A shared vocabulary for the agentic AI era.
Agentic AI introduced a wave of new problems — and new possibilities — that don't have names yet. This taxonomy gives them names. 103 terms across 10 categories, plus 23 principles that explain why the system behaves the way it does.
The Vocabulary
The pains, the gains, and the properties of working with agentic AI. Organized into 10 categories.
Context Quality
Problems and properties of the input/prompt.
Reasoning Cost
Extra model work caused by poor context.
Memory & State
What goes wrong when systems can't remember or trace — and what it looks like when they can.
Execution Quality
How well a system follows and completes its plan.
Architecture
Where planning, state, and control live.
System Transparency
How visible — or invisible — a system's reasoning, state, and actions are to the user.
Human-Agent Relationship
How the human-AI working relationship changes over time.
Evaluation
How AI work is judged.
Measurement
Metrics and scoring.
Human-AI Amplification
How AI extends human capability.
The Principles
Higher-order laws that explain why these patterns recur — and what to do about them.
The Context Compensation Principle
As context size and entropy increase, required reasoning grows nonlinearly.
The Bleed Principle
Unmanaged state costs compound. Every session without persistent memory costs more than the last.
The Divergence Principle
Agent autonomy and human understanding are inversely correlated without active inspection.
The Amplification Principle
AI's value scales with the quality of the system around it. Better context, memory, and structure unlock nonlinear returns.