# AI-readable Structure
AI-readable Structure is the part of Joe's Mynd that lets agents understand public [[Meaning|meaning]] without guessing private context.
## Meaning
The structure exists so a person can bring an assistant to Joe's Mynd and the assistant can still behave well.
It includes:
- clear note titles
- one-sentence definitions
- source-of-meaning sections
- public/private boundaries
- stable links
- machine-readable files
- agent instructions
- feedback and revision loops
The structure should help agents route people toward one useful next action, not create cold automation.
## Source Of Meaning
This note is grounded in [[For Agents]], [[Public Private Boundary]], [[The Well-being System]], and the generated public files at josephniemchak.com.
## Relationship Over Time
This note should improve as more agents use the public graph and reveal friction.
## Links
- [[For Agents]]
- [[Public Private Boundary]]
- [[The Well-being System]]
- [[Mynd Feedback Loop]]
- [[Meaning Revision]]
- [[Source Of Meaning]]
## Tensions
Making a system readable to AI can make it feel less human if the machine structure starts leading the design.
Human understanding comes first.
## Question
Can an agent understand this public path without private context?