Articles
I write about architecture decisions: what I built, what I discarded, and why. Building in public, not polished finished results.
-
Meaning Needs Rules: Why My AI Agents Aren’t Allowed to Search the Vector Index Directly
For weeks, my agents relied on an incomplete search index. After fixing it, they suddenly found notes they should never have been able to access without explicit permission.
-
One Truth, Many Agents: How an MCP Server Controls My Knowledge Vault
What happens when AI agents forget important rules in their prompts? I show how a custom MCP server controls access to my knowledge base, enforces writing rules, and ensures that multiple agents follow the same rules.
-
A Knowledge Base by Manifest: How an AI Agent Structures Your Company's Knowledge
An AI agent reads a Markdown manifest with eight questions and eleven rules and builds a complete enterprise knowledge store in a single session. Three layers, clear write permissions, and tested governance. No vector database as the foundation and no cloud dependency. This article explains the architecture and what it means for operating AI systems in Germany.