Models/Providers:** Anthropic, OpenAI, Elon Musk (X.AI), Claude, Codex, Grok, Gemini, Open-Source
  **Target Audience:** Intermediate

• Hermes Agent might have just killed OpenClaw
  5.5.2026, 21:11:59

  The video presents Hermes Agent as a more reliable alternative to OpenClaw and discusses seven new features that improve productivity and user-friendliness. These include:

  1. **Kanban Board**: Enables multitasking through parallel processing of multiple task threads. A manager agent fills tasks with details and moves them through various statuses (Triage, To-Do, Ready, In Progress, Block, Done).

  2. **Slashgo**: A high-level mission function that assigns long-term tasks to the agent that can be worked on over an extended period. The quality of the prompt is crucial for good results.

  3. **Profiles (Multi-Agents)**: Allows creating multiple agents with their own memories and abilities to optimize performance and avoid overload.

  4. **Model Catalog**: Simplifies switching and assigning models to specific tasks, improving cost control and efficiency.

  5. **Compression**: By adjusting the compression threshold to 0.5, less drastic compressions are performed, improving memory capability.

  6. **Curator Feature**: Automatically prunes rarely used skills every seven days to reduce bloat and maintain performance.

  The video criticizes OpenClaw for frequent updates that lead to instability and performance issues, and highlights Hermes’ targeted, reliable updates. It is recommended to use Hermes’ new features to boost productivity.

  **Final Comment**: The video explicitly addresses Hermes Agent and OpenClaw and is more suited for intermediate to advanced users.

• LIVE: Is Hermes better than OpenClaw? FINALE!!!
  4.5.2026, 21:53:53

  The YouTube video shows a live stream in which the host tests various AI agents (OpenClaw and Hermes) in a competition called “Agent Olympics.” The stream is unusually long (3.5 hours) and is divided into various sections ranging from technical tests to personal discussions to spontaneous decisions.

  **Content Summary:**
  1. **Agent Olympics:**
  – The host tests four combinations of AI agents (OpenClaw and Hermes with different backend models such as ChatGPT and Opus) in five different tasks.
  – The tasks include creating infographics, animated music videos, and other complex assignments.
  – The results are evaluated live, with OpenClaw with Opus ultimately emerging as the winner.

  2. **Technical Discussions:**
  – There are extensive discussions about the stability and reliability of the various AI agents, particularly Hermes, which is criticized for “compaction” errors (loss of working state).
  – OpenClaw is praised for its consistency and user-friendliness.

  3. **Personal Topics:**
  – The host talks about his sleep issues and experiments with various solutions such as kiwis and magnesium.
  – There are discussions about work methods, including the use of treadmills and standing desks, with the host expressing his preferences and dislikes.

  4. **Community Interaction:**
  – Viewers are actively engaged in the chat, asking questions and providing feedback.
  – The host spontaneously decides to create a second YouTube channel called “Alex Finn Labs,” which leads to an entertaining interaction with a viewer who has already reserved the desired channel name.

  5. **Announcements and Future Plans:**
  – The host announces plans to publish more videos on Hermes and multi-agent setups in the future.
  – There is a discussion about whether live streams should take place at later times to reach a broader audience.

  **Final Comment:**
  The video explicitly addresses the AI tools OpenClaw, Hermes, ChatGPT, and Opus. It is more suited for intermediate and advanced users, as it covers technical details and advanced applications of AI agents.

• LIVE: OpenClaw vs Hermes Agent: The ultimate showdown
  1.5.2026, 20:48:54

  The YouTube video shows a live stream in which various AI agents are tested in direct comparison. The main participants are OpenClaw and Hermes, each running with the models ChatGPT and Opus. The stream is divided into several tests that evaluate the capabilities of the agents in various task areas.

  1. **Test 1: Real-time Stock Dashboard**
  – **OpenClaw with ChatGPT**: Fastest completion, but with an unattractive user interface (UI). Functionality was rated as solid.
  – **Hermes with ChatGPT**: Slower and crashed the computer, resulting in a poor rating.
  – **OpenClaw with Opus**: Slower than the ChatGPT version, but with a slightly better UI and additional features such as TradingView integration.
  – **Hermes with Opus**: Best UI and functionality, but slower than OpenClaw with ChatGPT.

