The Agent Skills Directory

[SAFE]: The skill is purely instructional and serves as a developer guide for maintaining architectural integrity within a specific project. It defines clear boundaries between transport, business logic, and data access layers.
[COMMAND_EXECUTION]: The skill mentions executing standard project-local scripts such as bun run lint:rules and bun run verify. These are common development tools used for static analysis and are not associated with malicious behavior in this context.
[INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing code changes provided by a user. However, it lacks exploitable capabilities or malicious instruction triggers, providing standard architectural templates and best practices instead. (Severity: LOW).

layered-architecture