{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"AI News Today | Julian Goldie Podcast","title":"How to Build Your Own Agent Operating System","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/fbd4d9f1\"></iframe>","width":"100%","height":180,"duration":483,"description":"Build a No-Code Agent Operating System (Mission Control for Your AI Agents)\nThe episode shows how to build a non-technical “agent operating system” that acts as a mission control for multiple AI agents and workflows (SEO, video, text-to-speech, images) using tools like Claude, Hermes Agent, and open-source projects such as OpenJarvis. The presenter demonstrates plugging a CLI-based agent into a cleaner UI with features like a Kanban board, shared skills, optional memory via Obsidian, and local model support via Ollama, noting it can run locally or on a VPS. They outline a repeatable “loop method”: find an agent, ask Claude for a build recipe and downsides, order it to build (with bypass permissions), and verify it works—troubleshooting with screenshots if needed—highlighting that installation can take minutes (about six) and can be rolled back.\n00:00 Build Your Agent OS\n00:32 Why Mission Control Wins\n00:46 Pick a GitHub Agent\n00:56 Hermes Mission Control Demo\n01:22 Local vs VPS Setup\n01:43 Ask Claude for the Recipe\n03:05 Install and Configure Jarvis\n03:41 Avoid Overbuilding Workflows\n04:34 New Models Plug In Fast\n05:07 Test and Troubleshoot\n05:52 Rollback and UI Benefits\n06:17 The Loop Method Framework\n07:14 No Code Needed Mindset\n07:35 Get the Agent OS Pack","thumbnail_url":"https://img.transistorcdn.com/Tp0JSUlyiAe2ran44b13cjk4ImQYS1QEMKCAa9Q7go0/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9lNTYz/NGQxMTU2ZTM1MzU5/MDNiYjcwZjJmYTY5/ODJmOS5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}