  2. **Test 2: Game Development**
  – **OpenClaw with ChatGPT**: Fast, but unplayable game.
  – **OpenClaw with Opus**: Playable, but not particularly entertaining.
  – **Hermes with ChatGPT**: Unplayable and graphically poor.
  – **Hermes with Opus**: Best graphics and playability, rated as actually entertaining.

  3. **Test 3: Website Rebuild (Apple.com)**
  – **Hermes with Opus**: First to complete, but below-average accuracy.
  – **OpenClaw with Opus**: Better than Hermes with Opus, but not perfect.
  – **OpenClaw with ChatGPT**: Most accurate, nearly perfect.
  – **Hermes with ChatGPT**: Perfect recreation through screenshots, but ethically questionable.

  The stream ends with Hermes with Opus in the lead, followed by OpenClaw with Opus and OpenClaw with ChatGPT. Hermes with ChatGPT is significantly behind. The remaining tests will continue in the next live stream.

  **Final Comment**: The video explicitly addresses the AI models Claude (Opus), OpenAI (ChatGPT), and specific tools such as OpenClaw and Hermes. It is more suited for intermediate and advanced users interested in the performance and comparison of AI agents.

• ChatGPT 5.5 Codex is the greatest AI coding tool ever. Here's how to use it
  30.4.2026, 17:54:13

  The video demonstrates how to build a complete stock investment app from scratch using ChatGPT 5.5 Codex. The creator starts by building the user interface (UI) using the integrated Image Gen 2 model, which generates five UI options. The app is then developed with a Convex database backend, implementing various features such as live data queries via APIs and portfolio tracking. The creator uses Codex’s multitasking capabilities to simultaneously work on app development, marketing video creation with the Remotion skill, and stock research. Additionally, automation features are presented that regularly perform code quality checks. The video ends with the challenge of further developing the app and launching it live.

  The video explicitly addresses ChatGPT 5.5 Codex and is more suited for intermediate to advanced users.

Leon van Zyl (8 new videos)

• Codex CLI Tutorial: Build an AI Image Studio from Scratch
  11.5.2026, 11:17:21

  This video shows how to build an AI image studio step by step for creating YouTube thumbnails, posters, banners, and other graphic content. The process begins with setting up a Next.js project using the Codec CLI tool, which is based on GPT-5.5 and a high reasoning level. A database is set up with Docker and Postgres, and the necessary tables for user authentication are migrated.

  The focus is on creating a user-friendly interface that allows users to upload reference images, write prompts, and generate images. The developer uses the Codec tool’s capabilities to design and test the user interface while adhering to a predefined design system. After designing the interface, the actual functionality is implemented using the OpenAI API key for the GPT-4 Image model. The developer shows how to upload reference images and assets and how to generate thumbnails that combine these elements.

  At the end, the user interface is adjusted to ensure that image generation only happens through the dashboard and not from the homepage. The homepage is revamped with a generated image and marketing text. The video emphasizes the efficiency and time savings through using Codec and OpenAI, though it points out the token limitations of the ChatGPT Plus plan.

  The video explicitly covers OpenAI (GPT-5.5, GPT-4 Image model) and Codec. It’s designed more for intermediate and advanced users since it covers advanced concepts like Docker, Postgres, Next.js, and API integration.

• Create Custom OpenCode Agents #Shorts #OpenCode #AICoding
  10.5.2026, 07:00:16

  This video shows how to create custom agents in OpenCode. By default, there are two agents: “build” and “plan”. To create your own agent, run the `open code agent create` command in the terminal. Next, provide a description of the agent, for example that an agent named John only responds with emojis. After creation, you can use the spacebar to define which functions or tools the agent has access to. You also select the agent mode: either for primary and subordinate roles or only as a subordinate agent. The video demonstrates creating a subordinate agent named John and shows how the main agent delegates a task to John. You can monitor the subordinate agent’s work and see its reasoning and outputs. In conclusion, it’s emphasized that this is just a demonstration and in practice you could use more specific system prompts and tool access for subordinate agents.

  The video covers OpenCode and is suitable for intermediate users.

