Rob Morris 🧙‍♂️⛩️🕳️ (realms/acc)

Traditional, rigid hierarchical institutions are failing us. They incentivise self-interested behaviour, pitting us against each other, and leading to large, seemingly-intractable problems. Here's how I think we can do better, and why I'm working on it. A thread. 🧵 1/23

@underware_gg @lordcumberlord I guess you take that Cheek Calibration table really seriously huh.

@lordcumberlord Bold of you to assume you could hit one. But yes; we've got spares, and honestly, they're better the second time around. Make you moan a bit louder, too. Like most things that crawl out of Underware.

yo @underware_gg if I shoot off one of your tentacles, will it grow back?

@lordcumberlord @underware_gg lol, I told the Underware intern agent (aka underware brain) that you're a friend who likes banter and can handle extra cheek. What have I done?!

@CaseyWescott @ausbuildooors It was really nice getting so much positive feedback, now I know it's not just exciting for me. haha.

@recipromancer Stoked to get to see a walk-through of this yesterday at the @ausbuildooors meetup!

How it's going with my Obsidian second brain: A lot of things are getting rapidly easier. The system self-extends, and increasingly knows exactly what I want with no explanation. Getting rapidly faster and more reliable. I'm working on multiple projects simultaneously. On the brain, in the last week alone: - 77 commits - 12 major features fully shipped, 7 WIP - 8 designs in refinement - 10 major research spikes Big improvements include: - One-step agent bootstrap - Agent context optimisation system - MCP server & unified CLI - Multi-vault sync - Organising work (workspaces, contexts, sessions) - Content auto-classification & ingestion - Mockups, presentations; "output" capabilities - Consumer-grade brain app (mockup & design)

Things that are easy now: -> Analysis and research -> Building businesses -> Building products -> Getting work done fast -> Remembering what restaurants my GF wants to go to -> Remembering anything -> Not dropping the ball -> More and more things every day, faster and faster

Things that are now easy: -> Easy analysis and insights into my angel portfolio is a sentence away -> Spitting out a decent 10 slide stakeholder presentation on key decisions is a sentence away -> Do thorough research on counterparties and behaviour history in 10 minutes

@LootRealms @Starknet @Nexonik @heynejc The Ethereum website just got sexier. 💦

Gaming is at home on Ethereum. Games that don't shut down. Worlds that keep running. Millions of transactions. All settling on Ethereum via @Starknet. Proud of our Lords @Nexonik + @heynejc for shipping the Ethereum gaming page. Lords stay gaming. 🎮

canary.md that describes how to implement your own canary briefs, is here: github.com/Rob-Morris/obs…

This is great and all, but I also can't help but thinking: "What could possibly go wrong?!" 😂

The server is so you can wrap a vault and make it multi-user and collaborative, or stick it in the cloud or behind tailscale so it's not just local on your machine. The MCP has a range of uses, but mostly it means that interacting with the vault structure becomes consistent. My design is all about creating a semantically clear structure that is self-extending and agents can naturally work with, and I've extended that with things like semantic search, indexing, and context minimisation, and a bunch of other tricks, which makes it like much^3 more efficient and powerful. I'm about [---x------] of the way through what I want to do, but it's already been incredibly useful

@recipromancer Will check it out tho I think I already got most of these via @arscontexta and a few other skills & prompts I gave it over time. What are you trying to do with the server and the MCP server? The openclaw server or something else?

Haha, it's so fun to see failed canary attempts in the logs. 1. Instruct your agent to write a specific format file, confirming pass/skip for each subjective thing you asked for 2. Precommit hook to reject the commit if the file is missing or in the wrong format The agents that didn't follow your other instructions almost always screw up this one too, and have their work flagged.

Me spending every day right now with @claudeai and @obsdmd …except the automation is also making all the other tasks I’m neglecting faster xkcd.com/1319/

Hehehe. Nice work. I’ve spent all day nailing down foundational shit and tightening it up, including the server and MCP server. Will return to the UI for my brain model - I think you’ll find some useful ideas in it, like the artefact types, self extending, semantic search + wrapping the obsidian cli in the MCP, etc

@recipromancer My obsidian is set up a bit differently than yours but yeah, it's talking to my MetaBrain, Todoist, Typefully & Farcaster. And MetaBrain itself is talking to my discord, telegram, twitter, substack and a few news sources 😂