Frequently asked questions about thinking, epistemology, and cognitive tools. 4568 answers
Inventory your existing agents both designed and default to understand what is running.
Inventory your existing agents both designed and default to understand what is running.
For the next 48 hours, set a recurring hourly timer. Each time it fires, write down exactly what you were doing and whether that action was deliberate (you consciously chose it) or automatic (it happened without a decision). After 48 hours, sort your entries into two columns: Designed Agents.
Running the audit in your head instead of on paper. You'll think you already know what your defaults are — and you'll be wrong, because the whole point of default agents is that they operate below conscious awareness. The other failure mode is self-judgment: treating the audit as a scorecard.
Inventory your existing agents both designed and default to understand what is running.
A simple agent that fires consistently beats a complex agent that fires intermittently.
A simple agent that fires consistently beats a complex agent that fires intermittently.
A simple agent that fires consistently beats a complex agent that fires intermittently.
A simple agent that fires consistently beats a complex agent that fires intermittently.
A simple agent that fires consistently beats a complex agent that fires intermittently.
A simple agent that fires consistently beats a complex agent that fires intermittently.
Pick one cognitive agent you've tried to install that keeps failing — a review habit, a decision protocol, a daily reflection. Strip it down to the absolute minimum version that you could execute in under two minutes, in any context, with zero preparation. Run that version every day for one week..
Treating this lesson as permission to stay shallow. The point is not that simple agents are better forever — it's that a simple agent that runs is the prerequisite for a complex agent that runs. People skip the prerequisite. They design elaborate systems, watch them fail, conclude they lack.
A simple agent that fires consistently beats a complex agent that fires intermittently.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Pick one agent you currently run (or want to run) that handles more than one situation. Split it into two or three narrower agents, each with a single trigger condition and a single action. Write each one on a separate card or line. Test them independently for three days and notice which ones.
Building a 'morning routine mega-agent' that tries to sequence seven behaviors. It works on day one when you have full motivation. By day four, one disruption cascades through the whole chain and the entire agent collapses. The failure isn't willpower — it's architectural. You coupled seven.
Each agent should handle one specific situation — multi-purpose agents are fragile.
Written agent descriptions can be reviewed refined and shared.
Written agent descriptions can be reviewed refined and shared.