NEW Claude Agentic Operating System is INSANE!
The speaker presents a custom Claude-based agentic operating system that integrates multiple AI tools and agents into a unified dashboard, enabling automated workflows, team coordination, and memory management. The system allows users to combine models like Claude, Hermes, and others while maintaining flexibility to swap components without disrupting workflows.
Summary
The speaker describes building a comprehensive agentic operating system over two months that addresses fragmentation when using multiple AI tools separately. The primary motivation was consolidating disparate tools (terminal, multiple tabs, various AI models) into a cohesive interface with persistent memory and visual organization.
The system features several core components: a Hermes voice agent with real-time voice activation and browser control; Hermes Oracle for pulling latest news and social media content; multiple chat profiles for testing different models (Claude, Qwen, Grok, etc.); and a Mixture of Agents feature that the speaker claims performs comparably to or better than Fable 5 in many scenarios.
A key architectural feature is the 'Agent Mastermind' section, which enables group chat between agents who can brainstorm, read from an Obsidian vault, and automatically populate a pipeline with generated ideas. The system uses a 'Paperclip' organizational structure with CEO and department leaders coordinating sub-agents, enabling team-based AI work that isn't possible within single-model ecosystems.
The system emphasizes flexibility through abstraction—users can swap out models (e.g., replacing Fable 5 with Opus 4.8) in minutes rather than rebuilding entire workflows. Additional features include scheduled loop-based automation (Hermes Oracle refreshes every 24 hours), an automated memory system with thousands of notes that agents populate and retrieve for personalized responses, outreach tools with scheduled email sending, and integration with newly released tools via MCPs.
The speaker notes that every new feature unlocks additional possibilities, creating compound growth in system capabilities. The system was built using Claude itself, with each feature requested as a UI addition to the growing dashboard. The speaker offers this operating system through the 'AI Profit Board' community, which includes 194 pages of user testimonials, daily tutorials, beginner-to-expert courses, and weekly coaching.
Key Insights
- The speaker claims that Hermes Mixture of Agents performs at parity with or better than Fable 5 on many examples despite using older models combined together
- The fundamental problem the speaker solved was the lack of linkage between separate AI tools—when using terminal, different tabs, and Claude simultaneously, there is no shared memory or unified view of created content
- The speaker demonstrates that building a custom agentic OS enables model flexibility where swapping Claude models can be done in approximately 10 seconds by instructing an agent to replace the CLI, versus spending an entire day rebuilding workflows
- The speaker architected the system to run agents on autonomous loops with self-prompting capabilities, eliminating the need for manual prompting or quality control as agents now refresh themselves on schedules and quality-control their own work
- The speaker claims the memory system creates a positive feedback loop where agents automatically populate thousands of notes into the system, then retrieve and use that context to provide personalized responses and update the system with new information
Topics
Transcript
[0:00] Today I'm going to show you what I'm building with this Clawude agent operating system that combines for example Fable 5 uh everything that we've built with it. We've got for example Hermes uh voice agent Hermes Oracle that can pull in the latest news a full outreach tool mixture of agents and everything else that we're doing together and I'm going to show you exactly how this works step by step what we're building with it and the progress we made so far. So, I started building this out about 2 months ago, and it was really just like a kind of fun project where I was like, "Okay, let's try and just build out a basic mission…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Julian Goldie SEO
NEW Nvidia Autonomous AI is WILD!! 🤯
Nvidia announced Nemo Clo, a new autonomous AI agent system that operates independently without continuous prompting. Powered by Nemotron 3 Ultra (a 550 billion parameter model), the system is five times faster and cheaper than previous versions, with OpenShell providing secure sandboxed execution.
Laguna XS 2.1: New FREE + Opensource Local AI!
Julian reviews Laguna XS 2.1, a new free open-source local AI coding model from Poolside that performs comparably to Qwen 3.6 and outperforms Claude Haiku on benchmarks. He demonstrates its practical capabilities by building landing pages and functional apps, highlighting its speed, offline functionality, and multiple deployment options through local setup, Claude Code, or OpenRouter's free API.
How to Run Hermes FREE Forever!
The video demonstrates how to run the Hermes AI agent for free using Gemma 4, a local open-source model from Google, with significant speed improvements through MLX optimization. The setup works on Apple Silicon Macs or via free APIs on Open Router, enabling autonomous agents to work offline and privately without subscription costs.
This NEW Chinese AI is INSANE! (FREE + Open Source!)
Long Cap 2.0 is a new open-source Chinese AI model from a food delivery app company that offers 1 million tokens of free context memory, beats GPT-4.5 on SWE bench pro benchmarks, and uses efficient parameter activation to reduce computational overhead while maintaining high performance.
Claude Code is now FREE: Here’s how…
Google's new Gemma 4 model running on Ollama is 90% faster on Apple Silicon, enabling free Claude Code usage locally without token costs. The setup requires three simple steps: downloading Ollama, Gemma 4, and installing into Claude Code, with alternatives available via OpenRouter API for non-Mac users.