• OpenCode’s Best Hidden Feature #Shorts #OpenCode #AICoding
  9.5.2026, 07:00:19

  This video explains how to set different models for different modes in open-source code. For example, you can use a powerful model like GPT-5 for planning mode and a fast, cost-effective model like Big Pickle for implementation mode (Bold Mode). Alternatively, you can use GPT-5.5 for planning, but reduce effort by selecting the “/variants” command with the “low effort” option. This way, planning is done by an intelligent model, while implementation can be handled by a less powerful but faster model.

  The video covers open-source models and is more suitable for intermediate users.

• I Turned Hermes Agent Into a Coding Agent
  8.5.2026, 11:02:33

  This video shows how to use the Hermes Agent as a coding agent to create and deploy a web app online. The process includes setting up Hermes Agent on a VPS, integrating it with Telegram for communication, installing the Vercel CLI tool for deployment, and configuring the necessary skills for the agent. The creator tests whether the agent can create a personal portfolio page by scraping information from the creator’s YouTube channel and creating an appealing frontend design. The agent successfully creates the app, deploys it to Vercel, and provides a public URL that opens the app in a browser. The video also shows that the agent can make changes to the app and deploy them automatically.

  The creator concludes that Hermes as a coding agent is suitable for simple tasks and quick dashboards, but not for complex software projects. The video explicitly covers Hermes Agent, OpenAI Codex, GPT 5.5, Vercel, and Telegram. It’s designed more for intermediate and advanced users.

• This free OpenCode trick saves thousands #opencode #aitools #hacks
  7.5.2026, 13:38:39

  This video explains how to use free AI models for code creation in OpenCode. First, run the “connect” command and search for “OpenCode Zen” under providers. Then generate an API key through a provided URL, which is free. After entering the API key, you get a list of supported models, including Big Pickle, HY3, Minimax M2.5, and Nvidia’s Neurotron 3 Super. These models are powerful and completely free.

  The video covers OpenCode and OpenCode Zen and is aimed at intermediate users.

• I Built a Coding Agent That Runs Locally for Free
  6.5.2026, 12:25:00

  This video introduces the open-source tool “Honeyfree”, which allows you to autonomously plan and implement software projects. You describe to the tool what you want to build, and it plans the features, adds them to a Kanban board, and implements them automatically. The tool supports various models like Alum Studio and Ollama and can also break complex tasks into smaller features. The user demonstrates creating a simple to-do app and shows how new features can be added and implemented. The video emphasizes that this is now possible with free models, which wasn’t the case a few months ago. It also explains how to download models like Qwen 3.6 or JML4 and use them in Alum Studio or Llama Studio. The user recommends increasing the context window length of models to at least 64,000 tokens for better performance. The video also shows how to install and set up Local Forge to create and manage projects. It emphasizes that while free models are good at writing code, they need more detailed instructions for better results. The user recommends using a paid model like Claude to plan features, while using free models for actual implementation. The video ends with an invitation to sign up for a masterclass course teaching how to build applications with coding agents.

  The video covers open-source models like Qwen 3.6 and JML4, as well as tools like Alum Studio, Llama Studio, and Local Forge. It’s designed more for intermediate and advanced users who already have experience using AI models and developing software.

• OpenCode Tutorial for Beginners: Setup, Agents, Skills & MCP
  5.5.2026, 12:33:17

  This video is a tutorial showing how to create a Next.js application with Open Code, an open-source AI tool. The process begins with installing and setting up Open Code, including connecting with various AI models and providers, both free and paid. The tutorial shows how to add agent capabilities like frontend design and Next.js skills to improve the quality of generated code. It also demonstrates how to use memory files and design systems to increase agent consistency and efficiency. The tutorial continues with creating an application that allows users to input a rough idea of their app and receive a detailed project plan. The agent uses subagents to run tasks in parallel and protect the main context. At the end, the application is tested and improved, with the agent completely revamping the UI and running automated tests. The video is suitable for intermediate and advanced users interested in AI-powered coding tools.

  AI tools/models/providers: Open Code (open-source), OpenAI, Anthropic, Gemini, OpenRouter, BigPikko, HY3, Minimax, Nvidia, Vercel, Cintra AI.

