The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 11 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Vin: Autonomous Agents Can Be Steered Entirely via Knowledge Vaults
“And now instead of like managing an agent or talking to another human about working on something, I just focus on managing this vault. This is like the new source. I just continually try and make it so that this vault has all of the information needed so that …”
Vin: Using LLMs as thinking partners beats purely building autonomous agents
“I really, really, really like working with LLMs as a thinking partner. It's my favorite way of using LLMs. I know people like to use agents and LLMs to build things, but I really like using them to think alongside me.”
Vin: Web AI memory is uncontrollable and opaque to users
“A lot of people are using like Claude or ChatGPT on the web and it has things like memory, but you can't like control, you don't know what's in that memory, right? You don't know what it knows and what it doesn't know.”
Vin: Obsidian's bidirectional linking mirrors human associative thinking
“The reason why people love obsidian is because of these interrelationships. The idea that you could open a file and then, you know, I just open this file and then I'm like, oh, interesting. I mentioned Greg Eisenberg. I can click that and it goes to that file.…”
Vin: Interlinked AI vaults can surface blind spots in thinking
“It can start to surface patterns about what you're thinking about that you are not seeing for yourself.”
Vin: AI calendar agents work better with daily note context
“If an agent has that context, you can, it can more effectively give you information about what you should do or not do, or, and it can more effectively make decisions on what should be in your calendar or not in your calendar.”
Vin: Backlinking to specific mental models beats tagging broad topic terms
“Originally I was backlinking to general terms, podcasts or physical fitness or filmmaking. I'm realizing it's not the most useful way to use obsidian. The most important thing to do is to create notes for each of my patterns, theories, projects, or perspective…”
Vin: Writing creates the context required to effectively delegate to AI agents
“Is that writing right now is a big way of how you delegate things to agents. That's like a whole new, that's a whole new aspect of it. So if you can develop a writing habit, You have a lot more context that you can pass over to an agent, which then dramaticall…”
Vin: Strict rule bars AI agents from writing in his Obsidian files
“I don't want an agent to write into the files. Like I could easily get it to do that. Like I could just say, like even here, I've asked it to write a description of some commands that I can talk about today, but I don't want it to make a file to do this becaus…”
Vin: Obsidian CLI gives Claude Code access to vault note relationships
“It can give Claude code, not only those files that it can read and access, but it can also give Claude code information about the interrelationships of those files.”
Vin shows '/connect' command synthesizing disparate domains via Obsidian's link graph
“So there's connect, which is allow, it allows me to take two domains and connect them using the vaults link graph. So I can just say I did one here. And I just asked it to connect filmmaking and world building. And so it goes through and it reads all of these …”