The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard Maven build commands (./mvnw compile, mvn compile, ./mvnw clean verify) to validate project state and verify changes. These are typical operations for a development-focused AI agent skill and are restricted to build and test activities.
[EXTERNAL_DOWNLOADS]: The skill references standard, well-known libraries from the Micronaut ecosystem and official database drivers (e.g., io.micronaut.flyway, io.micronaut.sql, org.postgresql:postgresql). These dependencies are sourced from established repositories like Maven Central and are consistent with the skill's stated purpose.
[DATA_EXFILTRATION]: While the skill interacts with database configuration files (application.yml) which may contain sensitive connection strings, its instructions are focused on configuration and review rather than external data transmission. No unauthorized network operations or exfiltration patterns were identified.
[PROMPT_INJECTION]: The skill uses instructional language to enforce safety boundaries (e.g., "MANDATORY", "CRITICAL SAFETY"). These are used legitimately to ensure the agent follows a safe development workflow (compiling before editing) and do not attempt to bypass the underlying LLM's safety filters or instructions.

513-frameworks-micronaut-db-migrations-flyway