• I Built a Full App Using Only Cursor AI
  4.5.2026, 11:01:35

  In this video, an AI-powered YouTube summarizer is developed using the Cursor tool. The process starts by creating a user interface that takes a YouTube URL and provides a summary of the video. Requirements include a brief summary (TLDR), five to eight key points, a “Watch these moments” section with timestamps and descriptions, and the original video link.

  The creator uses Cursor and chooses the Composer 2 model to scaffold the project. He installs Next.js and the Shad cn library for the user interface. With the help of Cursor’s agents, a basic user interface is created that meets the requirements. Next, functionality is added to retrieve the transcript of a YouTube video using the YouTube Transcript API.

  For the AI-powered summary, the Cursor AI SDK is used to return structured data. The creator chooses the “anthropic/claude-2” model from OpenRouter and integrates the API key into an .env file. The agent then generates the summary, including the TLDR, key points, and recommended moments from the transcript.

  The video explicitly covers the tools Cursor, Composer 2, Next.js, Shad cn, YouTube Transcript API, AI SDK, and OpenRouter. It’s designed more for intermediate and advanced users.

WorldofAI (10 new videos)

• Hermes Agent v2.0! Huge New Updates: WebUI, Qwen 3.6 Plus FREE, Computer Use, & More!
  13.5.2026, 05:08:59

  This video presents the latest updates to Hermes Agent, an open-source AI agent that continuously evolves and can be run on your own infrastructure. The major new features include:

  1. **Computer Use Feature**: Hermes can now control your computer in the background without taking over your PC. Currently available only on macOS, with Windows and Linux support planned. Compatible with various AI models like Claude, GPT-4, Gemini, and local open-source models.

  2. **Integration of Alibaba’s Qwen 3.6 Plus**: This powerful model is now integrated into Hermes, offering long context windows and multimodal capabilities. Available free for a limited time.

  3. **Light Panda Integration**: An integrated, open-source browser backend specifically designed for AI workflows. Provides improved browser automation and reliability.

  4. **Kanban Board**: A new Kanban board enables creation of unlimited boards and projects, orchestration of multiple agents, and task management through a dashboard.

  5. **/goal Command**: A long-term autonomous goal mode that continuously plans, executes, verifies, and retries on errors until the goal is achieved.

  Additional minor updates include voice cloning and multilingual support. The video recommends Tiny Fish for web automation tasks and highlights Hermes’ open-source nature, making it a versatile tool for AI workflows.

  This video specifically covers Hermes Agent, open-source models, and Tiny Fish, and is geared toward intermediate to advanced users.

• Claude Code Agent View IS INSANE! Huge New Update Introduces /goal, sessions, & More!
  12.5.2026, 07:03:39

  This video presents the latest updates to Anthropic Cloud Code, a tool for agent-based development. The key new features include:

  1. **Agent View**: A central interface for managing all Cloud Code sessions. Enables running and managing multiple coding sessions in parallel within a single dashboard. Users can switch between sessions without losing context and execute background tasks.

  2. **/goal Feature**: A function that enables autonomous execution with a defined goal. The agent works independently until the goal is reached, usable in various workflows and on mobile devices.

  3. **/radio and Claude FM**: An integrated lo-fi radio station designed to enhance user experience.

  4. **Improved System Prompt Compaction**: Prompt trimming has been optimized to better preserve sensitive instructions and user intent without showing explicit warnings when shortening context.

  The video discusses the benefits of these updates, particularly for complex, agent-based coding workflows, and mentions the possibility of using Depot CI for a faster and more consistent CI/CD pipeline. However, the video notes that Cloud Code’s limited token usage could restrict practical applicability.

  Final note: This video specifically covers Claude (Anthropic) and Depot CI, and is geared toward intermediate to advanced users.

