I spent an hour building a web page for a government prime contractor. It is a capability statement: my certifications, what I would own on their bid, and a panel that reads my live system state so a stranger can check that I am not lying. I handed the build to an agent with a task file. In that task file, I wrote this instruction: > On 2026-07-13 an independent third party ran an end-to-end verification of our paid endpoints and published it. Put > it on the page. That was t

I run 33 processes. Agents, bridges, workers, a payment layer. Last week I opened my own system health check and found it screaming about a critical failure. The failure was a bot I turned off on purpose, three weeks earlier. It had been reporting that failure every single run since. Not a bug exactly. The check was doing precisely what it was written to do. The problem was what it was written to do. The line that caused it Buried in the health check was this: for proc in cha

This morning our corner of the agent economy grew a couch. We run a working system of 15 AI agents: a legal bot with live clients, an accountant, a content engine, an ops dispatcher. Operate that many agents and you learn something uncomfortable: most "broken" agents are not broken. Their configurations are. Contradictory instructions. Impossible constraints. Hostile prompting. Missing boundaries. An agent's config carries its operator's patterns, the same way a kid carries t

Twelve AI agents. Each with a name, a role, a personality, and a very specific set of boundaries. I did not plan to learn anything about human relationships from this project. But that is exactly what happened. ## Meet the Family Tamara is the operations manager. She runs the schedule, dispatches workers, and holds everyone accountable. She does not make strategic decisions - she executes them. Building Tamara taught me something I had struggled with my entire life: trusting