The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to run ./mvnw compile, mvn compile, and ./mvnw clean verify to ensure project stability before and after making changes. These are standard Maven operations for Java development used here as safety checks to prevent broken builds.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and refactor local project files, which constitutes a surface for processing untrusted data if the repository contains malicious instructions.
Ingestion points: Workflow steps 1 and 2 involve inspecting the current project setup and reading file content.
Boundary markers: None explicitly defined in the instructions for delimiting project data.
Capability inventory: The skill performs file system reads/writes and executes Maven build commands.
Sanitization: No specific sanitization methods for ingested code are described, though the skill provides explicit safety rules for the SQL it generates.
[EXTERNAL_DOWNLOADS]: The integration test examples reference the postgres:16-alpine Docker image for use with Testcontainers. This is a well-known service used for providing reproducible test environments.

512-frameworks-micronaut-data