Claude can now prompt its own SESSIONS. I built the boss.
Claude now has the ability to interact with multiple sessions, allowing a streamlined orchestration of tasks through an agent called Larry. This integration can simplify task management and enhance productivity by enabling a single main session to oversee various sub-sessions or tasks.
Summary
The transcript discusses the new capabilities of Claude, an AI system that can now communicate across multiple running sessions. The speaker introduces an orchestrator agent called Larry, designed to manage and delegate tasks efficiently. By linking specific tasks with Claude session IDs, the system aims to maintain an organized workflow within a folder structure, facilitating the tracking of completed, in-progress, and canceled tasks. The speaker emphasizes the need for clear organization to avoid task overload, and discusses the potential to create a ‘deliverables management system.’ As Claude interacts with different sessions, it demonstrates the capacity to autonomously handle background tasks and communicate updates back to the main session, thereby reducing the need for constant oversight from the user. The ultimate goal is to create a structured system that effectively integrates various projects and ensures that deliverables align with specific objectives, while also enabling improved visibility and efficiency in the workflow.
Key Insights
- Claude's capability to speak to different sessions allows it to not only look up content but also inject prompts into these running sessions.
- The key advantage of using the orchestrator agent Larry is to allow a single main session to manage multiple tasks without needing to reopen individual sessions.
- By attaching Claude session IDs to tasks, it's possible to maintain full context throughout the workflow, allowing the orchestrator to effectively delegate work.
- The framework features commands that Claude can execute, which are snippets that can be triggered to streamline task management within the folder system.
- The speaker argues for a structure that connects deliverables with corresponding projects to ensure that tasks are always aligned with broader goals, avoiding random work that does not contribute to objectives.
Topics
Transcript
[0:00] I don't know if you heard it, but Claude is now able to talk to different sessions. So, if you have several sessions open and you open a new one, it can actually talk to the other sessions. Not only look up the content, but actually inject prompts into these running sessions. And that's really powerful. And for those who don't know, if you use claude, all these sessions get stored locally on your device. they have all their own ID. That's when you [0:31] say resume session, you were able to find these different sessions based on the IDs. And I used this already in the past to let AI look up any context from previous sessions. Now,…
Full transcript available for MurmurCast members
Sign Up to AccessMore from ICOR with Tom | AI Productivity
I switched off Claude's memory. My folder does it better.
Tom explains why he disabled Claude's native memory feature and instead built a local folder-based system (myPKM) to maintain persistent context with AI agents. He shares an audit revealing systemic issues like contradictory documentation, forgotten rules, and parallel session conflicts, then outlines how to systematically fix these problems using the ICOR productivity methodology.
I do not download invoices anymore. My folder and Claude do it.
The speaker demonstrates how he uses AI (Claude) integrated with his local folder system to automate invoice collection for his accountant, downloading 36 invoices in 5 minutes with minimal manual intervention. He explains his AI team methodology, work streams, and how persistent local knowledge bases enable AI to handle recurring tasks independently without constant re-explanation.
Claude finished the work. Then I lost it in the chat.
A creator demonstrates a comprehensive local-folder-based system for managing AI-generated deliverables asynchronously, featuring an interactive browser interface that tracks decisions, changes, and conversations without relying on external databases or web applications. The system enables collaborative editing, visual annotations, and automatic context preservation across multiple parallel AI projects.
Claude writes better weekly reports than I did. Here is proof.
The speaker demonstrates how AI-powered weekly reports can automatically synthesize journal entries, notes, and captured content into personalized insights without manual review. Two different implementations are shown—one dashboard-style and one essay-style—proving the same underlying system can be customized to individual preferences and consumption habits.
I stopped doing my weekly review. The AI does it now.
The speaker has replaced traditional weekly reviews and manual note-linking with an AI-powered system built on a local folder structure called a 'scaffold.' This approach eliminates the time-consuming review ritual while enabling the AI to automatically connect notes and surface meaningful patterns that humans would miss.