---old---

True individual power in the age of AI. Essays, reviews, and conversations on creativity, empowerment, and design craft.

[---old--- @bitbrujo]

True individual power in the age of AI.

Essays, experiments, reviews, and guides on staying creative, capable, and in control — and the design craft that make it possible.

https://t.co/WpFAx22GCH

[---old--- @bitbrujo]

Full Guide -> https://t.co/Ycw1NQlrQE

#claudecode #mcp #secondbrain #ai #buildinpublic

[---old--- @bitbrujo]

Part 3 of my second brain series. This one's about the plumbing — how Claude actually does stuff beyond reading and writing files.

[---old--- @bitbrujo]

This system keeps expanding.

Built it for knowledge management. Now it handles reminders, event tracking, image generation, project incubation.

Start with the core. Add things when you discover a need. The system grows with you.

[---old--- @bitbrujo]

Setup takes 5 minutes. Add the server to your Claude config, restart, tell Claude when to use it.

Full guide with examples:
https://t.co/QVRlPxK4Pl

#claudecode #mcp #secondbrain #ai #agenticai

[---old--- @bitbrujo]

Memory, the Forgotten MCP -- Part 2: How to give your second brain's agentic system a relationship layer that persists across sessions

[---old--- @bitbrujo]

Reference servers aren't meant to impress you with features. They're meant to show you patterns.

The real value isn't the server itself — it's what becomes possible when you wire it into a larger system.

[---old--- @bitbrujo]

Briar — mesh messaging currently keeping Iranians connected, but is it safe?

I'm looking into Reticulum — resilient networking stack for LoRa, packet radio, serial lines. Trusted?

Decentralized comms isn't a niche. It's well designed insurance.

#briar #reticulum

[---old--- @bitbrujo]

I wrote up exactly how this works — the folder structure, the agentic loop, setup instructions.

Plus a starter repo you can clone and customize.

Full guide: https://t.co/RvetW3I72Y

#claudecode #secondbrain #ai #agenticai #buildinpublic

[---old--- @bitbrujo]

Commands are markdown files. Markdown files configure the agent.

You describe the behavior you want. The agent figures out implementation.

The barrier drops from "can you write code" to "can you communicate clearly."