PAI

Personal AI infrastructure provides the core architectural rules and governance for managing skills. It defines canonical skill structure, naming conventions, and global routing rules to ensure consistent loading and execution across system and user tiers.

You can customize AI skills without breaking the core architecture by using the two-tier extension pattern. This system versus user tier approach enables private customization and additions without altering foundational code or system operations.

To structure and canonicalize new skills, follow the foundational principles defined by the infrastructure. This involves adhering to canonical skill structure, naming conventions, frontmatter requirements, and global routing rules to ensure reliable loading and execution.

Yes, personal AI infrastructure supports private skill customization through its two-tier extension pattern. This pattern separates system and user tiers, enabling private extensions and component additions while maintaining overall architectural consistency and safety.

The two-tier skill architecture requires strict adherence to canonicalization rules and frontmatter definitions. While it enables safe extensibility, any new components must follow the established naming and routing rules to avoid breaking the overall infrastructure governance.

Your skill architecture needs canonicalization rules and frontmatter to ensure consistent skill loading and execution. These governance rules enforce structural consistency, safety, and proper context loading when skills are invoked across the infrastructure.