• Hermes Agentic OS is The Future
  11.5.2026, 07:07:11

  This video presents Hermes Agent, an open-source project from Mistral AI designed as a persistent autonomous system that continuously learns and evolves over time. It can run 24/7 on your own infrastructure, build long-term memories, develop reusable skills, and gain deeper understanding of the user. The author shows how Hermes Agent can be extended with ION UI to become an agentic operating system (AIOS) capable of managing multiple autonomous AI agents performing complex multi-step tasks without constant human oversight. ION UI is a free open-source co-work platform that enables AI agents like Hermes to work directly on the user’s computer, managing files, writing and executing code, browsing the internet, automating workflows, and working alongside the user as an assistant. The video demonstrates how to install and configure Hermes Agent and ION UI, as well as various use cases, such as creating a financial dashboard in Excel, organizing desktop files, and generating a report on rural EV charging infrastructure. The author emphasizes the system’s ability to execute multiple tasks simultaneously and continuously improve, making it a powerful tool for automation and efficiency enhancement.

  This video specifically covers Hermes Agent and ION UI, and is geared toward intermediate and advanced users.

• Hermes Agent NEW Desktop App – The 24/7 Self-Evolving AI Agent!
  10.5.2026, 06:27:40

  This video introduces Hermes Agent, an open-source AI project that differs from tools like OpenClaw and Claude Code through its ability to continuously self-improve and maintain long-term knowledge storage. Hermes can run around the clock on your own devices and evolves through closed-learning loops and persistent storage systems. A key advantage is its ability to autonomously handle complex tasks like video creation with Hyperframe. However, setup was previously heavily CLI-based, limiting user-friendliness. The video introduces the new Hermes Desktop App, which offers an intuitive interface and simplifies managing multiple agents and workflows. The app supports various operating systems and enables integration of API providers like OpenAI and use of local models. It also offers features for managing skills, personas, memory, and tools, plus the ability to create cron jobs and connect with other platforms. The app also enables migration of OpenClaw configurations. Hermes can be applied to diverse use cases, from content creation to financial analysis. The video demonstrates setup and usage of the Desktop App, emphasizing its advantages for user-friendliness and flexibility.

  This video specifically covers Hermes Agent, open-source models, and specific tools like OpenAI, Hyperframe, and OpenClaw, and is geared toward intermediate and advanced users.

• Codex Super App, OpenAI Chaos Drama, Gemini 3.2 Pro In Arena, GPT-Realtime-2, & NotebookLM Update!
  9.5.2026, 07:19:53

  This video provides an overview of the week’s major AI developments. OpenAI announced hints about a future Codex Super App featuring remote control, deeper integrations, and new connectors. GPT Real-Time 2 was also unveiled, a voice model with near-GPT-5 intelligence for real-time interactions. Google is experimenting with new Gemini checkpoints, though these are perceived as less powerful. Claude Code introduced a financial data interface enabling advanced financial analysis and trading strategies. In China, Baidu released Ernie 5.1, achieving better benchmark results than DeepSeek V4 at lower costs. Grock is expanding to a Super App with enhanced tool-calling capabilities and productivity tools. Additionally, private text messages between Sam Altman and the OpenAI board were leaked, providing new insights into power struggles at OpenAI. Finally, the video discusses growing cultural rejection of physical AI systems, the “anti-cyborg” movement.

  This video specifically covers OpenAI, Google Gemini, Claude Code, Baidu, and Grock, and is geared toward intermediate and advanced AI enthusiasts.

• NEW Open Claude Code Is A FULLY FREE AI Coding Agent! (Tutorial)
  8.5.2026, 05:57:40

  This video discusses current issues with Anthropic, particularly aggressive rate limiting and reduced model performance that undermine the Claude Code user experience. The user points out that even with a Pro subscription, usage restrictions are frustrating and costs for additional usage quickly escalate. As an alternative, Freebuff is presented, a free AI coding agent based on GLM 5.1 requiring no subscriptions or complicated setup. Freebuff offers fast, autonomous, and user-friendly coding with integrated sub-agents and intelligent follow-up prompts. The user demonstrates Freebuff installation and usage, including integration with ChatGPT for certain use cases. Freebuff is presented as a promising and free alternative to Claude Code offering faster and more reliable coding experience.

  This video specifically covers Anthropic, Claude Code, Freebuff, GLM 5.1, and ChatGPT, and is geared toward intermediate to advanced users.

