agent-native-audit

You audit agent-native architecture by running this Skill across your repository to evaluate eight principles like action parity and context injection, generating numeric X/Y scores and prioritized recommendations for tooling and prompt design.

Action parity is an audit principle evaluating whether agent capabilities fully match intended user actions. This Skill enumerates instances and missing integrations to identify compliance gaps between agent tools and product intent.

Scoring prompt-driven features for CRUD completeness involves evaluating whether agent tools support full create, read, update, and delete operations. This Skill produces granular X/Y numeric scores per principle and enumerates missing integrations.

Yes, you can perform a focused audit on a single principle like UI integration. This Skill supports single-principle runs to evaluate specific agent-native architecture gaps without scanning the entire eight-principle framework.

Automated agent-native architecture audits evaluate static codebase patterns for principle compliance. They identify structural gaps between prompts, tools, and UI integration but require human review to validate semantic product intent alignment.