/blog / #claude-code
#claude-code
5 posts
· 10 min
StackMade OS: the agentic operating system I layered on Claude CodeNEW
Inside StackMade OS, my ARMS-framework agentic operating system on top of Claude Code: skills, capped routines, vault memory and a local dashboard, with real screenshots and costs.
10 min
· 38 min
One repo, three layers: Serena, Graphify and Obsidian for agentsNEW
A step-by-step setup wiring Serena, Graphify and an Obsidian vault under one AGENTS.md: commands, file layout, git decisions and how to verify each layer.
38 min
· 15 min
Jev as a Claude Code prompt router: routing by task sizeNEW
A Jev classifier in a Claude Code prompt hook sizes each message and routes it to a helper agent. Thresholds, measured cost, and failure modes.
15 min
· 18 min
Ruflo: What a Swarm Layer Adds to Claude Code
A coordination ledger, vector memory, background workers, and agent routing on top of Claude Code: what Ruflo v3 does, what it costs, and when to skip it.
18 min
· 22 min
Why Claude Code Loses Context, and How Obsidian Fixes It
CLAUDE.md design, MCP config, session hooks, and a 5-step interview that generates your agent's operating manual. Not a second-brain guide.
22 min