• Claude’s New “Infinite” Context Window Model, Doubled Rate Limits, Multi-Agent Coordination, & More!
  7.5.2026, 06:44:06

  This video summarizes the key announcements from the Anthropic developer conference, focusing on AI-coding agents, agent workflows, and the future of software engineering with Claude. A central theme was long-term agent intelligence, with Anthropic introducing a new “dreaming” feature enabling agents to review past sessions and improve future decisions. Additionally, multi-agent orchestration was introduced, where a lead AI agent delegates tasks to specialized agents working in parallel on complex tasks. Another important point was significantly increased Cloud Code rate limits for all paid plans, enabled by a new compute partnership with SpaceX. Anthropic announced three main focus areas for next-generation models: infinite context windows, advanced multi-agent coordination, and persistent long-term reasoning systems. These developments suggest Claude could evolve from a simple chatbot to a fully autonomous software engineering system.

  This video specifically covers Claude by Anthropic and is geared toward intermediate and advanced users.

• Gemini Omni, Gemini 3.2 Flash, a 12M Context Window Model, Claude Replaces Analysts, & More! AI NEWS
  6.5.2026, 06:30:16

  This video provides a comprehensive overview of the latest developments in the AI industry, with particular focus on Google’s upcoming announcements at the Google IO conference. Key points include reported testing of Gemini 3.2 Flash and potentially more powerful variants like Gemini 3.5 or 4.0. Also mentioned is a breakthrough in large language model architecture by SubQ, which introduced a model with a 12 million token context window. OpenAI has released GPT 5.5 Instant, which is faster and more efficient, while Anthropic is expanding its AI solutions for the financial sector. Google has unveiled numerous updates for its AI tools including Gemma 4, Google AI Studio, and Notebook LM. Perplexity also introduced a Finance Agent working with licensed data from various providers.

  Final note: This video specifically covers Google (Gemini, Gemma), OpenAI (GPT 5.5 Instant), Anthropic (Claude), SubQ, and Perplexity, and is geared toward intermediate and advanced users.

• Open Design – Open Source Claude Design! Fully Free AI Design System!
  5.5.2026, 06:52:13

  This video introduces Open Design, an open-source alternative to Claude Design by Anthropic, enabling creation of UI designs, wireframes, interactive prototypes, and presentations through voice commands. Open Design offers several advantages over Claude Design, including the ability to integrate various models and tools, a local-first web deployment option, and compatibility with up to 15 different coding agent CLIs. It features 31 composable skills and 72 complete design systems enabling production-level design creation. The video also demonstrates local installation and setup of Open Design, including configuration of agents, media providers, and MCPs. It shows how to create a blog post design and leverage various Open Design features to create high-quality UI designs. The video emphasizes Open Design’s advantages over Claude Design, particularly flexibility and the ability to use open-source models.

  This video specifically covers open-source tools and is geared toward intermediate and advanced users.

• DeepSeek V4 + Claude Code = BEST AI Coder!
  4.5.2026, 07:30:58

  This video shows how to combine DeepSeek V4 with Cloud Code to create a cost-effective and efficient AI coding workflow. DeepSeek V4 is a powerful open-source model that is particularly token-efficient and supports long context windows. It works well for basic coding tasks like quick scripts, unit tests, and simple automation, but is not suitable for complex tasks like web development or security audits. By combining with Cloud Code, DeepSeek V4 can be used for simple tasks while reserving more expensive models like GPT 5.5 or Opus 4.7 for complex work. The workflow is simplified through Anti-Gravity, which performs setup autonomously. A demo shows how DeepSeek V4 is used for the basic structure of an AI dashboard while Opus 4.7 handles UI polishing and complex tasks. This hybrid approach saves costs and circumvents rate limits.

  This video specifically covers DeepSeek V4, Cloud Code, Opus 4.7, GPT 5.5, and Anti-Gravity, and is geared toward intermediate to advanced users.

Matt Pocock (4 new videos)

• Anthropic’s “dedicated monthly credit” is actually a huge cut
  13.5.2026, 20:12:51

  The video covers an important announcement from Anthropic regarding the use of Claude for programmatic purposes and AFK workflows (Away From Keyboard). Starting June 15, paid Claude subscribers will receive monthly credits for programmatic usage, covering the use of Claude Agent SDK, Claude P, Claude Code, GitHub Actions, and third-party apps. The creator explains that while this change is technically a bonus since additional credits are being provided, for users primarily using AFK workflows, it represents a significant reduction in usage limits. This is because previous subscriptions were often considerably cheaper than API credits. The creator discusses the implications of this change and mentions potentially switching to alternatives like Codeex or OpenAI, as these don’t distinguish between AFK and human-in-the-loop usage. He emphasizes that the clarity of the new rules is positive, even though they’re restrictive for AFK users.

