Obsidian Is the Surface. Here Is Where myPKA Fits.
The speakers clarify the architecture of their myPKA (Personal Knowledge Assistance) system, explaining how it consists of three independent layers: AI LLM services (Claude, ChatGPT), a local folder structure with markdown files, and a front-end interface (such as Obsidian). They emphasize that Obsidian is merely one optional tool for navigating the underlying system, not a replacement or lock-in, and demonstrate how the same core methodology can be implemented with different front-ends.
Summary
Tom and Parker discuss the relationship between Obsidian, myPKA, and the ICO (Information and Capability Optimization) methodology. They clarify widespread confusion by explaining that myPKA is not a tool but a conceptual architecture launched in April that defines how AI can interact with a local folder to build persistent memory and knowledge systems.
They present the three-layer architecture: (1) The AI Layer consists of cloud-based LLM services like Claude, ChatGPT, or Gemini accessed via APIs; (2) The Local Folder Layer is a file system stored on the user's machine containing markdown files, agent instructions, work streams, and SOPs—this is tool-agnostic and can be accessed via any file explorer, terminal, or application; (3) The Front-End Layer is the interface used to navigate and interact with the folder, which can be Obsidian, a custom-coded UI, VS Code, terminal, or any markdown reader.
The speakers demonstrate their own implementations to show flexibility: Tom uses Obsidian as his front-end with the same local folder structure, while Parker has built a custom front-end using a database approach and HTML outputs, yet both follow identical ICO methodology principles. They emphasize that understanding these three layers as separate entities enables users to swap any layer without losing functionality or being locked into a specific tool.
They announce significant updates to their learning platform, including revamped ICO courses with 62 new animated video lessons, a completely new myPKA course explaining the folder structure in detail, and new community features including member video uploads enriched with AI. The AI Team folder contains pre-built agents (Larry the orchestrator, Pen for journal entries, and others) that help manage knowledge and action within the system.
The speakers also discuss ongoing optimization work with different LLMs, noting that Claude excels at proactive sub-agent orchestration while Codeex is better for certain creative tasks, and they're experimenting with forcing AI to use skills-based approaches for stricter rule adherence. They highlight the benefits of recommending Obsidian as the primary front-end: it removes the barrier to entry for non-technical professionals, provides a cohesive workspace, allows for easy plugin development to add features, and enables community members to create and share custom plugins.
Key Insights
- The confusion around myPKA stems from people misunderstanding that AI, local storage, and the front-end interface are three completely separate layers that can be mixed and matched independently rather than a single integrated tool
- Parker and Tom both follow identical ICO methodology principles but use completely different implementations—Tom uses Obsidian with local files while Parker uses a custom-coded front-end with database storage—proving the methodology is tool-agnostic
- Claude proactively uses sub-agents and can orchestrate them multiple levels deep, while Codeex requires explicit instruction to use sub-agents and only goes one level deep, making Claude more reliable for the orchestration model they've designed
- The decision to recommend Obsidian was motivated by removing the barrier that forced non-technical professionals from industries like marketing to learn software development just to use the productivity system
- Obsidian's plugin architecture allows the team to deliver isolated features without affecting other parts of the application, enabling rapid iteration and allowing community members to safely create and share custom plugins
Topics
Transcript
[0:02] Hello everyone. Welcome back to another podcast builder. What the heck is going on? See, I should rather generate myself using AI than talking. Might be more accurate in the meantime. Well, again, welcome back everybody to another session today talking about AI. No, not true. >> What a surprise. >> Sorry, I I was [laughter] [0:37] all right. Well, now everything is here. Sorry, guys. So, here. No focus. See, that's why multitasking is a lie. All right. So, let's do this properly everybody. Today we will talk about something that we realized in the past week and since we launched this new Obsidian approach and how to use AI in Obsidian and these things to clarify a…
Full transcript available for MurmurCast members
Sign Up to AccessMore from The AI Productivity Podcast
Obsidian is just a shell. Your knowledge folder is the real system.
Tom and Paco discuss how Obsidian is merely a shell for implementing the tool-agnostic iCore methodology, emphasizing that the underlying knowledge folder structure and methodology matter far more than any specific application. They introduce iCore for Life, a free Obsidian scaffold with custom plugins designed to help users understand and implement the productivity system's concepts and workflows.
Stop copying our AI setup. Copy how we think.
Tom and Paco discuss their different approaches to building AI-integrated productivity systems, emphasizing that the methodology (IICOR) should be implemented flexibly based on individual thinking styles and workflows rather than copied exactly. They showcase various visualization tools for managing deliverables, team knowledge, and work streams while stressing the importance of documentation and local folder structures over relying on external databases.