Learn
Field notes from the town library.
Short, practical guides to running more than one AI agent, and keeping them safe while you do.
What is AI agent orchestration?Coordinating several AI agents toward one goal, explained.Read moreMulti-agent coordinationThe five problems every multi-agent setup has to solve.Read morePrompt injection, for agentsHow injected text hijacks agents, and the defenses that work.Read moreThe dual-LLM patternA quarantined model reads untrusted text; the privileged one never does.Read moreTaint tracking for agentsMark runs that read untrusted data and limit what they can do.Read moreHuman in the loopWhere to put the human in an agent workflow, and how to keep it fast.Read moreBYOK, explainedWhat bring-your-own-key means for AI tools, and what to check.Read moreTask leasesThe simple trick that stops AI agents duplicating work.Read moreRun multiple Claude Code agentsA practical setup for several Claude Code sessions on one project.Read more
Move your agents in
Your first month of Pro is free.