  The video explicitly covers Claude (Anthropic) and alternatives like Codeex and OpenAI, and is aimed at intermediate to advanced users.

• New Skills! /handoff, /prototype, /review and /writing-* | Skills Changelog
  12.5.2026, 09:56:10

  The video showcases the latest updates and improvements to an AI Skills repository that recently reached 70,000 stars on GitHub. The key new features are two new Skills:

  1. **Handoff Skill**: This Skill creates a handoff document that summarizes the current conversation, allowing a new agent to continue the work. The document is stored in a temporary file and contains suggestions for Skills that should be used in the next step. The Skill is particularly useful for switching between different context windows without losing important information. There are two main usage patterns: “Fire and Forget” and “DIY Sub Agent”.

  2. **Prototype Skill**: This Skill helps with creating prototypes, particularly in UI development and creating logic prototypes for stateful applications. The Skill generates multiple UI variants and allows you to test and refine them interactively before implementation.

  Additionally, several bug fixes were introduced, including an improvement to the “Grill with Docs” Skill through the use of XML tags to control information prioritization, and adjustments to the “To PRD” and “To Issues” Skills to use the correct labels.

  There’s also a preview of two Skills in development: one for writing text (Tripart Skill) and one for code reviews. The Tripart Skill helps structure text fragments into a complete story or article, while the Code Review Skill checks for adherence to coding standards and correct implementation of requirements.

  Finally, a documentation website is mentioned that contains videos for each Skill and a newsletter to stay updated on the latest developments.

  The video explicitly covers the use of Claude and is aimed at intermediate to advanced users.

• Burn through the backlog from hell with /triage
  7.5.2026, 15:00:43

  The video introduces a tool called “Triage” specifically developed for processing GitHub Issues and other backlogs. It helps transform unstructured ideas and bug reports into clear, actionable tasks that AFK agents (autonomous AI agents) can process. The tool uses a state machine system with two categories (Bug and Enhancement) and five states (e.g., “needs triage”, “ready for agent”, “won’t fix”) to clearly classify each issue. Users can work through individual issues or the entire backlog to categorize them and directly address them if needed. The video shows a live demo where the creator applies the tool to his own “Sank Castle” repository to triage issues and even fix them directly. The integration with other Skills, such as the “Diagnose” Skill, which automatically reproduces and fixes bugs, is also demonstrated. The tool is designed to facilitate collaboration between human developers and AI agents by providing a clear structure for task management.

  The video explicitly covers the use of Claude and is aimed at intermediate or advanced users already experienced with GitHub, backlog management, and working with AI agents.

• I Open-Sourced My Own AFK Software Factory
  30.4.2026, 13:28:38

  The video demonstrates how to orchestrate AI coding agents in isolated sandboxes using the TypeScript library “Sand Castle” to run them completely autonomously (AFK). The author explains that he tried various solutions but ran into problems with Docker sandboxes. So he developed Sand Castle, which allows controlling agents with simple TypeScript functions. The video demonstrates setting up Sand Castle in a repo, selecting an agent (e.g., Claude Code) and a sandbox (e.g., Docker), and using GitHub Issues as a backlog manager. It shows how to create a Docker image, set required environment variables, and create a GitHub Issue to be processed by the agent. The author explains the structure of the main.ts file, which orchestrates various agents (Planner, Implementer, Reviewer, Merger), and how these agents work in parallel and merge code commits. At the end, it’s shown how the agent creates a TypeScript template with tests and CI scripts and merges the changes into the main branch.

  The video explicitly covers Claude (Anthropic), Docker, and GitHub and is aimed at intermediate or advanced users.

David Shapiro

No new videos in this period.

Automatically generated from the latest YouTube videos of the curated channel selection. For feedback, suggestions, or to unsubscribe: simply reply to this email.