Open Second Brain Is Stable, and Dark Factory Got Its Next Layer
Open Second Brain has reached a stable version, Hermes can now use it as native memory, and the new Hermes Workflows plugin turns my manual playbooks into autonomous work graphs.
All posts
-
-
Dark Factory and Open Second Brain: autonomous development and the memory it runs on
A short talk about my two projects — Dark Factory and Open Second Brain. Before the presentation I asked the orchestrator to build a working chat app from scratch — and by the time I started speaking, the deployed site was already live at its URL. Inside: how both projects are wired together and why memory turned out to be a hard requirement, not a nice-to-have.
-
How I Built the First Dark Factory Workflow
Hermes on a VPS, a kanban graph of 13 tasks, reviews between stages on different profiles, a mini-brainstorm with a human over Telegram, and a bare-HTML deploy at the end. The first working Dark Factory stack — survived several debug runs and now ships through every stage without manual dispatching.
-
How I gave my AI agent a wallet - and why it immediately needed a memory
open-second-brain 0.8.0 and Pay Memory: how I let an agent pay for external APIs through pay.sh, and why the key part turned out to be not the payment itself but a clear record of every cent spent.
-
How I Built OpenSecondBrain
The story of open-second-brain: how Hermes on a VPS, Obsidian, MCP, CLI, and several agent runtimes came together into a small file-based memory for AI agents.
-
How I built this blog with coding agents
The story of launching techmeat.dev: from a spontaneous idea and a set of skills to Astro, Cloudflare Pages, brainstorming, and the first version of the blog.